Enhancing Indonesia S Power System

Enhancing Indonesia’s
Power System
Pathways to meet the renewables
targets in 2025 and beyond
INTERNATIONAL ENERGY
AGENCY
The IEA examines the IEA member IEA association
full spectrum countries: countries:
of energy issues
including oil, gas and Australia Argentina
coal supply and
Austria Brazil
demand, renewable
Belgium China
energy technologies,
electricity markets, Canada Egypt
energy efficiency, Czech Republic India
access to energy, Denmark Indonesia
demand side Estonia Morocco
management and Finland Singapore
much more. Through France South Africa
its work, the IEA Germany Thailand
advocates policies that Greece Ukraine
will enhance the Hungary
reliability, affordability Ireland
and sustainability of
Italy
energy in its
31 member countries, Japan
11 association countries Korea
and beyond. Lithuania
Luxembourg
Mexico
Netherlands
New Zealand
Norway
Poland
Please note that this Portugal
publication is subject to Slovak Republic
specific restrictions that limit
its use and distribution. The Spain
terms and conditions are Sweden
available online at Switzerland
www.iea.org/t&c/
Republic of Türkiye
United Kingdom
United States
This publication and any
map included herein are
without prejudice to the The European
status of or sovereignty over Commission also
any territory, to the
delimitation of international participates in the
frontiers and boundaries and work of the IEA.
to the name of any territory,
city or area.
Source: IEA. All rights

reserved.
International Energy Agency
Website: www.iea.org
Enhancing Indonesia’s Power System Abstract
Pathways to meet the renewables targets in 2025 and beyond
Abstract
This report was prepared on the basis of the framework for collaboration
established by the International Energy Agency (IEA) and the Ministry of Energy
and Mineral Resources (MEMR) of Indonesia on the topic of power system
enhancement and renewable energy integration, and in support of the
implementation of the upcoming Presidential Decree on renewable energy. It is
part of the assistance provided by the IEA towards Indonesia’s efforts to reform its
energy sector and is consistent with IEA’s forthcoming Energy Sector Roadmap
to Net Zero Emissions in Indonesia. The overarching objective of the assignment
was to assist Indonesia in tackling short-term power system challenges, by
achieving key targets such as reaching a 23% share of renewable energy in the
national electricity mix by 2025 in a secure and affordable fashion, and by making
grids progressively smarter. The assignment included the organisation of a
number of workshops for Indonesian stakeholders and a techno-economic study
performed by the IEA. It benefited from the support of the state-owned utility
Perusahaan Listrik Negara (PLN). This public report summarises the information
gathered from the workshops and presents the results of the study in a set of
recommendations for Indonesia.
IEA. All rights reserved.
PAGE | 3
Enhancing Indonesia’s Power System Acknowledgements, contributors and credits
Acknowledgements, contributors
and credits
This report was prepared by the Renewable Integration and Secure Electricity
(RISE) Unit of the International Energy Agency (IEA). It was co-ordinated by
former RISE analyst Peerapat Vithayasrichareon and Jacques Warichet, under
the guidance of César Alejandro Hernández Alva, Head of the RISE Unit. The
authors were Craig Hart, Luis Lopez and Jacques Warichet as well as former RISE
colleagues Randi Kristiansen and Peerapat Vithayasrichareon. The model was
developed by Craig Hart. Other IEA colleagues also provided valuable inputs and
feedback, including Praveen Bains, Piotr Bojek, Kieran Clarke, Elizabeth
Connelly, Timothy Goodson, Jeremy Moorhouse, Alison Pridmore, Vida Rozite,
Thomas Spencer and Gianluca Tonolo.
The IEA would like to thank PLN and the Ministry of Energy and Mineral
Resources for supporting the study and for providing data, information and
feedback during the project.
Elspeth Thomson edited this report. Colleagues in the Communications and

Digital Office (CDO), notably Astrid Dumond, Isabelle Nonain-Semelin and
Therese Walsh, helped produce the report and website materials.
The authors are grateful for the comments and feedback from Indonesian and
international experts, in alphabetical order: Meg Argyriou (Climate Works),
Hakimul Batih (OECD), Sean Collins (IRENA), Morten Egestrand (DEA), Rizky
Fauzianto (RMI), Philip Godron (Agora Energiewende), Shigeru Kimura (ERIA),
Florian Kitt (ADB), Randi Kristiansen (UN ESCAP), Raul Miranda (IRENA),
Barbara O’Neill (NREL), Peter-Philipp Schierhorn (Energynautics), Cecilia Tam
(OECD), Agus Tampulolon (IESR), Elizabeth Tinschert (GIZ), Fabby Tumiwa
(IESR), Peerapat Vithayasrichareon (DNV), Nicholas Wagner (IRENA) and
Matthew Wittenstein (UN ESCAP).
PAGE | 4
Enhancing Indonesia’s Power System Table of contents
Table of Contents
Executive summary .................................................................................................................. 6
Chapter 1. Introduction .......................................................................................................... 10
Meeting Indonesia’s target of a 23% share of renewables in the electricity mix by 2025 ....... 10
Power system flexibility to support clean energy transitions in Indonesia ............................... 11
Report scope and structure ...................................................................................................... 12
Chapter 2. Indonesia’s power sector and challenges ........................................................ 14
Indonesia’s power sector is a major source of emissions… .................................................... 14
… but Indonesia has ambitious targets for electricity access and decarbonisation ................ 15
The power sector is structured around state-owned utility PLN as the single buyer ............... 18
Overly optimistic demand forecasts and conservative reliability standards created generation
overcapacity ............................................................................................................................. 20
Indonesia has substantial energy resources but renewables are underutilised ...................... 23
PLN relies on biomass co-firing to achieve renewable target with little use of solar PV ......... 24
Power pricing regulation limits attractiveness of renewables investment ................................ 26
There is untapped flexibility potential in the Sumatra and Java-Bali systems ......................... 28
Contractual structures limit the flexibility of the young thermal fleet ........................................ 34
Summary of the flexibility challenge in Indonesia .................................................................... 36
Chapter 3. Energy transition pathways for Java-Bali and Sumatra in the short term .... 38
Modelling of the Java-Bali and Sumatra systems .................................................................... 39
Alternative pathways towards 2025 and their impacts on the sustainability targets ................ 43
The potential of solar PV to provide a low-risk pathway to achieve the 2025 renewable targets
................................................................................................................................................. 46
Flexibility requirements increase with higher shares of VRE ................................................... 50
Higher shares of solar PV lead to lower operating costs ......................................................... 56
Electrification of end-uses can help accommodate more solar PV.......................................... 61
Contractual flexibility of thermal plants enables a higher share of solar PV ............................ 65
Review and discussion of assumptions ................................................................................... 67
Chapter 4. Solutions for enhanced power systems ........................................................... 69
New contractual structures can harness the flexibility potential of Indonesia’s thermal fleet .. 70
In-depth assessment of needs can support Indonesia’s own smart grids strategy ................. 73
Enhanced operating practices can help realise the full benefits of technical and contractual
flexibility .................................................................................................................................... 78
Annex ....................................................................................................................................... 85
Abbreviations and acronyms .................................................................................................... 85
Units of measure ...................................................................................................................... 86
PAGE | 5
Enhancing Indonesia’s Power System Executive summary
Executive summary
Indonesia is a fast-growing economy, expected to become the 4th largest in the
world by 2050. To meet the growing energy demand, the government has set
ambitious sustainability targets and pledged to meet net zero emissions by 2060
or earlier. The power sector will play a major role in the energy transition, but is
today the country’s largest contributor to emissions from fossil fuel combustion.
Over 60% of the electricity is supplied by a young fleet of coal-fired power plants
whose installed capacity will meet a significant share of demand for years to come
unless steps are taken now to mitigate their emissions. Gas currently represents
almost 20% of electricity generation.
Indonesia has abundant natural resources and a huge potential for renewables,
especially hydro, geothermal and solar PV. The national electricity plan states a
target 23% share of renewables in the electricity mix by 2025 (up from 14% in
2021). To meet this target, the Electricity Supply Business Plan of the state-owned
utility PLN (RUPTL 2021-2030) states it will meet the target with new hydro,
geothermal and biofuel-firing capacities, and with biomass co-firing in coal plants.
Implementing the plan may be challenging due to delays in the construction of
these large generation projects. The plan forecasts relatively little use of solar PV
due to the currently higher cost of this technology in Indonesia. Globally, however,
solar PV has become increasingly competitive and its deployment can be quite
rapid thanks to short construction times. This recently prompted the government
to draft a new regulation promoting rooftop solar as a way to meet the 23% target.
This leads to the main research question in this study: could a higher share of
solar PV fill the gap and help meet the 2025 renewables target?
The study focuses on the two main systems of Java-Bali and Sumatra, where 80%
of the demand is located, and assesses their performance across a number of
scenarios in terms of system behaviour, costs and emissions.
The central scenario of the study assumes that all uncommitted and unallocated
capacity from the RUPTL, which includes 2.5 GW of new sources of renewable
(RE), and the bioenergy portion of designated co-firing capacity would be replaced
by utility-scale solar PV. This would require solar PV to reach a capacity of
17.7 GW (against 2.8 GW in the RUPTL) and an annual share in electricity of 10%
(against 2% in the RUPTL) in the combined systems of Java-Bali and Sumatra in
2025. Even though such a capacity deployment looks ambitious, the scenario
serves as an illustration of the potential role that increasing shares of variable
renewables could play to help Indonesia reach its objective to attain a
decarbonised and diversified electricity mix.
The main finding is that the existing assets in Java-Bali and Sumatra are capable
of accommodating a 10% share of solar electricity by 2025 using flexibility means
PAGE | 6
existing in the system. The coal and hydro plants are completely capable of
delivering the needed flexibility and continue to ensure system stability and
adequacy. A relatively high level of PV curtailment in Sumatra (14% yearly) is
however observed, at periods of low demand and high solar infeed due to the
contractual inflexibility of power purchase agreements (PPAs), as explained
further. No investment in additional grids or storage capacity is required. However,
this amount of variable generation requires updates to operating practices such
as the appropriate forecasting and representation of these forecasts in system
operation decisions, and the ability to monitor and control the operation of solar
PV plants, including the ability to curtail where necessary.
On the cost side, the picture is more nuanced. Solar PV brings fuel savings from
both fossil fuels (5.5-7%) and biomass (which comes at a premium) but the current
regulations in Indonesia do not allow solar PV to compete in the short term when
considering the total system cost. However, the authorities have options. Long-
term plans for PV deployment would allow a local industry to develop and offer
cheaper rates. Removal of subsidies to coal generation and the introduction of
carbon pricing would further improve the business case of all renewable sources.
Given the focus of the study on 2025, these longer-term initiatives are not studied
in detail but could be the subject of a further study. Another aspect not included in
this study is the benefits of enhanced grids and interconnections, such as the
expected interconnection between Java-Bali and Sumatra in 2028. The role that
increased interconnection among Indonesia’s main islands could play in the long
term is addressed in IEA’s upcoming Energy Sector Roadmap to Net Zero
Emissions in Indonesia.
A key barrier to accommodating variable renewables in the Indonesian power

system is contractual inflexibility. Take-or-pay (ToP) obligations in PPAs between
PLN and independent power producers (with guaranteed offtake obligations) and
in fuel supply contracts for gas generators reduce incentives for thermal units to
be flexible and affect the overall efficiency of the system. The PPA constraint is
significant since the capacity of coal IPPs in Java-Bali is equal to two-thirds of the
peak demand in 2025. With the assumption of a 60% guaranteed offtake each
year, this significantly reduces the room in the generation mix for renewables.
These contractual constraints are a barrier not only to variable renewables but
also to any new renewable capacity, even dispatchable (hydro, geothermal), and
lead to higher system costs. Removing these constraints, at least partially, would
therefore provide room for renewables, reduce costs and help reduce emissions.
To provide concrete recommendations on the contractual structures and amount
of contracts to revise, more contractual data would be required.
The study does not look in detail at the role of biomass co-firing, a key contributor
in PLN’s plans to meet the 23% target in 2025. It notes, however, that biomass
co-firing in PLN’s existing coal plants at low blending rates (10-20% as currently
PAGE | 7
considered, as these require no retrofit of the assets) may exacerbate the coal
dominance in the electricity mix. Reliance on a significant share of biomass
generation through co-firing in selected plants as specified in the RUPTL would
require these plants to run almost flat out, while for every unit of energy from
biomass, nine units from coal are forced into the system. This reduces the system
flexibility and increases operating costs through interactions with other contractual
inflexibilities such as gas ToP contracts. A more thorough study would be required
to assess the role of biomass in Indonesia’s electricity mix.
The study also looks at the role of electrification of end-uses, like cooking and road
transport, which are part of the government strategy to decarbonise the economy,
reduce oil imports and improve air quality and emissions. Given the over-sized
thermal capacity, increased electricity demand reduces the curtailment of solar PV
(mainly in Sumatra) but also leads, in Java-Bali, to an increase in power sector
emissions, supporting the need to decarbonise electricity as end-uses are
electrified. Despite this, overall emissions are still reduced with the electrification
of road transport, driven predominantly by efficiency gains in the move from diesel
internal combustion engines to electricity for two- and three-wheelers.
The overall conclusion is that, from a system integration perspective, Indonesia

can aim for higher shares of renewables than those listed in the current plans for
2025 and beyond, especially when considering a mix of variable renewables and
other dispatchable technologies. However, investment in these renewable
capacities faces the risk of low-capacity factors due to the very high amount of
thermal capacity in the system with inflexible contractual structures. A priority for
the Indonesian power sector is to review the contractual arrangements, while
respecting investors rights, and ensure that the thermal fleet is used as closely to
actual technical capabilities as possible.
Key recommendations
• The power sector should put renewables, in particular solar PV, at the centre of
planning and start adapting operating practices to enable more generation from
variable renewables.
• The authorities should take actions to improve the financial competitiveness of
solar PV with respect to other technologies in Indonesia. In the short term,
constraints such as the local content might be reconsidered.
• The authorities could also take wider action to level the playing field between
coal and other technologies by removing the implicit and explicit subsidies to coal,
for example the price cap on coal supply, and supporting a shift towards low-
carbon sources with some form of carbon pricing.
• The power sector should make best use of the current assets in the system and
allow or incentivise the generation fleet to operate according to its technical
capabilities.
PAGE | 8
• The authorities should prepare new contract structures with embedded flexibility
for all new PPAs and fuel supply contracts. In addition, to complement the
upcoming coal phase-out programmes, they should consult with stakeholders in
order to design an approach that brings additional flexibility from existing
generation assets.
PAGE | 9
Enhancing Indonesia’s Power System Chapter 1 - Introduction
Chapter 1. Introduction
Meeting Indonesia’s target of a 23% share of
renewables in the electricity mix by 2025
Indonesia is today the seventh largest economy in the world, the twelfth largest
energy consumer and the ninth largest emitter of CO2 from fuel combustion. Its
economic growth over the next decades will be spectacular, going from a GDP per
capita of around USD 13 000 in 2021 (which stands at 70% of the global average),
to USD 40 000 by 2060, a level equivalent to today’s Japan, taking it to the range
of advanced economies, and within the top five global economies when total GDP
is considered.
In 2021, Indonesia announced a commitment to achieve net zero emissions by

2060, or earlier with support from advanced economies. The measures that could
be taken to achieve this target include introducing energy efficiency measures,
increasing the share of renewables, reducing the use of fossil fuels, increasing the
adoption of electric vehicles and electrification of residential end-uses. The
transformation of the power sector would therefore be critical in achieving this
goal.
Emitting 224 million tonnes of CO2 in 2019, the power sector in Indonesia is the
country’s largest emitter, accounting for 38% of emissions from fuel combustion.
Coal is the largest means of power generation, accounting for around 60% of the
country’s total electricity output. The emissions intensity of Indonesia’s power
sector is highest among the Southeast Asian countries, at 760 tonnes of CO2 per
kWh in 2019. The policies and regulatory frameworks of the power sector therefore
need to be realigned towards the net zero goals. In the short term, in its National
Electricity General Plan (RUKN), the Directorate General of Electricity (DGE)
under the MEMR set a target to achieve at least a 23% share of renewables in the
electricity mix by 2025 (up from around 14% in 2021). While Indonesia has
substantial renewable energy potential, particularly geothermal, hydro and solar
generation, only a small proportion has been utilised or planned, offering a range
of options to decarbonise the power sector.
Indonesia is the largest archipelago in the world and there are significant
differences among the islands. Java is the most populated island and its system,
interconnecting the three islands Java-Madura-Bali (hereafter Java-Bali or JVB) is
by far the largest power system. Together with neighbouring island Sumatra
(SUM), these two systems represent 80% of Indonesia’s electricity demand. Until
these two systems are interconnected (from 2028 at the earliest), Java-Bali and
Sumatra are distinct systems with different features. Java-Bali is a densely-
PAGE | 10
populated island with a coal-dominated electricity system while Sumatra is a more

rural island with a high potential for solar and hydro power. As interconnections
between the islands grow in the future, Sumatra has the potential to play a role as
a renewables hub. Therefore, an examination of Java-Bali and Sumatra brings to
light the many and diverse potentials and challenges relating to renewables growth
in the Indonesian power sector, while already representing four-fifths of the
country’s electricity demand.
Power system flexibility to support clean

energy transitions in Indonesia
Although Indonesia has high potential for hydro and geothermal generation, solar
PV is expected to play a growing role. Currently, solar PV comes at higher costs
than other renewable technologies. However, wind and solar PV are expected to
account for a growing share of the new additions globally as their cost
competitiveness continues to increase. . Integrating variable renewables (VRE),
like wind and solar PV, into the power system can be challenging due to their
unique technical properties: variability because the output varies over time
depending on the availability of primary resources (wind or sun); and uncertainty
as the output cannot be perfectly forecasted, especially at longer lead times.
Systems such as in South Australia, Denmark or Ireland, have demonstrated that
this challenge can be addressed in cost-effective and reliable ways. To help
governments deploy efficient measures in the right order of priorities, the IEA has
developed a framework to capture the evolving impacts and understand the
challenges of integrating VRE into the power system.
In particular, solar PV produces electricity according to diffuse solar radiation,

which is affected by the seasons and of course daylight, requiring other resources
to take over in the evening. Higher reliance on solar PV therefore requires
sufficient flexibility in the electricity system to keep supply and demand in balance
across all relevant timescales, ranging from seasons to hours and minutes. Power
system flexibility refers broadly to all the attributes of a power system that allow
the system operator to reliably and cost-effectively balance demand and
generation in response to variability and uncertainty. Key components of power
system flexibility include technical flexibility, contractual/institutional flexibility and
operational practices. These components must be considered simultaneously in
the effort to enhance system flexibility.
PAGE | 11
The building blocks of power system flexibility
Report scope and structure

This report focuses on the two largest systems of Java-Bali and Sumatra and how
they may evolve by 2025 to meet the target of a 23% share of renewables in the
electricity mix. In particular, the ability of these two systems to accommodate more
variable renewable energy (VRE) is analysed. Given the significance of these two
systems, the differences between them and the nature of the challenges,
recommendations can be drawn for Indonesia’s power sector over the next
decade as it will play a key role in the country’s pathway towards carbon neutrality
by 2060.
This study has three main parts. Chapter 2 provides the context of the existing
Indonesian power sector. It gives a general description of today’s power system
before diving into the challenges relating to system flexibility that must be
addressed for the integration of higher shares of VRE such as solar PV. Chapter
3 focuses on the two largest systems of Java-Bali and Sumatra, which together
account for 80% of the country’s demand for electricity. It presents a techno-
economic analysis to assess the capability and challenges that these two systems
face in integrating VRE, to meet the renewable targets of 2025. A central scenario
looks at the potential of solar PV to fill the gap between the current trajectory and
the target. Alternative scenarios address the electrification of transport and
PAGE | 12
residential cooking – separate but parallel goals of the Indonesian government –

and how they can contribute to VRE integration. Another key aspect that is studied
is the inflexibility of the current contractual structures for power generation.
Chapter 4 builds on the key outcomes from the previous two chapters to identify
some measures that would support reaching the short-term sustainability
objectives and prepare for more ambitious targets. These measures include
enhancements to contractual structures, power system planning and operation
practices, and deployment of smart grids.
PAGE | 13
Enhancing Indonesia’s Power System Chapter 2 - Indonesia’s power sector and challenges
Chapter 2. Indonesia’s power

sector and challenges
Key findings – Power sector in Indonesia

• Over the past few decades, Indonesia’s economic growth and electrification
progress have been remarkable, enabling near-universal access to electricity.
However, overly optimistic demand growth forecasts and conservative
planning practices relying on high generation capacity margins, as well as a
lack of interconnections among islands has led to overcapacity of thermal
plants, far in excess of peak demand for years to come on the main island of
Java.
• Driven by the high share (60%) of coal-fired generation, the power sector
represents 38% of Indonesia’s emissions from fossil fuels combustion today,
but the MEMR has set an ambitious target for a 23% share of renewables in
the electricity mix by 2025 (up from 14% in 2021).
• The RUPTL (Electricity Supply Business Plan) 2021-2030 relies on new hydro
and geothermal capacities, and on biomass co-firing in about a third of the
coal-fired capacity (at the co-firing rate of up to 10% biomass) to reach the
target of a 23% share of renewables in the electricity mix by 2025, but relatively
little on solar PV.
• Despite the increased competitiveness of this technology globally, solar PV in
Indonesia remains unattractive compared to other technologies due to low
deployment levels and local regulations, such as the local content obligation
which keeps PV prices high. Regulated fuel and electricity prices result in the
need for state subsidies to the state-owned utility PLN and further limit the
attractiveness of private investment in new renewable capacities.
• A key barrier to accommodating VRE in the Indonesian power system is
contractual inflexibility. PLN has to dispatch the young thermal fleet of coal-
fired IPPs according to minimum offtake obligations which cover a significant
share of the demand.
Indonesia’s power sector is a major source of

emissions…
Indonesia is an archipelago of more than 17 000 islands and with a population of
270 million. The distribution of population and resources among the different island
PAGE | 14
groups is uneven. Likewise, the levels of economic development vary across the
different island groups leading to different energy system features. Indonesia is a
fast-growing economy with growing energy use. Between 2000 and 2019, GDP
grew from USD 395 to 1 049 billion (2015 market exchange prices), while its total
primary energy supply grew from 6.53 to 10.09 exajoules (EJ).
Due to the large share of fossil fuel generation, the power sector’s CO₂ emissions
(224 MtCO₂) represent the largest portion (38%) of the country’s total emissions
from fuel combustion. The emissions intensity of the power sector has increased
over the past decades, contributing over 220 million tonnes of CO2 in 2019.
Power sector emissions and CO2 intensity in Indonesia and in selected Southeast
Asian countries, 2000 and 2019
Source: IEA (2021), World Energy Statistics and Balances.

… but Indonesia has ambitious targets for

electricity access and decarbonisation
Indonesia’s national energy policy, the Kebijakan Energi Nasional or KEN, is a
high-level strategy adopted in 2014, which lists targets for 2050. The main drivers
for this energy policy are energy security and energy independence, to be
achieved through energy conservation and supply diversification. It is important to
note that the current energy policy, which pre-exists the net zero pledge, aims to
reduce the role of oil in the primary energy supply but is open to expanding
coal and gas due to their abundance in the country. The targets for renewables in
primary energy are aligned with the ASEAN targets.
PAGE | 15
Selected targets under the 2014 National Energy Policy
Unit 2020 2025 2050

Primary Energy Supply EJ >16 >42
Share of Primary Energy
New and renewable energy % >23 >31
Oil % <25 <20
Coal % >30 >25
Gas % >22 >24
Installed power capacity GW >115 >430
Electricity access rate % 100
Electricity consumption per capita kWh/capita 2 500 7 000
Note: New and renewable technologies include non-renewable sources such as nuclear, hydrogen, coal bed methane,
liquefied coal and gasified coal.
Source: Government of Indonesia (2014), National Energy Strategy (KEN).
To put these targets into the perspective of the most recent data relative to year
2020, almost full electricity access was achieved and the installed power capacity
amounted to around 72 GW. Electricity consumption per capita was slightly less
than 1 000 kWh and total energy supply was around 10 EJ. The discrepancy
between these last values and the targets illustrates the excessively optimistic
demand forecasts, an issue that is further explored below.
In its updated Nationally Determined Contribution (NDC), the country aims to

reduce total economy-wide emissions to 2 034 MtCO₂-eq (unconditional) or
1 683 MtCO₂-eq (conditional on international support) by 2030. This entails
achieving emissions for coal, oil, gas and power of 1 355 MtCO₂-eq
(unconditional) or 1 223 MtCO₂-eq (conditional) from a business-as-usual
assumption of 1 669 MtCO₂-eq by 2030.
For the power sector, a peak of 349 MtCO₂-eq is planned by 2030 in order to
reach net zero emissions by 2060, up from 224 MtCO₂-eq in 2019. In 2021, the
government announced in its NDC the goal to meet net zero emissions by 2060,
or sooner with support from developed countries. Five strategies were identified
to achieve this target: 1) increasing the share of renewables, 2) reducing the use
of fossil fuels, 3) promoting electric vehicles, 4) electrification in the residential and
industrial sectors and 5) utilising carbon capture, utilisation and storage (CCUS).
The power systems across the country are very diverse. Highly developed
provinces such as Jakarta, Central Java and Bali have electricity access rates of
100%, while the less developed eastern provinces (in Nusa Tenggara) have rates
just below 90%. Diesel-based generators are common in the isolated small-island
grids, while coal, hydro and geothermal resources are prevalent in the larger
systems of Java-Bali and Sumatra.
PAGE | 16
Population, electricity consumption and electricity access rates in the major island
groups, 2021
Electricity
Population Consumption
access rate
(million) (GWh) (%)
Java-Madura-Bali 156 175 035 99.60
Sumatra 58 38 111 99.69
Kalimantan 16 11 352 98.86
Sulawesi 20 11 423 99.23
Maluku, Papua, Nusa Tenggara (MPN) 18 6 192 94.13
Total 268 242 113 96.7
Note: Electricity access rates in the different provinces of the eastern island group MPN vary greatly.
Source: PLN (2021), RUPTL 2021.
Indonesia’s abundant resources for power generation enabled it to increase

electricity access rates rapidly despite the higher costs associated with
archipelagic layouts. Most notably, the role of coal expanded from a 36% share in
2000 to almost 60% by 2019, when total generation grew from 93 TWh to 294
TWh.
Generation in Indonesia and in selected Southeast Asian countries by fuel, 2000 and
2019
350 Biofuels
TWh
Solar and wind

300
Geothermal
250 Hydro
Gas
200
Oil
150 Coal
100
50
0
2000 2019 2000 2019 2000 2019 2000 2019 2000 2019
Indonesia Malaysia Philippines Thailand Viet Nam
Source: IEA (2021), World Energy Statistics and Balances.

The transmission system is mostly made up of 150kV transmission lines, with a

limited number of high voltage 500kV lines found only in the Java-Madura-Bali
system, but which form the strong backbone of its transmission network. As the
transmission system is long and thin, voltage stability is a concern, especially
within the 150kV network. In 2005 and 2019, transmission outages led to
widespread blackouts in Java.
PAGE | 17
Apart from Java-Madura-Bali, the islands currently have separate systems.

However, there are plans to build more interconnections between islands and with
the neighbouring countries in the future under the ASEAN Power Grid (APG). The
first planned interconnection is between Java and Sumatra, from 2028. These
interconnections and further reinforcements to the transmission systems will play
a key role in enabling renewables-generated power to be shared across larger
balancing areas.
Transmission lines in the Java-Bali and Sumatra system
This map included is without prejudice to the status of or sovereignty over any territory, to the delimitation of international frontiers and
boundaries and to the name of any territory, city or area.
Source: MEMR geoportal.
The power sector is structured around state-

owned utility PLN as the single buyer
Indonesia’s power sector is currently regulated by Law 30/2009, or the 2009
Electricity Law. It allocates roles and responsibilities to the institutions and market
players, and defines rules for permitting, tariff setting and planning of the power
sector.
The MEMR or Kementerian Energid dan Sumber Daya Mineral (ESDM) is the
main policymaker and regulator for energy affairs, including electricity.
PAGE | 18
The DGE under the MEMR specifically focuses on the development of the
electricity sector and coordinates with other directorates such as oil and gas, coal
and minerals, and new and renewable energy. The national electricity policy is set
by the DGE through the National Electricity General Plan (Rencana Umum
Ketenagalistrikan Nasional or RUKN) and takes input from KEN and from the
National Energy Master Plan (Rencana Umum Energi Nasional or RUEN). It is an
indicative 20-year demand outlook with government policy targets such as
electricity access and interconnectivity rates, share of renewables and uptake of
electric vehicles. The DGE set the goal for at least a 23% share of new and
renewable energies (NRE) in the electricity mix by 2025. The electricity system
transformation also needs to fulfil the three key sustainability pillars, which are
Availability (security of supply), Affordability (least cost) and Acceptability
(environmental sustainability).
The 2009 Electricity Law designates PLN, a state-owned vertically-integrated

utility, as responsible for generation, transmission, distribution and retail. Several
attempts have been made to restructure and unbundle the electricity sector but
these have been unsuccessful. The Constitutional Court stated in 2015 that the
unbundling of electricity services would not be permitted if it meant that the state
would have less control over the sector as a result.
Several independent power producers (IPPs) are allowed to participate by selling

energy directly to PLN, which acts as a single buyer. Under the 2009 Electricity
Law, private businesses may be given the right to provide electricity for public use
through the electricity business licences or Izin Usaha Penyediaan Tenaga Listrik
(IUPTL), especially for renewable energy projects or those with construction or
fuel supply risks. In the Java-Bali system, the generation capacity from IPPs in
2025 will be about 19 GW, accounting for a 54% share. However, several IPPs
reported difficulties in terms of inconsistent application of regulations as well as
lack of transparency and clarity in PLN’s rules for procuring electricity from power
plants, 1 resulting in limited private participation in the power sector, and
subsequently the limited development of renewable energy.
In contrast, transmission and distribution infrastructure remains within the

ownership of PLN. Private operators are allowed to undertake build-operate-
transfer (BOT) or build-lease-transfer (BLT) schemes in order to alleviate the
investment burden of PLN in grids while retaining the latter’s control.
1
OECD (2021), Clean Energy Finance and Investment Policy Review of Indonesia, Green Finance and Investment, OECD
Publishing, Paris, https://doi.org/10.1787/0007dd9d-en
PAGE | 19
Overly optimistic demand forecasts and

conservative reliability standards created
generation overcapacity
Under the 2009 Electricity Law, electricity supply business planning (Rencana
Usaha Penyediaan Tenaga Listrik or RUPTL) is drafted by PLN every year in
coordination with other stakeholders and has to be approved by the MEMR. The
latest RUPTL was issued in October 2021 for the period 2021-2030 (hereafter,
RUPTL 2021).
It is important to note that demand projections, both the peak demand and annual
consumption, of Indonesia have been consistently overestimated. This is a
common problem among utilities under regulatory frameworks allowing risks
(costs) to be passed on to consumers or tax payers. The demand forecast has
been a top-down approach based on historical elasticities and GDP growth for
each sector. In addition to the observed over-estimation of economic growth, this
approach is not able to reflect energy efficiency improvements.
Historical data and projections of demand (left) and peak load (right) in the RUKN and
RUPTL, 2011-2030
600 160
Demand Peak Load Historical
GW
TWh
550 140 RUKN 2008

500 RUKN 2019
120
450 RUPTL 2015
100 RUPTL 2019
400
RUPTL 2021
350 80
300
60
250
40
200
150 20
100 0
2012
2014
2016
2018
2020
2022
2024
2026
2028
2030
2012
2014
2016
2018
2020
2022
2024
2026
2028
2030
Source: IEA analysis of MEMR (2008), RUKN 2008; MEMR (2019), RUKN 2019; PLN (2014), RUPTL 2014; PLN (2019),
RUPTL 2019 and PLN (2021), RUPTL 2021.
High demand forecasts contributed to the overbuilding of power plants, which has
led to high reserve margins, particularly in central and eastern Java. Excessive
capacity unnecessarily increases the system costs of a power system and can
delay the uptake of cheap renewable capacity as revenues are prioritised towards
payments for existing assets, mainly excess fossil fuel power plants.
PAGE | 20
Historical reserve margins in the Java-Bali power system

70% Java-Bali
Reserve margin
60% Western Java
50%
Central Java
40%
Eastern Java and
Bali
30%
20%
10%
0%
2017
2018
2019
2020
Source: IEA analysis of the RUPTL (2017; 2018; 2019; 2021).
There are several reliability criteria that Indonesia has set out in its electricity
policy. The main reliability criterion is to maintain the reserve margin at a
minimum of 35%, which is based on the loss of load probability (LOLP) of less
than 0.274%, or one day per year in which the peak demand cannot be met 2 within
the area of a control centre unit. The reserve margin criterion is higher than
international standards for power systems based on fossil fuels. In Java-Bali, the
reserve margin is forecast to be in the range of 40-60% for the next decade. In
particular, generation is being built in the western part of the system, where the
capital Jakarta is located. Given the high margins in the eastern part of the island,
one wonders if grids would be needed instead of generation. Also, moving towards
a fully probabilistic method, such as those used in Australia and Texas, would
result in lower system costs while maintaining adequate system reliability.
On the distribution side, the System Average Interruption Duration Index (SAIDI)
and System Average Interruption Frequency Index (SAIFI) are monitored to
assess the reliability of service. Since 2016, the interruptions have been more
frequent and their resolution has taken longer, resulting in worse indices.
Compared to its ASEAN neighbours, the Indonesian power sector in 2019 had
higher values of SAIDI (18.95 hours per year, compared to the ASEAN average
of 10.27 hours per year) and SAIFI (11.51 times per year, compared to the ASEAN
average of 8.71 times per year). A smart grid programme is now aiming to improve
service reliability and reduce the instances and duration of interruptions.
2
The relationship between the LOLP and the reserve margin is the result of an assessment which depends on a number of
parameters such as the outage rate of power plants and the size of the plants relative to the system size. The same LOLP
may lead to different reserve margins in different systems. For comparison, a LOLP of one day per year corresponds to
margins below around 15% in some US systems.
PAGE | 21
SAIDI and SAIFI indicators in Indonesia, 2011-2020

30 16
SAIDI SAIFI
instances / customer
hours / customer
14
25
12
20
10
15 8
6
10
4
5
2
0 0
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
The main operation principles of PLN are designed to ensure the reliability
(credible contingency), cost-effectiveness (economic dispatch and losses) and
quality of the system (frequency and voltage excursions). PLN entities act as
system operators in their respective territories with the authority to call upon
generators to ramp up or ramp down production, or start-up and shut down, to
shed load, and to curtail intermittent generation.
The operating practices in Indonesia are deterministic, in line with traditional

practices as applied to systems dominated by baseload fossil generation. The
power system operates at 50.00 Hz with a normal frequency range of ± 0.20 Hz,
with allowable operation during disturbances ranging between 47.50 Hz and 52.00
Hz. The spinning reserve requirement is based on N-1 criterion to cover the loss
of the largest unit in the system. Protection schemes including load shedding,
under frequency relay and islanding modes, are employed to ensure system
stability. The voltage stability standard is maintained at +/- 5% for 150kV and
500kV systems through reactive power management. One of the key challenges
in Java-Bali is the voltage stability due to the network topology, being long and
thin. For transmission, the country implements static and dynamic N-1 security
criteria, ensuring the ability to deliver the same amount of energy based on load if
a transmission circuit goes out. As some islands will experience growing
penetration of VRE, their system’s stability (the ability of the power system to
recover from disturbances on very short time scales and maintain the state of
operational equilibrium) and reliability criteria, operational standards and
contribution of technical assets to provide system services may need to be revised
towards more probabilistic approaches.
PAGE | 22
Indonesia has substantial energy resources

but renewables are underutilised
Coal is a strategic resource for Indonesia. It is a major exporter of coal to India,
the People’s Republic of China (hereafter, “China”), Japan, Malaysia and the
Philippines. In 2020, it produced 563.7 million tonnes of coal, exported 405 million
tonnes, and consumed 131 million tonnes domestically – 79% of which went to
the power sector.
Coal supplies 60% of the electricity demand. Domestic coal supply is secured
through a Domestic Market Obligation (DMO) established by Regulation 25/2018
(which supersedes Regulation 34/2009) requiring coal producers to reserve a
minimum percentage – which may vary each year – for domestic sales aimed at
keeping the cost of electricity low. For 2021, 25% of domestic production was set
as the obligation, with USD 70/tonne as a price cap. 3 This local obligation at a set
price is in fact a fuel subsidy that affects the overall efficiency of the
Indonesian energy system. Given the currently high prices on international coal
markets, this local obligation also represents an opportunity cost for Indonesia.
However, this consideration did not prevent Indonesia from banning coal exports
in January 2022 as PLN's reserves were approaching historic lows.
Gas is also a significant resource in Indonesia, and the country exports both
pipeline gas and LNG. In 2020, the power sector consumed 12% of total net gas
production, while gas itself was responsible for 19% of total power generation.
Indonesia also has significant renewable energy potential, but only a small
percentage has been realised. Geothermal and hydropower have been the main
sources of renewable power generation to date. According to the latest Electricity
Supply Business Plan (RUPTL 2021), these two sources combined are expected
to contribute to 4.6 GW of the 10.6 GW additional NRE capacity needed to achieve
the country’s target of a 23% share of NRE in power generation by 2025. IEA’s
estimates, as well as those of several other organisations (for example, the
Renewable Energy Pipeline by the Danish Energy Agency, the Review of
Renewable Energy Potentials in Indonesia and Their Contribution to a 100%
Renewable Electricity System by TU Delft, Indonesia’s Vast Solar Energy
Potential by the Australian National University) support much higher potential for
wind and solar compared to the RUPTL, but this potential is unevenly distributed
across the islands, especially wind power. As for utility-scale PV, the Java-Bali
system has a potential in the range of 59 GW, compared to more rural Sumatra
which has a potential exceeding 600 GW.
3
Decree of the Minister of Energy and Mineral Resources Number 139.K/HK.02/MEM.B/2021 on Meeting Coal Needs.
PAGE | 23
Renewable energy potential according to the RUPTL and IEA, and current installed
capacity in Indonesia
Realisation of
Potential Potential
Installed capacity potential
according to according to
in 2020 according to
RUPTL IEA
RUPTL
GW GW % GW
Geothermal 29.5 2.13 7.2
Hydro 75.1 5.64 7.5
Mini- / Micro hydro 19.4 0.50 2.6
Bioenergy 32 1.89 5.9
Utility-scale solar (4.8 kWh/m² per day) 207.9 0.158 0.08 1 500
60.6 500
Onshore wind 0.154 0.25
(≥ 4m/s) (≥ 4.5m/s)
Marine 18 0 0
Notes: Installed capacity includes on-grid and off-grid capacity. Installed capacity of solar does not include solar powered
public street lighting (16.04 MW) and solar powered energy saving lamps (10.90 MW).
Sources: PLN (2021), RUPTL 2021; MEMR (2020), Handbook of Energy and Economic Statistics of Indonesia;
IEA analysis of the Global Wind Atlas and Global Solar Atlas; and 2020 land-use data from the European Space Agency
(ESA) Climate Change Initiative (CCI).
Bioenergy is one of the main focus points for development in the country. There
are currently 45 MW of biogas installations planned for 2024, and ongoing trials
for biomass use. The trials involve using waste pellets, wood pellets, woodchips,
palm kernel shells, sawdust and rice husks. There are plans to make use of
different co-firing rates, namely, 6% for pulverised coal (PC), 40% for circulating
fluidised bed (CFB) and 70% for stoker-type boilers. One of the challenges in
developing biomass co-firing at the existing thermal power plants is the large
amount of biomass supply that would be required and the need to deploy the
corresponding sustainable supply chain as explained below. Another
consideration is the lower efficiency of co-firing in coal plants which increases the
total costs of the power system. On the other hand, this type of retrofit would allow
the re-use of existing dispatchable assets.
Since 2015, Indonesia has had plans for waste-to-energy, but these are yet to turn
in effective generation.
PLN relies on biomass co-firing to achieve

renewable target with little use of solar PV
To achieve the targets set by the government, including a 23% share of
renewables in the electricity mix by 2025 and emissions reductions, PLN
formulated two generation scenarios – optimal mix and low-carbon 4 – for 2021-
2030 in the RUPTL. In both of these two scenarios, the renewables growth
4
PLN’s optimal mix scenario for 2030 considers a mix of 64% coal, 11.5% gas, 23% NRE, 0.4% fuel oil and an additional
potential for 1.2% NRE. The low-carbon scenario for 2030 considers a mix of 59.8% coal, 15.6% gas and 24.2% NRE.
PAGE | 24
between 2021 and 2025 would be 45-46 TWh, with biomass taking a large share
of growth at around 40-42%, followed by hydro and geothermal at 21-25% and 20-
24%, respectively. At the same time, the electricity generated from non-
renewables is also expected to grow by 24 TWh. On the other hand, the share of
solar PV in 2025 is less than 2% despite its declining costs globally and short
installation time, due to the currently high costs of solar PV and PLN’s concerns
about the variability of VRE. 5
Growth in renewables generation in the RUPTL optimal and low-carbon scenarios,

2021-2025
20 000
Change in generation, GWh
18 000 Optimal
16 000
14 000
12 000 Low-Carbon
10 000
8 000
6 000
4 000
2 000
0
Hydro Geothermal Solar Wind Waste Biomass
Sources: IEA analysis of PLN (2021), RUPTL 2021.

To achieve its renewable energy targets, Indonesia is aiming to leverage biomass

co-firing in its existing coal plants that are located primarily in the Java-Bali system.
By 2025, Indonesia expects 13 million tonnes of biomass to be co-fired to achieve
the renewable generation required to meet the 23% target. While this volume falls
within the IEA estimates of the available wood residues at slightly more than 18
million tonnes in 2025, it illustrates the challenge for Indonesia to deploy a
sustainable biomass supply chain to support this target. Co-firing up to 10% would
be the maximum rate that Indonesia aims to implement with its existing fleet, and
up to 20% in power plants built in the future. These low blending ratios present the
advantage of obviating the need for more investment in retrofitting plants.
5
Tambunan H., et. al. (2018), Maximum Allowable Intermittent Renewable Energy Source Penetration in Java-Bali Power
System, 2018 10th International Conference on Information Technology and Electrical Engineering (ICITEE), 2018, pp. 325-
328, doi: 10.1109/ICITEED.2018.8534845.
PAGE | 25
Growth in renewables generation by main island group in the RUPTL Low Carbon
Scenario, 2021-2025
14 000
Change in generation, GWh
Hydro
12 000 Geothermal
Solar
10 000 Wind
8 000 Waste
Biomass
6 000
4 000
2 000
0
Java- Sumatra Kalimantan Sulawesi Maluku-
Madura- Papua-
Bali Nusa Tenggara
Source: IEA Analysis of PLN (2021), RUPTL 2021.

Indonesia also aims to leverage generation from new hydro and geothermal in
Java-Bali and in Sumatra. For the 11.5 TWh of hydro and 9 TWh of additional
thermal generation needed between 2021 and 2025, the planned additional
capacity of 3.16 GW hydro and 1.4 GW geothermal would be sufficient.
The expected growth from wind and solar between 2021 and 2025 is limited at 1.4
TWh and 4.5 TWh, respectively, in the RUPTL. These moderate targets are based
on PLN’s own assessment of their ability to manage variable generation. The
RUPTL considers only utility-scale solar projects. In the meantime however, in
order to accelerate the deployment of renewables and meet the 23% target by
2025, the MEMR has drafted a new regulation (MEMR Regulation 26 of 2021)
dedicated to rooftop solar, which is expected to come into force shortly.
Power pricing regulation limits attractiveness

of renewables investment
PLN and the MEMR use the main indicator biaya pokok penyediaan (BPP), the
average cost of production, as a measure of cost-effective performance of the
power sector. This indicator considers the total system costs associated with
providing a kilowatt-hour of electricity and is calculated on the basis of average
historical accounting.
The total BPP is made up of three components: BPP-pembangkit (generation),

BPP-transmisi (transmission), BPP-distribusi (distribution) with generation being a
major factor for the total BPP. Each of these factors is calculated yearly at sub-
system level and may diverge between regions. In 2020, total BPP (weighted
PAGE | 26
average) was IDR 1348/kWh (USD 92.5/MWh) 6 and BPP-generation was IDR
1027/kWh (USD 70.4/MWh) 7 or 76% of total BPP.
In order to ensure electricity affordability, the MEMR controls the generation

component through fuel and electricity purchase price regulation. On the one
hand, BPP-generation is used as a benchmark for power purchases from IPPs.
On the other hand, the cost of coal is capped at USD 70/tonne through the DMO
resulting in about USD 1.82 to 2.13 billion worth of operating costs effectively
subsidised by the coal producers. Without this subsidy, the electricity generation
cost could be around IDR 100 to 150/kWh (USD 7 to 11/MWh) more expensive. 8
For gas, Regulation 34K/16/MEM/2020 allocates gas volumes to the power sector
and Regulation 91K/12/MEM/2020 sets maximum purchase prices. Given the
large share of coal in the electricity supply, the resulting BPP-generation of the
systems remain low.
As renewables purchase prices are also regulated, it introduces stiff competition

with fossil-fuel based generation. Regulation 50 of 2017 amended by Regulation
4 of 2020 9 sets a maximum purchase price for hydro and geothermal of 100% of
the current BPP-generation of the local system, while for electricity generated from
solar PV, wind, biomass, and ocean wave it is limited to 85% of the BPP-
generation. Local content requirements on power infrastructure development (MOI
Regulation 52/2012) and specifically for solar PV (MOI 5/2017) are estimated to
increase the CAPEX by up to 50% due to the higher module production cost
compared to global markets. Hence, regulation based on a purchase price cap
that is artificially lowered by coal purchase subsidies limits the attractiveness of
renewable investments.
The limited investment in VRE hinders the Indonesian electricity sector from taking
advantage of zero marginal costs which can help reduce the subsidies it is
currently spending. In addition to coal purchase subsidies, consumer tariffs are
also subsidised. To cover the costs of providing electricity to the general public,
low-income households, rural, or remote areas, the government provides a direct
subsidy to PLN to cover expenses not covered by current consumer tariffs. In
recent years, this has ranged between USD 3 and 4 billion.
6
PLN, Statistics book 2020. 1 USD = 14 582 IDR market exchange rate in 2020.
7
Decree of the Minister of Energy and Mineral Resources Number 169.K/HK.02/MEM.M/2021 on cost of electricity supply
for year 2020.
8
Article published on Bisnis.com on 23 March 2018 with the title “Coal Prices Set, Electricity Production Costs Can Be
Reduced by Rp300/kWh” on the consequences of the capped price on coal for power generation.
9
Regulation of the Minister of Energy and Mineral Resources number 4 of year 2020 about second amendment to Regulation
50 of 2017 concerning the utilisation of renewable resources for electricity supply.
PAGE | 27
Historical electricity subsidies, average subsidy rate and BPP, 2010-2020
9 120.0
USD MER billion
Total subsidy
(USD MER/MWh)
8
Average rate
(USD billion)
100.0
7 Subsidy
(USD/MWh)
6 80.0
5 BPP
60.0 (USD/MWh)
4
3 40.0
2
20.0
1
0 0.0
2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020
Notes: BPP = biaya pokok penyediaan or average cost of production. MER = market exchange rate. The average subsidy
rate is based on total subsidies paid over total electricity sales.
Source: IEA analysis of PLN (2014), RUPTL 2014; PLN (2017), Annual Report 2017; and PLN (2021), RUPTL 2021.
Given that increasing the share of VRE would have system-wide impacts and
benefits, it would be important to assess them in the context of these regulatory
control points that distort market values. Reflecting the real market value of coal
can allow PLN to find a more optimal use for its subsidies such as helping lower
the upfront costs of VRE integration. If solar PV displaces coal-fired generation,
excess coal could be sold at higher prices on international markets. Currently, the
record high prices on coal markets represent an opportunity cost for Indonesia.
And although this is short-lived, the prices are expected to stabilise to around USD
20/tonne above the cap in 2022-2024 according to IEA’s coal market report
(December 2021).
There is untapped flexibility potential in the

Sumatra and Java-Bali systems
Key components of power system flexibility include technical flexibility,
contractual/institutional flexibility and operational practices. These components
must be considered simultaneously in the effort to enhance system flexibility.
Prominent technical flexibility resources include power plants (both conventional

and variable renewables), electricity grids (including interconnectors), energy
storage (pumped hydro and batteries) and distributed energy resources (including
demand response and electric vehicles). Grids play a major role in terms of
flexibility: they allow the sharing of resources across larger areas, a feature which
has even more value for variable renewables as their spatial diffusion helps
smooth down variability. Since grids take time to be built, little emphasis is put on
PAGE | 28
grids in the study of Java-Bali and Sumatra in 2025 presented in Chapter 3, but
they will certainly play a growing role in the coming decades.
Contractual/institutional flexibility is the flexibility provided by underlying

institutions in the power sector, including laws and general practices. It plays a
key role in facilitating the optimal use of the technical flexibility resources. Most
advanced economies have power markets to provide signals of the time and
locational value of resources and to incentivise flexibility. For emerging economies
with vertically-integrated structures, such as Indonesia, state-owned electric
utilities are the single buyer, purchasing bulk power from private producers as well
as their own subsidiaries under rigid contractual structures, which can often limit
the use of technical flexibility resources.
In addition to technical and contractual flexibility, modern system operational

practices provide another mechanism to foster the more flexible use of technical
assets, particularly on the supply side, which help to address technical and
economic concerns about a high share of VRE during the clean energy transition.
Some of the key operational practices include real-time monitoring and dispatch,
forecasting and system services. System planning practices could also be
improved so that a long-term flexibility strategy is developed and implemented.
Understanding flexibility requirements, from short-term to long-term, can support

the effective utilisation of, and levels of investment in, different flexibility resources
and the services they provide. This flexibility is particularly important to
accommodate high shares of VRE. The hour-to-hour variations in net demand
(demand minus non-dispatchable VRE generation, a measure for the system
flexibility requirements) of the system and the gap between minimum and peak
net demand are robust indicators of the challenges and flexibility requirements on
hourly and daily timescales, respectively. Currently the shares of VRE in both the
Java-Bali and Sumatra systems are still less than 0.1%. Therefore, the profiles of
net demand are nearly identical to the respective load profiles.
In the Java-Bali system, the daily demand patterns are influenced by the
residential, business and industrial sectors. A typical weekday consists of
morning, afternoon and evening peaks with the minimum demand in the early
morning. On Sundays and public holidays, the majority of the demand is from the
residential activities and the demand profile is relatively stable throughout the day
with the daily peak occurring in the evening. The demand patterns are slightly
different between the dry season (April to October) and wet season (November to
March). In the wet season, there is a greater variability in the demand patterns in
terms of higher ramp rates and a larger gap between daily minimum and peak
demand.
PAGE | 29
Typical demand profile in the Java-Bali system on weekdays and Sundays, dry (left)
and wet (right) season, 2019
Source: IEA analysis of PLN Data.

The highest 1-hour upward ramp in the Java-Bali system occurs on weekday
evenings at 2 615 MW (equivalent to 44 MW/minute) or 12% of the daily peak
demand. The highest 3-hour upward ramp occurs on weekday mornings from
around 07:00 to 10:00 which is 3 739 MW (equivalent to 21 MW/minute) or 15%
of the daily peak demand. Given the size of the Java-Bali system, these ramping
requirements are relatively small.
In Sumatra, the typical daily demand patterns for both weekdays and Sundays are
similar since the residential sector represents the majority of overall demand at
around 60%. The demand is relatively stable during the day with a sharp rise in
the evening peak at around 6-7 pm. The highest hourly ramp is 1 040 MW, which
is equivalent to 22% of the daily peak demand, while the 3-hour upward ramp is 1
440 MW or around 30% of the daily peak demand. Although such ramping
requirements are more challenging than those in the Java-Bali system, they are
still manageable given the reasonable share of hydropower and gas-fired
generators in Sumatra, which are considered relatively flexible. In other systems,
such as those in California and India, 3-hour ramps can be as high as 60-70% of
the daily peak demand.
PAGE | 30
Demand profile during periods with high ramping requirements in the Sumatra system,
2018
6 000
MW
5 000
4 000
3 000
2 000
1 000
0
Thu Thu Fri Fri Sat Sat Sun Sun Mon Mon Tue Tue Wed
16/08 16/08 17/08 17/08 18/08 18/08 19/08 19/08 20/08 20/08 21/08 21/08 22/08
00:00 12:00 00:00 12:00 00:00 12:00 00:00 12:00 00:00 12:00 00:00 12:00 00:00
Note: The maximum hourly ramping requirement in the Sumatra system occurred on Sunday 19 August from 17.00-18.00,
while the maximum 3-hour ramping requirement occurred on Friday 17 August from 15.00-18.00.
Source: IEA Analysis from PLN Data.
Typical ramping requirements in Java-Bali and Sumatra systems
Java- Bali system Sumatra
2019 2025 2019 2025
1-hour 3-hour 1-hour 3-hour 1-hour 3-hour 1-hour 3-hour
Max ramp
2 615 3 739 3 520 4 300 1 096 1 476 1 564 2 065
up (MW)
% of daily
12% 15% 13% 16% 22% 31% 19% 29%
peak
Season Wet Wet Dry Wet Dry Dry Dry Dry
Weekday Weekday Weekend Weekend Weekend Weekend Weekday Weekend

Period
evening morning evening evening evening evening evening evening
Max ramp
down 3 212 4 460 7 107 9 319 1 042 1 568 1 280 2 194
(MW)
% of daily
17% 23% 35% 46% 22% 31% 17% 29%
peak
Wet-Dry
Season Dry Dry Dry Wet Dry Dry Dry
Transition
Weekend Weekday Weekend Weekend Weekend Weekday Weekday Weekday
Period
evening evening evening evening evening evening evening evening
Note: Analysis is based on demand data from four representative days, the daily peak load in Java-Bali in 2019 and a full
year of hourly demand data for Sumatra in 2018, projected to years 2019 and 2025. The dry season is April to October. The
wet season spans November to March.
The gap between daily minimum and peak demand in the Sumatra system is in
the range of 1 000 – 2 100 MW in 2019, which can be as high as 40% daily peak.
In 2025, the gap is expected to increase to 1 500 – 3 000 MW or 38% of the daily
peak. The larger the gap between minimum and peak demand, the greater the
PAGE | 31
flexibility requirements and operational challenges for the system, which typically
translate into frequent starts/stops of power plants.
With the smaller system size and the demand patterns, it appears to be more
challenging to meet the flexibility needs of Sumatra compared to Java-Bali from
the system operations perspective. Hence integrating VRE, particularly solar PV,
into the Sumatra system would require greater flexibility resources. This is
explored in detail in Chapter 3 where we discuss the scenarios with high solar PV
in 2025. The planned interconnection between the two systems (at the earliest by
2028) will contribute to smoothing out variability and reducing the ramp rates of
the combined system.
Dispatchable power plants, both fossil-fuel and hydropower, have been the
primary flexibility resources in many power systems around the world, including
Indonesia. They are operated subject to their technical capability, which typically
includes the minimum stable level at which a specific generator can operate, the
rate at which power output can be adjusted (the ramp rate), the start-up and
shutdown times, and the constraints on how often a generator can be cycled
(minimum up/down times as well as number of start-ups). Dispatchable power
plants will be required to vary their generation outputs more significantly due the
growing share of VRE and the associated variability and uncertainty of net demand
as a result.
In Indonesia, power plants are still dispatched based on a merit order established
according to the traditional categorisation of baseload, intermediate and peak.
This categorisation is not only subject to technical characteristics, but also
contractual constraints (more information provided in the next section).
Geothermal and coal-fired plants operate as baseload leading to relatively stable
outputs throughout the day. Gas-fired power plants are dispatched to
accommodate changes in the demand throughout the day, similar to hydropower
plants, subject to water inflow patterns. In addition, gas-fired generation also
makes up the main difference in generation between the wet and dry season when
hydro availability is low.
PAGE | 32
Typical weekly generation profiles by technology in the Java-Bali system in the dry
season, wet season and week of Eid (June), 2019
IEA. All rights reserved
With the rising share of VRE in the power system, thermal power plants in
Indonesia, which were originally designed to operate as baseload, are expected
to operate more flexibly to accommodate the variability arising from VRE. A range
of strategies can make existing conventional power plants more flexible. These
can be categorised into two areas: changes to operational practices, including
contractual structures, and investment in flexibility retrofits. From international
experience, the operating characteristics of conventional power plants can be
significantly improved after retrofitting. However, the coal-fired power plants in
Indonesia are relatively young and have flexible characteristics.
PAGE | 33
Average operating characteristics of conventional generation technologies
Minimum operating levels (% Ramp rate

Warm start time (hours)
of capacity) (MW/minute)
Technology
Typical Retrofit Typical Retrofit Typical Retrofit
Coal 37% 20% 21 60 6 2.6
CCGT 45% 30% 21 56 1.6 0.5
OCGT 35% 20% 29 60 0.7 0.3
Hydro 15% - 60 - N/A -
Notes: CCGT = combined-cycle gas turbines; OCGT = open-cycle gas turbine.

Sources: IEA (2017), Energy Technology Perspectives 2017; NREL (2012), Power Plant Cycling Costs 2012; Gonzalez-
Salazar et al. (2018), “Review of the operational flexibility and emissions of gas- and coal-fired power plants in a future with
growing renewables”; Siemens (2017), Flexibility of Coal and Gas Fired Power Plants; Agora Energiewende (2017), Flexibility
in Thermal Power Plants.
Hybrid power plants, which combine two or more technologies, are an emerging
trend in Southeast Asia, including Indonesia, Thailand and Viet Nam. The 145 MW
Cirata hydro floating solar PV power plant in Java-Bali combines solar PV with a
hydro plant. It is one of the largest floating solar PV systems in the world and has
the potential to contribute to system flexibility from the technical perspective by
balancing the variability from the PV cells, particularly during the rainy season. At
the time of writing, this project is expected to be in operation by the end of 2022,
and is viewed as a pilot for the development of hybrid floating solar PV plants in
many other islands to meet the renewable targets.
Electricity grids are another main flexibility resource that enables different sources
to be shared across the regions. For Indonesia, the existing transmission system
is regarded as one of the barriers to flexibility, due to the archipelago and volcanic
characteristics of the islands. In the Java-Bali system, which is the main system in
Indonesia, the transmission grid has a radial structure (long, thin and low density)
from east to west with the demand centres in the north and south of the island,
separated by mountains and volcanoes in the centre. Reinforcements to existing
grids as well as interconnections between islands and with neighbouring countries
could deliver significant flexibility improvements. A study of the ASEAN Power Grid
also demonstrated that achieving sustainable development goals requires
developing both grids and renewables, lest the excess thermal capacity be
exported and thereby increase the overall emissions in the region.
Contractual structures limit the flexibility of

the young thermal fleet
In its role as single buyer in the Indonesian power system, PLN procures power
from its own generation assets as well as IPPs and leased generation assets. IPPs
enter into long-term PPAs with PLN, providing them the budget security for project
PAGE | 34
financing. As in other emerging economies, physical PPAs for thermal generation

have historically been an effective tool to ensure investment in generation to meet
the rapidly growing power demand in Indonesia. For the system, the cost of
dispatching generators can be viewed as comprising two components: a capacity
payment and an energy payment. If the capacity payment is large and the energy
payment is low, as can be the case for coal-fired generation in the absence of
costs for environmental externalities, the majority of the generation is paid for up
front, and consequently the system operator has an incentive to keep running the
generation for which they have already paid the costs.
In physical PPAs, it is common practice to agree on a certain amount of generation

(minimum offtake obligation) at a fixed price in order to secure the income of
the generator. This legal agreement suits systems seeking a baseload from a
thermal fleet. Unfortunately, the generator has no financial incentive to provide
flexibility and supplement the infeed from VRE when the system needs it, which is
an important feature to enable integration of higher shares of variable renewables.
For thermal generation, the guaranteed take obligations are often relatively high,
well above the technical capabilities (minimum stable levels) of the assets. The
PPAs often also specify other operating parameters in ranges that restrict their
flexibility: ramp rates and start-up time. This barrier for flexibility has also been
observed in other countries like India and Thailand.
Given the relatively young thermal fleet in Indonesia, the physical PPAs can
potentially limit the use of technical flexibility for many years to come. According
to the RUPTL, the total capacity of coal-fired IPPs in the Java-Bali system is
planned to grow from 10 GW in 2020 to 19 GW by 2025, while the peak load is
expected to grow from 25 GW to around 30 GW by 2025. Based on a typical
lifespan of 30 years for coal plants, only two units would retire before 2030 and 14
GW would still be in operation beyond 2040. PPAs are usually established for a
duration covering the technical lifetime of the plant.
Fuel supply contracts can also be a cause of inflexibility. Many thermal plants
enter into long-term contracts for fuel supply. These are especially prevalent with
fuel suppliers that need revenue certainty to support investment in related
infrastructure. A notable example of this is with gas suppliers, where gas pipelines,
LNG terminals and storage are all capital intensive, long-term investments. In
Indonesia, ToP 10 clauses in gas supply contracts are common 11. This effectively
means that the fuel can be considered a sunk cost, modifying the marginal cost of
10
Take-or-pay [obligations] is the term most commonly used in the gas market, while guaranteed (off)take obligations are
used in the power market. Both terms imply a financial commitment to either pay for gas or run generation physically at a
power generator.
11
Hakam, D. F., and Asekomeh, A. O. (2018). Gas Monetisation Intricacies: Evidence from Indonesia. International Journal
of Energy Economics and Policy, 8(2), 174–181.
Retrieved from https://www.econjournals.com/index.php/ijeep/article/view/6005.
PAGE | 35
the plant, as it has to be paid for irrespective of generating power with the fuel or
not. In practice, these plants may be dispatched ahead of low-cost renewable
technologies until carbon pricing is introduced. From a system operator
perspective, it may be a rational choice under these circumstances to curtail wind
and solar to run coal in a system with minimum take obligations, due to the
uncertainty in forecasting wind and solar and the imperative to maintain system
stability. However, from an environmental perspective it would still be optimal to
dispatch renewables ahead of the coal fleet.
Summary of the flexibility challenge in

Indonesia
Increasing the share of VRE requires increasing the flexibility of Indonesia’s power
system, particularly in Java-Bali. The main flexibility challenges of most power
systems, including in Indonesia, can be grouped into three main building blocks:
technical, contractual/institutional and operational practices. Chapter 3 examines
some of these challenges in greater detail in order to assess the impact they have
on the ability of the systems of Java-Bali and Sumatra to accommodate higher
VRE shares in 2025, and on the corresponding costs and emissions. Chapter 4
explores a few options to address some of these challenges.
Technical inflexibility
• Inflexible thermal power plants: As some generation technologies (coal,

nuclear, geothermal) are more efficient at high capacity factors, their higher costs
per unit of output may be a barrier to flexibility. As shown in Chapter 3, the thermal
power plants in Java-Bali and Sumatra will be no obstacle in 2025 because the
ramping requirements remain within the capability of the existing assets, without
the need for retrofits.
• Grid infrastructure: The failure of the Ungaran-Pemalang transmission line in
Central Java contributed to the August 2019 blackout that cost an estimated IDR
90 billion in damage. It demonstrated that the network has weak connections
relative to the power demands. The RUPTL 2021, stated plans to create a new
500-kV Northern Java corridor that would run parallel along the existing lines in
Banten, West Java, Central Java and East Java.
• Geographical topology: The shortest distance between the two coastal cities of
western and eastern Java, Cilegon and Banyuwangi, is 960 km, or approximately
the same distance between Washington DC and Chicago under the PJM system
of Pennsylvania, New Jersey and Maryland. The general direction of electricity
flows is from the east, where there is abundant coal and excess capacity, towards
the higher demand regions in the west. Such a distance, combined with limited
transfer capacities, can pose challenges in terms of voltage stability.
PAGE | 36
Contractual and institutional frameworks
• The PPAs in Indonesia often have clauses specifying minimum take obligations
which are relatively high. This is one, or the main, source of system inflexibility in
Indonesia because it forces PLN to prioritise power generation from these plants
when optimising its dispatch. Such conditions make it uneconomical for PLN to
make use of VRE with its lower operating costs. With increasing shares of VRE,
there could be more VRE curtailment.
• Fuel supply contracts can indirectly limit the operational flexibility of power plants
through minimum offtake and penalties, especially when the fuel suppliers require
additional certainty in their investments. In Indonesia, ToP clauses, which are
common in its gas supply contracts, also constrict PLN’s ability to take advantage
of the operational flexibility of gas-fired power plants.
System operational practices

• Overly optimistic demand forecasts: The excess capacity in the Java-Bali
system and high reserve margin have been in part due to overly optimistic demand
forecasts. These forecasts resulted in programmes such as the 10 000 MW Fast
Track Programme 1 in 2006, the 17 918 MW Fast Track Programme 2 in 2010,
and the 35 000 MW Fast Track Programme in 2014 where regulations and permits
were simplified and financing was made available for both PLN and IPPs to build
up new capacity rapidly. Lower than expected electricity demand resulted in the
high observed reserve margins, forcing PLN’s power plants to run at low capacity
factors. Adding renewable capacity appears uneconomical unless some thermal
capacity is retired or is allowed to run at even lower capacity factors.
• Control systems and real-time monitoring: There is still limited deployment of
advanced monitoring infrastructure (AMI) and real-time control of generators,
electricity networks and high-voltage substations. AMI deployment provides better
insights into demand patterns, as well as more efficient operations and better
modelling of the power system. The latter is important as the system operator can
better assess the system performance, and plan and prepare for a wider array of
scenarios. PLN is currently upgrading control rooms to include load frequency
control and automatic generation control in the Java-Bali system to balance supply
and demand.
• Reserve margin and system services requirement: Although the probabilistic
technique of LOLP is used as the reliability criterion to determine the required
reserve margin in Indonesia, the existing approach is still largely based on
deterministic criteria. The high reserve margin criteria (35%), together with the
optimistic demand forecast, has resulted in overcapacity.
• Grid code: Prior to the grid code revision of 2020, there were no explicit terms
covering generation from VRE, nor specifications on fault ride-through, and low-
and high-voltage ride-through. A distribution grid code is under preparation to
bring further improvements.
PAGE | 37
Enhancing Indonesia’s Power System Chapter 3 - Energy transition pathways for
Pathways to meet the renewables targets in 2025 and beyond Java-Bali and Sumatra in the short term
Chapter 3. Energy transition

pathways for Java-Bali and
Sumatra in the short term
Key findings – Java-Bali and Sumatra power systems in 2025

• Delays in deploying new geothermal and hydro capacity, and in establishing a
sustainable supply chain to support biomass co-firing in coal plants (which
comes at a price premium) put the realisation of the target for a 23% share of
renewables in electricity in 2025 at risk. This risk could be mitigated by
deploying solar PV at a faster pace.
• To meet the 23% renewables target, the SolarPlus Scenario, in which 17.7 GW
of solar PV in the combined systems of Java-Bali and Sumatra (against 2.8
GW in the RUPTL) are deployed to compensate for non-committed new
capacities, a possible failure to deploy biomass co-firing appears to be viable
from a system operation perspective. Flexibility requirements are higher
compared to the Base Scenario (where the assets are consistent with the
RUPTL), especially in the smaller Sumatra system, but remain within the
capabilities of the existing and planned dispatchable assets.
• Compared to the Base Scenario, the SolarPlus Scenario leads (without taking
into consideration carbon pricing) to 2% savings in operating costs per year
(essentially through avoided fuel costs), which could be used to implement new
flexibility measures in the future, and a 4.5% decrease in CO2 emissions, at
the expense of a 14% PV curtailment rate in Sumatra.
• The total costs of the SolarPlus Scenario are higher than the Base Scenario
due to the higher investment cost for PV technology. Authorities can take
various measures to bring down the cost of solar PV with a concrete plan to
deploy PV.
• Given the abundant generation capacities, the Java-Bali and Sumatra systems
can accommodate the electrification of cooking and road transport to meet the
ambitious national targets. Even though most of the additional demand in Java-
Bali is fulfilled by the thermal fleet, the cross-sectoral emissions are reduced
by 1% with all new loads electrified, while solar curtailment is reduced by 6%.
• The main barrier to higher gains from solar PV is the inflexibility in the PPAs
with coal IPPs which cover a significant share of the demand. Removing
contractual inflexibilities would more than double the emissions savings from
PAGE | 38
the SolarPlus Scenario (relative to the Base Scenario) and reduce solar PV
curtailment to negligible rates.
• Given the over-built thermal capacity in the system, adding new generation
capacities to the system leads to stranded assets. This study does not examine
the financial implications of the excess capacity. It only illustrates the limited
room for renewables in the Java-Bali and Sumatra systems in 2025. In the
absence of a coal phase-out programme, and despite the expected economic
growth, the efficiency of any new generation project will be affected.
This chapter provides a detailed techno-economic assessment of the capability of

the current Java-Bali and Sumatra systems to integrate the target of a 23% share
of renewables in the electricity mix by 2025, notably with solar PV. Thanks to its
potential speed of deployment, solar PV technology was identified as a robust
choice to fill the gap between the actual deployment track of renewable generation
and the 2025 target. Wind power is kept at the levels planned in the RUPTL.
For the purpose of illustrating system performance, this study looks at capacities
of solar PV which are six times higher than the current plans (17.7 GW instead of
2.8 GW), as a high-end estimate of how much solar PV capacity would be useful
to fulfil the target, regardless of how realistic the deployment pace would be to
reach these levels. In order to assess the ability of these systems to handle the
challenges relating to an increase in VRE deployment, in the short term the impact
of the transition to VRE was analysed from the perspectives of the three key
sustainability pillars of the Indonesian energy transition policy, namely, Availability
(security of supply); Affordability (least cost) and Acceptability (environmental
sustainability).
The metrics used to quantify the system benefits and value of flexibility options
are: reductions in the cost of fuel, operation and maintenance (O&M), and carbon
dioxide (CO2) emissions. Other metrics include the level of VRE curtailment,
contribution of different sources to system adequacy and system services
including system inertia and ramping requirements.
Modelling of the Java-Bali and Sumatra

systems
To better understand the flexibility requirements of the Java-Bali and Sumatra
systems as Indonesia transitions towards higher shares of VRE, a production cost
PAGE | 39
model 12 of the two power systems was developed. Key inputs were forecasted
demand profiles, the techno-economic characteristics of the power plants,
hydropower energy constraints, transmission network constraints and VRE
generation profiles.
The modelling of the power systems in Java-Bali and Sumatra is predominantly

based on data requested and received from Indonesian stakeholders (the MEMR
and PLN) with key information on the different components of the power system.
In the absence of specific data for either Indonesia or a specific component of the
power system, either public domain information or assumptions based on best-
practice were made to allow for completeness of the model.
The study is based on the planned systems in Java-Bali and Sumatra in 2025 as
per the latest Electricity Business Plan (RUPTL 2021) that aims to achieve the
national renewable target of a 23% share of renewables in the electricity mix by
2025. 13 Therefore, the supply and demand outlook as per the RUPTL in 2025 form
the basis of the model and the Reference or Base Scenario apart from one
notable exception: the share of biomass in the electricity mix in the Base Scenario
is 2.2% (against 3.5% in the RUPTL) although the installed capacities are the
same. With a level of bioenergy contribution as in the RUPTL, the 23% target
would not be achieved. Since co-firing takes place in PLN’s coal plants at the
blending ratio of 10%, reaching the RUPTL bioenergy share requires a yearly
capacity factor of those plants (with a capacity of around 12 GW) around 80%,
leading to an even greater share of coal-fired power and higher fuel costs. In
contrast, the Base Scenario assumes economic dispatch of the thermal fleet with
the contractual constraints described below.
The model was developed with an hourly temporal resolution, using projected
demand profiles based on demand forecasts for both annual energy and peak
demand from the RUPTL 2021, with load shapes based on historical profiles. 14 In
addition, the model captures the techno-economic characteristics of both existing
and planned power plants, fuel supply (including hydropower) constraints,
transmission and VRE production profiles. While the model is deterministic,
operating reserve requirements are captured based on load risk and renewable
forecast errors to ensure sufficient capacity is reserved for both balancing and
spinning reserves. In addition, constraints in PPAs with coal IPPs and ToP
contracts for the supply of gas are implemented using assumptions derived from
12
A production cost model simulates detailed operation of a given power system by co-optimising hourly (or sub-hourly)
dispatch and procurement of reserves.
13
The 23% share of NRE is at the country level. Different island-power systems are expected to have different contributions
to achieving this NRE target. In the RUPTL’s Low Carbon Scenario, Sumatra’s NRE share is 43.6% while that of Java-Bali
is 17.1%. Given that the combined systems represent 80% of the country’s demand for electricity, the study aims to fulfil the
23% target by covering Java-Bali and Sumatra together.
14
Historical profiles for the Java-Bali system are based on the demand for four representative days from 2019, up-scaled
using the daily energy profile across the entire year (365 days). Meanwhile the historical profile for the Sumatra system
included a full hourly profile (8 760 hours) for 2018.
PAGE | 40
historical (2019) data. The annual minimum offtake for coal IPPs (set to 60%) 15
was based on the range of capacity factors of IPPs in 2019 in both Java-Bali and
Sumatra while allowing the simulation outcomes for the Base Scenario in 2025 to
meet the renewables target. Gas contracts are assumed to include daily ToP
obligations at the regional level scaled up from the 2019 consumption according
to the capacity growth of gas-fired plants.
The Java-Bali and Sumatra systems are disaggregated into five and two regions,
respectively. For the Java-Bali system, the modelling regions are in accordance
with the control regions of PLN, while for the Sumatra system, two regions are
assumed: North (SMN) and South (SMS), based on the provinces and the location
of the key bottleneck in the north-south 275 kV transmission corridor between
Jambi and Sumetara Barat provinces. The high-level transmission network
topologies (150 kV, 275 kV and 500 kV) connecting different operating regions are
modelled. In 2025, the Sumatra and Java-Bali systems will still not be
interconnected and will therefore be operated independently of one another.
15
Experience from neighbouring countries supports values above 60%. Such capacity factors, however, would not allow
meeting the 25% targets in 2025 unless unrealistic assumptions about the capabilities of other types of units in the system
were made. No capacity factor constraint is added to the plants owned by PLN and its subsidiaries.
PAGE | 41
Representation of the Java-Bali and Sumatra power systems
This map is without prejudice to the status of or sovereignty over any territory, to the delimitation of international frontiers and boundaries and
to the name of any territory, city or area.
Source: IEA Analysis of PLN (2021), RUPTL 2021.

PAGE | 42
Alternative pathways towards 2025 and their

impacts on the sustainability targets
In addition to the Base Scenario in 2025, the analysis considers additional
scenarios that explore different pathways for Indonesia to meets its renewables
target in 2025. The additional scenarios revolve around the following main
sensitivities:
• Renewables mix, based on both the status and lead-time of renewable

technologies in the RUTPL and ensuring that Indonesia meets its renewable
targets for 2025
• Additional demand due to electrification of new end-uses (clean cooking and
electric vehicles)
• Contractual constraints on thermal plants which limit their flexibility.
Three scenarios look at various mixes of renewable sources around the Base
Scenario. Two of them focus on the role of biomass co-firing: how the system
performance and renewable targets are affected by either the imperative of
meeting the electricity share of biomass in the RUPTL (Enforced Co-firing
Scenario) or if the biofuel supply chain was not in place (No Co-firing Scenario).
The central scenario of the study is the SolarPlus Scenario, the purpose of which
is to provide a sense of the potential value and impact of solar PV in the power
systems compared to the other renewable and non-renewable technologies for
which there are planned but uncommitted capacities in the RUPTL 2021. Due to
the long construction times of hydro and geothermal, and the uncertainty around
deploying the biofuel supply chain logistics for biomass co-firing, solar PV is
identified as a technology with high potential to ensure Indonesia meets the
renewable targets for 2025. This is especially so under the right enabling
conditions, which can allow for relatively short leadtimes for its deployment.
Alternative scenarios (electrification scenarios) then consider the potential to

include targets for electrification of end-uses (clean cooking, electric vehicles)
which, up until now, have not formed part of the RUPTL despite being on the
government agenda, and their impact on CO2 emissions, in particular. Indeed, the
government strategy aims at reducing the use of oil products in final energy
consumption through electrification of cooking (to replace traditional biomass and
LPG) and road transport.
Finally, another set of scenarios (scenarios with contractual flexibility) consider

the impact of removing the assumed contractual constraints in PPAs and on gas
supply contracts.
PAGE | 43
Scenario settings in 2025
Storage Share of RE (VRE)

Scenario Electricity mix Demand
and EVs Sumatra and Java-Bali
2025 Enforced co-firing RUPTL RUPTL RUPTL 23% RE (1.9% VRE)
RUPTL
2025 Base (with economic RUPTL RUPTL 22% RE (1.9% VRE)
dispatch of co-firing)
2025 Without co-firing No Co-firing RUPTL RUPTL 20% RE (1.9% VRE)
No co-firing
2025 SolarPlus RUPTL RUPTL 25% RE (10% VRE)
High share of solar PV
2025 SolarPlus No co-firing 25% RE (10% VRE)

RUPTL Clean cooking
+ Clean cooking High share of solar PV
2025 SolarPlus No co-firing
EVs RUPTL 25% RE (10% VRE)
+ EVs High share of solar PV
2025 SolarPlus No co-firing
EVs Clean cooking 25% RE (10% VRE)
+ Clean cooking + EVs High share of solar PV
No co-firing
2025 SolarPlus High share of solar PV
RUPTL RUPTL 28% RE (10% VRE)
With contractual flexibility Thermal plants
contractual flexibility
Enforced co-firing: This scenario is the most compliant with the plans in the
RUPTL. The coal-fired plants where biomass co-firing takes place (owned by PLN,
with a capacity of 12 GW) are forced to run at least 80% of the time to equal the
bioenergy share in final energy in the RUPTL.
Without co-firing: This scenario assumes that biomass co-firing is not yet
possible in 2025 due to either fuel supply logistics, or other technical reasons that
would prevent the designated coal-fired capacity to co-fire the expected 10%
biomass in coal plants for generation purposes. In this scenario, the capacity
assigned to co-firing in the RUPTL 2021 is therefore maintained as normal coal
capacity and those plants are not forced to run to meet a given capacity factor.
SolarPlus: This scenario assumes that all non-committed generation capacity

from the RUPTL 2021 (according to the unfulfilled government quota) for
deployment up until 2025 is replaced with utility-scale solar PV generation of
approximately the same energy output. Similar to the No Co-firing Scenario,
biomass co-firing is maintained as coal capacity. In addition, most of the non-
committed generators are geothermal and hydropower plants, and hence are
regarded as a large uncertainty in Indonesia’s ability to meet its national renewable
target of 23% by 2025 due to long lead times. This scenario therefore leads to a
solar PV capacity of 18 GW 16 for the combined systems of Java-Bali and Sumatra.
16
Hourly solar generation profiles are simulated from selected sites in Sumatra and Java-Bali based on resource potential
distance to the grid and distance to demand centres. No rooftop solar is considered.
PAGE | 44
The next three scenarios are variants of the SolarPlus Scenario and have the
same generation capacity. The demand is increased in comparison to the RUPTL
Base Scenario, which does not include these new end-uses of electricity despite
being part of the government strategy. The SolarPlus Scenario and the
electrification variants all entail a very optimistic deployment pace for new assets
(solar PV and new electric uses), but these scenarios are illustrative of the impact
of these technologies on the system.
(Electric) clean cooking in Indonesia will be scaled up substantially as there is a

nationwide programme to promote electric cooking to reduce dependency on
traditional biomass and imported LPG. The increase in electric cooking will have
an impact on the power system with the increase in both total and peak demand
of electricity. In this case, we modelled demand profiles for electric cooking in
various regions in Sumatra and Java-Bali according to IEA’s Sustainable
Development Scenario (approximately 20% of households).
Electric vehicles (EVs) will have a major role to play in clean energy transitions
across Indonesia. There are policies to support the development and adoption of
EVs, including the national EV roadmap with an ambitious target for 2050. By
2025, more than 370 000 electric 4-wheelers and 11.8 million electric 2-wheelers
are targeted to be on the roads of Indonesia. The uptake of EVs will have
implications for the power sector with the rise in both total and peak demand. At
the same time, EV charging represents an opportunity to accommodate a higher
share of solar PV, especially if managed charging is considered, which is not the
case here.
We also consider a scenario that combines the electrification of both end-uses

(clean cooking + EVs).
Finally, we explored a variant of the SolarPlus Scenario where the contractual

constraints of thermal power plants are relaxed (Contractual Flexibility
Scenario). In all the previous scenarios, assumptions were made about the extent
of the inflexibility in contracts. In particular, the minimum capacity factor of IPPs
was set to a level (60%) which was compatible with reaching the 23% target in the
Base Scenario, and gas units were constrained by a contract requiring a daily take
of gas at the regional level matching the volumes of the RUPTL. In this scenario,
the annual capacity factor of IPPs is allowed to be lower than 60% and gas supply
constraints are set to annual levels instead of daily.
PAGE | 45
Impact of the scenarios on plant capacity factors
90% 2019
80%
70%
60% 2025 Base Scenario
50%
40%
30% 2025 Enforced Co-firing
Scenario
20%
10%
2025 No Co-firing
0% Scenario
2025 SolarPlus Scenario
The potential of solar PV to provide a low-risk

pathway to achieve the 2025 renewable
targets
Between 2019 and 2025, the RUPTL 2021 expects the annual electricity demand
to grow by 19% in the Java-Bali system and by 52% in the Sumatra system, while
peak electricity demand would increase by 18% (4.9 GW) and 49% (2.6 GW) in
Java-Bali and Sumatra, respectively. In order to both achieve the renewable
targets and meet growing electricity demand, the RUPTL details the necessary
capacity additions between 2021 and 2025, and calls for certain coal plants to
begin co-firing biomass at the ratio of 10%.
In the Java-Bali system, capacity additions according to the RUPTL 2021 are still
focused on coal, but with notable shares of gas as part of its plan to increase
flexibility. Solar and wind capacity additions are also expected but will contribute
only a small part of the system in the next decade. In terms of generation capacity,
renewable energy represents around 25% of the total capacity (excluding co-
firing), of which wind and solar PV represent about 5%.
The Enforced Co-firing Scenario is established according to these projections,

with approximately 23% of generation coming from RE, of which 2% comes from
VRE. The Reference (Base) Scenario has the same capacities but does not force
co-firing plants online, falling short of the 23% target.
PAGE | 46
Annual demand (left) and peak (right) in the Java-Bali and Sumatra systems in 2019
and 2025 according to the RUPTL
250 35
TWh
GW
30
200 2019
25
150 20
100 15
10
50 2025 Base
5
Scenario
Java-Bali Sumatra Java-Bali Sumatra

Capacity additions in the Java-Bali and Sumatra systems according to the RUPTL
60 Java-Bali 16 Sumatra PSH

GW
GW
14 Bioenergy
50
12 Wind
40
10 Solar
30 8 Geothermal
6 Hydro
20
Gas
4
10 Oil
2
Coal
0 0
2020 Additions Retirements 2025 2020 Additions Retirements 2025
Notes: Biomass co-firing is included in the coal capacity (in existing plants). PSH = pumped storage hydro.
Source: IEA analysis of PLN (2021), RUPTL 2021.
PAGE | 47
Capacity mix and the share of annual generation for the different modelling scenarios
in 2019 and 2025
100% Storage
90%
80%
70% Wind
60%
50% Solar
40%
30%
Bioenergy
20%
10%
0% Geothermal
2025 Enforced Co-firing
2025 No Co-firing
2025 Enforced Co-firing
2025 No Co-firing
2019
2019

2025 Base Scenario
2025 Base Scenario

Hydro
Scenario
Scenario
Scenario
Scenario
Oil
Gas
Co-firing
Coal
Energy Capacity
Note: The energy contribution of the biomass co-firing capacity is split into its different fuel components of coal and bioenergy.
PSH = pumped storage hydro.
The main assumption regarding the share of renewables in the SolarPlus

Scenarios is that the 23% RE target in 2025 is achieved in the combined systems
of Sumatra and Java-Bali.
Total installed capacity in the SolarPlus Scenario (and variants) is 81 GW for the
Java-Bali and Sumatra systems, which is higher than the 68 GW capacity in the
Base Scenario as a result of the lower capacity factor (CF) of solar PV plants
(~18%) relative to both geothermal (80%) and hydropower (~40%) plants in the
modelled system. The total VRE capacity in the SolarPlus Scenario is therefore
18 GW across both Java-Bali and Sumatra, compared to just 3 GW in the Base
Scenario.
The overall share of VRE is around 10% (15% in Sumatra and 9% in Java-Bali).
PAGE | 48
Installed capacity of each generation technology across the different modelling

scenarios in 2019 and 2025 in the Java-Bali and Sumatra systems
90 Storage
GW
80 Wind
70
Solar
60
50 Bioenergy
40 Geothermal
30
Hydro
20
10 Oil
Gas
2019 2025 Base 2025 Enforced 2025 No Co-firing 2025 SolarPlus
Scenario Co-firing Scenario Scenario Co-firing
Scenario Coal
Note: PSH = pumped storage hydro.
Share of renewables and variable renewables (VRE) in electricity production, 2019 and
2025 scenarios

PAGE | 49
Flexibility requirements increase with higher

shares of VRE
As the share of VRE in the Indonesian power system increases, there will be an
increase in the need for flexibility to operate the system, driven by the specific
characteristics of VRE resources. For example, the variability in supply due to
changing weather conditions across multiple timescales (for example, daily,
weekly or seasonal) will begin to drive the operation of the power system.
Examples of this could be in the way that dispatchable generation is operated, or
the way that power flows across the grid.
Several indicators can demonstrate the growing need for flexibility in a system.
One way in which increased VRE deployment, and more specifically solar PV,
affects power system operation is through increased daily cycling of dispatchable
generation in order to accommodate the solar production and the greater swings
in net demand. The technical constraints of generators may limit the flexibility of
certain generators, due to aspects such as minimum generation levels for stable
operation, constraints in the cycling of units (for example, minimum up or down
times), start-up times and ramp rates.
One of the implications of larger shares of solar PV is that systems begin to

experience higher ramping requirements, both in terms of instantaneous ramp
rates as well as over more prolonged periods such as 3-hour ramps. In 2019, with
negligible deployment of VRE capacity, ramping requirements were driven solely
by variations in demand. However, by 2025 ramp rates increase for all scenarios.
Notably, the Java-Bali and Sumatra systems both have sufficient flexibility to
handle these ramp rates through planned and existing dispatchable capacity.
Change in the ramping requirements under different scenarios in 2019 and 2025

PAGE | 50
Another indicator of flexibility requirements is the gap between daily minimum

and peak net demand, which is indicative of the swings in net demand, and which
widens with the deployment of more VRE. This gap between daily minimum and
peak net demand drives the operation of dispatchable generation and leads to
more frequent start-ups, shutdowns and the cycling of generators in order to
accommodate the intra-daily variability in the supply-demand balance. In 2019,
this gap is driven primarily by daily variability in demand, and is 20-31% in the
Java-Bali system and 25-40% in the Sumatra system. In 2025 (Base Scenario),
this daily gap increases in absolute terms but so does the peak demand.
Therefore, this gap actually remains almost in the same range in both the Java-
Bali (19-31%) and Sumatra (26-39%) systems. However, with an accelerated
deployment of solar PV (SolarPlus Scenario), this daily gap grows significantly in
both the Java-Bali (24-60% of daily peak demand) and Sumatra (38-71% of daily
peak demand) systems.
Evolution of the daily gap between minimum and peak net demand as a percentage of
peak net demand in the Java-Bali and Sumatra systems
Java-Bali Sumatra
90% 90%
80% 80%
70% 70%
60% 60%
50% 50%
40% 40%
30% 30%
20% 20%
10% 10%
0% 0%
Oct
Nov
Dec
Mar
May
Apr
Jul
Jan
Jun
Aug
Sep
Feb
Oct
Nov
Dec
Mar
May
Apr
Jul
Jan
Jun
Aug
Sep
Feb
2019 2025 Base Scenario 2025 SolarPlus Scenario
While these larger swings in net demand do pose a challenge to these systems,
both have sufficient flexibility in 2025 to accommodate these changes. The
Sumatra system will experience greater variability than the Java-Bali system due
to its demand shape, its higher share of VRE (15%) in the system and its smaller
system size. However, the model shows that Sumatra would be able to
accommodate these higher shares of solar PV due to the inherent flexibility of its
thermal and hydro fleet, with no negative impact on the system reliability (no
unserved energy) and at the expense of solar PV curtailment (14.5% annually in
Sumatra, equivalent to 3.4% over both the Java-Bali and Sumatra systems
combined) occurring in periods of both low demand and peak solar production
PAGE | 51
around midday. Curtailment takes place when demand drops below the minimum
generation levels, which are defined by the system needs (reserves and flexibility
needs) and technical and contractual constraints of the thermal fleet. Curtailment
could be reduced at the cost of spilling hydro resources from run-of-river plants. 17
Generation profiles by fuel type during the period of minimum net demand in the
SolarPlus Scenario for the Java-Bali and Sumatra systems, in 2025
Java-Bali
30 000
MW
25 000
20 000
15 000
10 000
5 000
18 Mar 19 Mar 20 Mar 21 Mar 22 Mar 23 Mar 24 Mar

00:00 00:00 00:00 00:00 00:00 00:00 00:00
Sumatra
8 000
MW
7 000
6 000
5 000
4 000
3 000
2 000
1 000
18 Mar 19 Mar 20 Mar 21 Mar 22 Mar 23 Mar 24 Mar

00:00 00:00 00:00 00:00 00:00 00:00 00:00
Geothermal Bioenergy Coal

Gas Hydro Oil
Storage Solar Wind
Unserved Energy Curtailment Total demand
Net demand
17
In our model, run-of-river plants are assumed to have priority over solar PV. This assumption is true if there are contractual
agreements between plant operators and PLN with ToP clauses.
PAGE | 52
The Java-Bali system has 9% VRE penetration in the SolarPlus Scenario, and can
reliably integrate VRE due to both the larger size of the system and the high
complementary of solar PV production with its demand profile. As a result, there
is no curtailment of VRE, nor any negative impact on reliability.
Contribution of each generation technology to system services in the Base and

SolarPlus Scenarios, in 2019 and 2025
2019 2025 Base Scenario
Inertia
Spinning
} Stability
reserve
Ramping flexibility
Peak capacity /
adequacy
Energy
0% 20% 40% 60% 80% 100% 0% 20% 40% 60% 80% 100%
2025 Base Scenario 2025 SolarPlus Scenario
Inertia
Spinning
} Stability
reserve
Ramping flexibility
Peak capacity /
adequacy
Energy
0% 50% 100% 0% 20% 40% 60% 80% 100%
Coal Gas Hydro Geothermal Oil Bioenergy & other renewables Variable renewables Storage
Notes: The contributions are for the Java-Bali and Sumatra systems together. The contributions to stability (inertia and
spinning reserve) are calculated for the 100 hours with the lowest inertia. Inertia is based on the contribution from spinning
rotors. Inertia and spinning reserves are among contributors to stability, although detailed technical studies are required to
capture all of its components. Ramping is calculated from the contribution to the top 100 hourly ramps. Peak
capacity/adequacy is based on the contribution to capacity needs in the modelled year. Energy is the share in annual
generation. These measures aim to give an illustration of the diverse aspects of electricity security, but do not encompass all
relevant components or potential technology contributions. Demand response is not included, though it has the potential to
contribute to the services.
PAGE | 53
With the higher shares of VRE in the SolarPlus Scenario, the system will also
become reliant on coal plants in meeting peak demand and ramping requirements,
as well as in providing system services such as stability and balancing.
System stability is a potential concern as the penetration of solar PV increases.

There are many components to stability, but the analysis focuses on inertia and to
a lesser extent on the contribution to spinning reserves. In case of sudden
imbalances, the inertia of online power plants supports system stability as it limits
the rate of change of frequency by liberating kinetic energy from the rotating
masses. If synchronous generators are taken offline and displaced by VRE,
system inertia decreases because VRE is decoupled from the grid via inverters
which do not inherently provide inertial services. Practices are however evolving
and solutions exist.
In both the Java-Bali and Sumatra systems, inertia in 2025 in the SolarPlus
Scenario is lower than in the Base Scenario. Given the large size of the Java-Bali
system, inertia remains comfortable, 18 above 40 GWs, even in the lower range.
The Sumatra system, on the other hand, is more vulnerable to the decline in
system inertia due to the relatively small system size and the nature of the demand
profiles. Thermal generators also play a role to damp oscillations through their
power system stabiliser. Displacing these generators may also lead to a reduced
capability to damp oscillations. 19
One approach that has been used by many systems is to set up minimum level
inertia requirements. A detailed grid stability study is required to determine such
requirements, similar to the studies done in Ireland and Texas that have
determined maximum penetration of non-synchronous generation and minimum
inertia levels, respectively.
18
A comfortable inertia level is about one order of magnitude higher than the minimum level required to ensure that the rate
of change of frequency following typical outages (largest unit in the system) does not exceed the trigger value to disconnect
generation (to avoid equipment damage) or shed load (to halt the frequency drop and restore the frequency).
19
Inverter-based resources do not inherently provide stability services but can be designed to fulfil this role. In Great Britain,
the system operator National Grid ESO successfully procured stability services from a wide range of resources including
inverter-based generation.
PAGE | 54
System inertia based on high and low inertia estimates of power plants in the Java-Bali
and Sumatra systems in the Base and SolarPlus Scenarios, 2025
Java-Bali
200 000
MWs
150 000
100 000
50 000
Oct
Nov
Dec
Mar
May
Apr
Jul
Jan
Jun
Aug
Sep
Feb
Oct
Nov
Dec
Mar
May
Apr
Jul
Jan
Jun
Aug
Sep
Feb
Inertia (lower range) Inertia (higher range)
Sumatra
50 000
MWs
40 000
30 000
20 000
10 000
Oct
Nov
Dec
Mar
May
Apr
Jul
Jan
Jun
Aug
Sep
Feb
Oct
Nov
Dec
Mar
May
Apr
Jul
Jan
Jun
Aug
Sep
Feb
Inertia (lower range) Inertia (higher range)

Note: Inertia estimates are based on the assumed inertia of each generation technology (coal plants 4-6 MWs; CCGT and
hydropower 2-4 MWs). Detailed grid stability studies are required to provide a definite assessment of system inertia and
specific requirements of the Java-Bali and Sumatra systems.
The higher share of VRE in 2025 under the SolarPlus Scenario will lead to a
decrease in the capacity factor of coal plants and power sector emissions,
compared to the Base Scenario, due to the replacement of a small portion (about
1 GW) of uncommitted thermal capacity, and the higher utilisation of more flexible
gas-fired capacity. In the No Co-firing Scenario, the utilisation of coal plants further
increases. This results in higher CO2 emissions. In the Enforced Co-firing
Scenario, emissions are even higher as coal plants are forced to run at the
expense of gas plants.
PAGE | 55
CO2 emissions in the different scenarios
180
million tonnes
160
140
120 Sumatra
100
80
60
40
20
2019 2025 Base 2025 2025 No Co- 2025 Java-Bali

Scenario Enforced Co- firing SolarPlus
firing Scenario Scenario
Scenario
Avoided annual CO2 emissions in the scenarios compared to the Base Scenario
6%
4%
2% Sumatra
0%
-2%
-4%
-6%
-8% Java-Bali
-10%
2025 Enforced Co- 2025 No Co-firing 2025 SolarPlus
firing Scenario Scenario Scenario
Higher shares of solar PV lead to lower

operating costs
One of the results from deploying a higher share of solar PV in the generation mix
in 2025 is the reduction in the overall operational costs of the system compared to
the generation mix proposed the RUPTL. These operational costs consist of fuel
PAGE | 56
costs, ramp costs, 20 start and shutdown costs, and variable operating and
maintenance (VO&M) costs. While a carbon price, if included, would further
improve the case for renewables, this was not considered to be implemented in
2025.
Replacing the non-committed plants as planned in the RUPTL (geothermal,

hydropower, biomass, coal and gas) with solar PV in both the Sumatra and Java-
Bali systems can lead to a net reduction in yearly power system operational costs
of almost USD 430 million or 5.3% compared to the Base Scenario. The
operational cost savings could be used for implementing flexibility measures to
accommodate the higher share of solar PV, such as demand response
programmes or ancillary services contracts with generators and industrial users.
Annual power system operational costs by cost category across all scenarios
10 000 VO&M Cost

USDm
9 000
8 000
7 000 Start and Shutdown
6 000 Cost
5 000
4 000 Ramp Cost
3 000
2 000
1 000 Gas ToP Cost
2025 Base 2025 Enforced 2025 No Co- 2025 SolarPlus

Scenario Co-firing firing Scenario Scenario Fuel Cost
Scenario
Note: Gas ToP costs and coal minimum offtake costs are the sunk costs related to unused resources according to contracts.
The main cost saving component in the SolarPlus Scenario is fuel costs due to
the replacement of a small amount of uncommitted thermal generation (660 MW
coal, 330 MW gas) on an energy basis by solar PV in addition to the replacement
of the biofuel cost for co-firing. The fuel savings amount to USD 432 million (or
5.5%). On the other hand, the start and shutdown cost and ramp costs increase
slightly since conventional plants are required to cycle up/down and start/stop
more frequently to accommodate the variability of solar PV generation.
20
The ramp cost reflects the wear and tear costs as a result of increased cycling. This cost is based on international data.
PAGE | 57
Annual system operational cost savings relative to the 2025 Base Scenario
600
USDm
VO&M Cost
400
200 Gas ToP Cost
- 200 Fuel Cost

- 400
- 600 Start and Shutdown Cost
- 800
-1 000 Ramp Cost
2025 Enforced 2025 No Co- 2025 SolarPlus
Co-firing firing Scenario Scenario
Scenario Total savings
Note: Gas ToP costs and coal minimum offtake costs are the sunk costs related to unused resources according to contracts.
The results from the scenarios with enforced co-firing and without co-firing show
that the use of biofuels for co-firing has a cost premium attached to it, due to the
higher cost of biofuels in comparison to coal. Without biomass co-firing, the annual
operational costs would be around USD 66 million (or 0.9%) lower than the Base
Scenario, as biomass is more expensive than coal. There is also large uncertainty
as to what may be the actual cost of these biofuels. 21 Forcing the co-firing plants
online to meet the 23% target with bioenergy comes at the expense of gas-fired
generators that are not using the gas according to contractual volumes. This raises
the costs significantly (+10% with respect to the Base Scenario).
Sensitivity analyses on total system cost components

To address the key question of the competitiveness of solar PV, an analysis of the
total system cost was performed. Firstly, the costs were compared between the
Base and SolarPlus Scenarios, with a sensitivity analysis on some key
parameters. Secondly, conditions were identified that would allow the SolarPlus
Scenario to break even with the Base Scenario.
An initial comparison of the SolarPlus Scenario with the Base Scenario under a
set of baseline assumptions shows a net spend of USD 444 million in total system
costs when ramping up solar PV despite the lower operating costs. The main
factors contributing to the differential between the two scenarios are the fixed
operations and maintenance (FO&M) costs (USD 105 million per year, which
21
Biofuels for co-firing consist of 9% wooden pellets and 1% residual waste. The costs of these fuels are based on costs
from the Institute for Energy Economics and Financial Analysis (2021), Indonesia’s Biomass Cofiring Bet. The cost of wooden
pellets is estimated from the domestic market and from Viet Nam. The cost of residual waste is based on the average of
community-scale and industrial scale costs.
PAGE | 58
include the periodic maintenance of the assets and are independent of the output
of the plant), the additional investment (USD 795 million per year) and the savings
from avoided fuel (USD 439 million per year).
Additional total system costs of the 2025 Base and 2025 Solar Plus Scenarios
compared to 2019 (left) and net costs and savings of the 2025 Solar Plus Scenario
compared to the 2025 Base Scenario (right)
Notes: As the SolarPlus Scenario involves replacement of non-committed power plants with solar PV, the Best Case Scenario
assumes higher cost assumptions for fuel and CAPEX of coal, biomass, hydro and geothermal along with lower CAPEX
assumptions for solar PV. Worst Case entails lower cost assumptions for fuel and CAPEX of non-PV plants, along with higher
CAPEX assumptions for solar PV.
WACC=8%; VO&M costs = Variable operations and maintenance costs; FO&M = Fixed operations and maintenance costs.
Sources: Investment cost ranges from DEA (2021) Technology Data for the Indonesian Power Sector; NREL (2021) Cost
Projections for Utility-Scale Battery Storage: 2021 Update; IRENA (2020) Renewable Power Generation Costs in 2020; Fuel
cost ranges from IEEFA (2021a) Indonesia’s Biomass Cofiring Bet; IEA Bioenergy (2019) Future Prospects for Wood Pellets
Market; Argus (n.d) Indonesia International Market Prices; PLN (2021) RUPTL 2021 (baseline coal price = 70 USD/tonne
and 55 USD/tonne for 5 000 kcal/kg and 4 200 kcal/kg, respectively); IEA Coal Market Report (2021).
A sensitivity analysis of three main parameters – fuel cost, CAPEX of solar PV,
CAPEX of non-PV power plants – shows the comparative impacts based on how
Indonesian energy policy might evolve in the coming years.
Fuel costs translate to net savings for the SolarPlus Scenario. Due to avoided cost
of coal and other fuels such as biomass, the potential savings could amount to
USD 562 million if the power plants were paying international market prices,
instead of the artificially limited savings of USD 439 million due to the DMO price
cap of USD 70/tonne. 22
22
Assumptions on baseline and maximum prices for the fuels are: biomass waste = USD 25 /tonne to USD 29.3 /tonne;
biomass wood = USD 97.3/tonne to USD 115/tonne; bituminous coal = USD 70 /tonne to USD 90 /tonne; sub-bituminous
coal at USD 55 /tonne to USD 84.5/tonne. The maximum market prices for coal uses the expected equilibrium price over the
next 2-3 years instead of the current record high prices.
PAGE | 59
The CAPEX expenditure for new solar PV 23 relies on module and hardware costs
(42% of total CAPEX) and soft costs such as developer margin and financing (23%
of total CAPEX). At approximately USD 1073/kW, the cost of constructing utility-
scale solar PV in Indonesia was higher than the global average of USD 883/kW in
2020. As highlighted earlier, local content regulations are estimated to increase
the levelised cost of energy (LCOE) of solar PV by up to 50%. Allowing cheaper
module imports from efficient manufacturing countries, relaxing tax and import
duties, as well as developing financing and procurement strategies to lower the
weighted average cost of capital (WACC) and other soft costs could allow
Indonesia to lower the cost of investing in solar PV. If Indonesia manages to at
least lower the CAPEX to the global average, it could lead to investment cost
savings of up to USD 145 million. On the other hand, if supply chain issues push
up the cost of PV modules, it could result in an additional USD 47 million in
investment costs.
The CAPEX of non-PV power plants is also a factor in the feasibility of the
SolarPlus Scenario. Avoidance of higher construction costs, which are likely, entail
a comparatively lower investment cost for SolarPlus. Construction of new coal-
fired power plants could face increased financing risks due to initiatives from global
financial institutions to reduce new builds of coal and Indonesia’s own net zero
pledges. Likewise, investment in new geothermal and hydro capacities may also
be higher than historic costs, since the best locations are already developed. 24
The most favourable situation for SolarPlus is when the CAPEX for coal,
geothermal and hydro are higher while that for solar PV is lower. Meanwhile, the
least favourable situation is when the CAPEX for non-PV plants are lower while
that for solar PV is higher.
The 2025 SolarPlus Scenario would be more expensive with the baseline
assumptions. However, it would breakeven with the 2025 Base Scenario if the
following conditions were met: (i) the annualised investment costs for new coal-
fired plants were at least 43% higher (USD 2 000 /kW) than the current baseline
assumptions (USD 1 400 /kW), (ii) investment in solar PV is brought down to
global average cost, and (iii) power plants were exposed to market prices of coal,
gas and biomass.
23
Assumptions on baseline, minimum and maximum CAPEX for solar PV are: USD 1 073/kW, USD 883/kW and
USD 1 134 /kW (assuming a 10% increase in module costs and developer margins)
24
The assumptions on baseline, minimum and maximum CAPEX for non-PV plants are: coal = USD 1 400 /kW,
USD 1 400 /kW, USD 2 000 /kW; biogas = USD 2 651 /kW, USD 2 190 /kW, USD 4 356 /kW; biomass = USD 1 410 /kW,
USD 1 109 /kW, USD 2 143 /kW; geothermal = USD 4 468 /kW, USD 2 704 /kW, USD 5 785 /kW; medium-hydro =
USD 2 215 /kW, USD 1 203 /kW; USD 4 240 /kW.
PAGE | 60
This calculation illustrates the importance for the Indonesian authorities to figure
out how to improve the competitiveness of solar PV. With an LCOE of
USD 65/MWh, it is lower than the BPP-generation benchmark of USD 70.4/MWh,
but not as low as 85% as stipulated in the ministerial regulation. Having a clear
plan to deploy solar PV would support the growth of a local supply chain which,
as it develops, could offer better rates. The authorities could also collaborate with
international finance institutions to improve the investment environment for
renewable capacities.
Electrification of end-uses can help

accommodate more solar PV
Clean cooking (referring here to electric cooking) and electrification of road
transport are part of the government strategy to decarbonise the economy, reduce
imports of oil products and improve air quality. Both will increase the annual
electricity and peak demand in both Sumatra and Java-Bali, although the load
profiles of these end-uses are different.
The scenario with clean cooking assumes that one-fifth of households switch from
LPG to electric cooking by 2025. This leads to an increase in demand and peak
load by 2.2% and 4%, respectively, over the combined systems of Java-Bali and
Sumatra. The scenario with EVs assumes 370 000 electric 4-wheelers and 11.8
million electric 2-wheelers by 2025, as per the government targets, and leads to
an increase in demand and peak load by 1.2% and 1%, respectively, over the
combined systems of Java-Bali and Sumatra. Non-managed charging is assumed,
as tariff structures currently do not differentiate the time of use. Due to the existing
thermal capacity, a system with an accelerated deployment of solar PV as in the
SolarPlus Scenario could meet the demand with both clean cooking and EV
deployment by 2025, without any additional capacity or demand-side response.
In this analysis, only the impact on emissions and solar PV curtailment are
considered. Cost impacts are not studied. In particular, local grid reinforcements
may be necessary, but the model does not contain enough details about the
distribution network to perform a hosting capacity study. Such a study would be
necessary for a complete overview of the opportunity for fast electrification.
PAGE | 61
Annual energy (left) and peak demand (right) in the Java-Bali and Sumatra systems
with the electrification of new loads for clean cooking and electric vehicles in 2025
250 35
TWh
GW
2025 SolarPlus
30
200
25
2025 SolarPlus -
150 20 Clean Cooking
100 15
2025 SolarPlus -
10 EVs
50
5
2025 SolarPlus -
All New Loads
Java-Bali Sumatra Java-Bali Sumatra
While curtailment in the 2025 SolarPlus Scenario was at 14% in Sumatra (and
negligible in Java-Bali), increased electrification from clean cooking and EV
deployment reduces this curtailment by 6% and 2%, respectively, as the demand
at midday in Sumatra is also slightly increased.
Change in solar PV curtailment rate with newly electrified loads
16%
14% Java-Bali
12%
10%
8%
Sumatra
6%
4%
2%
0% Combined
2025 SolarPlus 2025 SolarPlus - 2025 SolarPlus - 2025 SolarPlus - systems
Clean Cooking EVs All New Loads
However, as the majority of the additional load is being met by dispatchable

thermal capacity in Java-Bali, and almost exclusively coal capacity, the carbon
intensity of the electricity system also increases. Meanwhile, in Sumatra, the
majority of new demand is met by otherwise spilled hydro, with some of the
demand being met by gas and a small amount of curtailed solar PV. When looking
PAGE | 62
at the power sector alone, carbon emissions would increase in the Java-Bali
system, and emissions would decrease in the Sumatra system.
Increase in generation by technology due to newly electrified loads, relative to 2025

SolarPlus
7
TWh
2025 SolarPlus
6
5
4 2025 SolarPlus -
3 Clean Cooking
2
2025 SolarPlus -
1
EVs
Coal Hydro Solar Storage

2025 SolarPlus -
All New Loads
Increase in generation by technology for the Java-Bali and Sumatra systems when
comparing a scenario with and without newly electrified loads under the 2025
SolarPlus Scenario
8 Storage
TWh
7 Wind
6
Solar
5
Bioenergy
4
3 Geothermal
2 Hydro
1
Oil
Gas
-1
Java-Bali Sumatra Coal
PAGE | 63
The increase in carbon emissions in Java-Bali highlights the need for faster
decarbonisation of the power system to support environmental and sustainability
goals when electrifying other new demands.
Reduction in carbon intensity of the power system with newly electrified loads, relative
to the 2025 SolarPlus Scenario
25
kg/MWh
20
15 Java-Bali
10
5
-5
Sumatra
- 10
2025 SolarPlus - Clean 2025 SolarPlus - EVs 2025 SolarPlus - All New
Cooking Loads
Note: In the 2025 SolarPlus Scenario, carbon intensity is 597 kg/MWh and 422 kg/MWh in the Java-Bali and Sumatra
systems, respectively.
On the other hand, taking into account the avoided emissions from the otherwise
LPG-based cooking and internal combustion engine-based road transport leads
to a brighter figure. Reduction in emissions for road transport is particularly high,
as electric mobility brings huge efficiency gains in motorcycles, at around 0.03
kWh/km compared to 0.16 kWh/km for internal combustion engines. Despite the
high carbon content of electricity in Indonesia, the emissions from 2-wheelers are
brought down from 38 kgCO2/km for internal combustion engines to under 17
kgCO2/km from electric motorcycles.
The emissions reductions from switching from LPG to electrified cooking are
limited given the lower efficiency of electric cooking compared to LPG. If electric
cooking replaces traditional biomass use instead of LPG, the reductions would be
greater and would also include reduced NOx and SOx emissions.
If incentives are established to incentivise EV charging outside peak hours,

especially during the day when PV infeed is the highest, more benefits can be
delivered, leading to greater emissions reductions and lower curtailment rates.
PAGE | 64
Absolute reduction in carbon emissions of the energy sector with newly electrified
loads, relative to the 2025 SolarPlus Scenario
4
million tonnes
3 Sumatra
2
1
Java-Bali
-1
-2
-3 Total reduction
2025 SolarPlus - Clean 2025 SolarPlus - EVs 2025 SolarPlus - All New
Cooking Loads
Contractual flexibility of thermal plants

enables a higher share of solar PV
Inflexible contractual arrangements for the thermal fleet can limit the room for
renewables in the generation mix due to distortion in the merit order dispatch,
leading to less efficient operation of the system. All scenarios until now were
constrained by three contractual inflexibilities assumed as follows: (1) a yearly
capacity factor of at least 60% for coal IPPs (PPA inflexibility or coal contractual
inflexibility), (2) a fixed daily gas offtake at the regional level (gas supply
constraints or gas contractual inflexibility), and (3) output of thermal IPPs at any
time set at 50% of the nominal capacity. The consequence of these constraints
was spilled hydro from run-of-river plants and curtailment in solar PV, mainly in
Sumatra.
Contractual flexibility would: (1) create more room for renewables by removing
minimum offtake constraints on coal IPPs, (2) allow gas units to fully use their
flexibility by operating with annual offtake constraints, and (3) allow newer IPPs to
operate at lower minimum output levels, with faster ramp rates and shorter start-
up times. With respect to the 2025 SolarPlus Scenario, relaxing these inflexibilities
can lead to cost savings and emissions reductions as otherwise curtailed low
carbon, low marginal cost generation is provided space to operate. The scenarios
in the study consider removal of the first two constraints, as well as their
combination. The third item is in any case recommended, as it does not carry
additional cost and could improve flexibility, independently from the first two items.
The cost benefit of removing these constraints is not addressed since the
renegotiation of the contracts may come at a cost which is difficult to assess.
The constraint on the coal minimum offtake requirements appears to have the
largest impact with respect to the SolarPlus Scenario. Removing this constraint
PAGE | 65
leads to savings up to 14.2 tonnes of CO2. Curtailment in Sumatra drops from 14%
to negligible amounts (<0.3%).
Relaxing gas contractual constraints, moving from a daily ToP contract for gas
offtake (per region) to an annual arrangement, without removing PPA inflexibility,
would result in a part of the gas generation being displaced by cheaper coal
generation and leading to an increase in emissions. This is because there is no
carbon pricing. The combination of removing both inflexibilities, however, allows
for both cost savings and emissions reduction benefits to be realised.
Change in annual generation by technology under different contractual arrangements

with capacity according to the 2025 SolarPlus Scenario
15 Storage
TWh
10 Wind
5 Solar
Bioenergy
-5 Geothermal
- 10 Hydro
- 15 Oil
2025 SolarPlus - Coal 2025 SolarPlus - Gas 2025 SolarPlus - Coal & Gas
Contractual Flexibility Contractual Flexibility Gas Contractual
Flexibility Coal
CO2 savings by fuel type under different contractual arrangements relative to the 2025
SolarPlus Scenario
16
million tonnes
14 Oil
12
10
8 Gas
6
4
2
Coal
-2
2025 SolarPlus - Coal 2025 SolarPlus - Gas 2025 SolarPlus - Coal
Contractual Flexibility Contractual Flexibility & Gas Contractual Total reduction
Flexibility
Note: The total emissions savings of the SolarPlus Scenario were 7 million tonnes compared to the Base Scenario, and 19
million tonnes compared to the Enforced Co-firing Scenario.
PAGE | 66
Comparison of curtailment under different contractual arrangements under the 2025

SolarPlus Scenario
16%
14%
Java-Bali
12%
10%
8%
6%
4% Sumatra
2%
0%
2025 2025 2025 2025
SolarPlus SolarPlus - SolarPlus - SolarPlus -
Coal Gas Coal and Gas Combined systems
Contractual Contractual Contractual
Flexibility Flexibility Flexibility
Given this study’s timeframe of 2025, the introduction of carbon pricing is not
analysed. However, carbon pricing (if appropriately set) would start to favour
efficient gas plants over inefficient coal plants in the merit order. A consequence
is that gas infrastructure could be supported without the current ToP mechanism
in gas supply contracts. The introduction of carbon pricing and the moving away
from ToP could thus happen simultaneously, supporting both renewables and
increased system flexibility. This would further liberate gas for sale on the
international markets.
Review and discussion of assumptions

This study illustrates a number of possible trajectories to meet the 2025
renewables target and the impact of some key sensitivities. The quantitative
outcomes depend on a number of assumptions and could be improved with actual
data. However, the qualitative conclusions appear robust, as well as the order of
magnitudes. For completeness, a few assumptions are discussed here.
The electricity demand in all scenarios is based on the RUPTL 2021. It was stated
in Chapter 2 that optimistic demand forecasts in the past are among the reasons
for the over-built thermal fleet and pose a barrier to integrating new RE generation.
If demand in 2025 is below the RUPTL projections, the existence of the contractual
and technical constraints at the thermal power plants means that the share of
renewables in the electricity mix and the potential benefits of all the alternative
scenarios are reduced.
Technical and contractual constraints on thermal generators have a significant

impact on the results. The instantaneous minimum output of coal generators was
PAGE | 67
set at 50%, which is a little higher than the technical capability of modern plants.
Changing this value has limited impact. The yearly capacity factor of coal IPPs
was set at 60%, a value at the lower end of the range experienced in neighbouring
countries. If actual PPAs were even more inflexible on the value of the minimum
take obligation (for example, 65-70%), the share of renewables in the electricity
mix and the potential benefits of all the alternative scenarios would be reduced,
except for the scenario illustrating the impact of removing the contractual
constraints.
PAGE | 68
Enhancing Indonesia’s Power System Chapter 4 - Solutions for enhanced power systems
Chapter 4. Solutions for enhanced

power systems
Recommendations for enhancing the power system

• To support its ambitious sustainability targets, Indonesia should make an
actual commitment to renewables in the power sector and put them at the
centre of planning: set even more ambitious targets, improve their
competitiveness (not only as an industry product but also as a means of
electricity generation), and adapt planning and operating practices to take into
account their characteristics.
• Given the inhibiting role of contracts on the flexibility of the young and large
Indonesian thermal fleet, there is a real urgency to enter into a contract reform
for coal-fired plants, possibly in parallel with a coal phase-out programme. To
capture efficiently the latent flexibilities from generation assets, the new
contracts should move away from energy-only pricing towards a pricing that
values better the system value of the plants with, at least, a part that capture
the role in meeting peak demand.
• Renegotiation of existing contracts may be difficult; it should be supported by
an in-depth analysis of the flexibility needs for the next decades (at least until
2040) and take into account the evolution of the thermal fleet and the benefits
of smart technologies and practices which may be introduced progressively in
the coming years.
• In the perspective of future VRE growth, operation efficiency would benefit from
deploying closer to real-time operations with system-wide forecasts of wind and
solar power, and dispatching and activation of reserves in intraday.
• The MEMR can learn from PLN’s smart grid pilots when designing Indonesia’s
own smart grids strategy which relies not only on new hardware, software and
algorithms, but also on new skills and enhanced planning and operating
practices.
• In the next decades, Indonesia will need more grid infrastructure to
accommodate the demand growth and more VRE: incentives can be deployed
not only to build more grids (and attract private financing), but also to use the
assets efficiently through enhanced planning and operation practices.
• More transparency in the planning process under the RUPTL and the allocation
of projects and contracts by PLN would bring benefits through increased
stakeholders’ confidence and better assessment of the value of new assets
and resources.
PAGE | 69
Providing access to clean, secure and affordable electricity to all Indonesians will
require the government, state-owned companies and private companies to work
together and deploy the wide range of solutions needed to improve – on the one
hand – the efficiency of the large systems and – on the other – the quality of
electricity in smaller systems or remote areas. Enhanced institutions and practices
– at various levels – will play a critical role in achieving both objectives.
The previous chapter showed that the systems of Java-Bali and Sumatra are able
to accommodate enough solar PV to fill the gap between the committed renewable
generation and what is needed to meet the 2025 RE targets. In 2025, the annual
share of solar PV would then be approximately 10% (compared to 2% in the
RUPTL 2021). More ambitious levels of VRE would require the development of
more extensive and smarter grids. While investment in assets is an essential part
of Indonesia's transition, a good share of progress can be made without significant
investments in infrastructure by making better use of the existing assets. In this
chapter, we present a range of solutions aimed at helping Indonesia deploy more
VRE and meet the country’s sustainability goals. These solutions span across the
“hard” (equipment and infrastructure) and the “soft” (commercial structures and
improved practices in planning and operations) aspects of the enhancements to
the Indonesian power sector.
New contractual structures can harness the

flexibility potential of Indonesia’s thermal
fleet
A power system may have the technical capabilities to provide adequate flexibility
but, due to commercial structures this flexibility cannot be utilised. This is the case
in Indonesia. Aligning commercial flexibility and technical flexibility would enable
Indonesia to transition to a clean energy system and integrate a higher share of
low-carbon resource.
The PPAs which require PLN to take power from the IPPs well above the minimum
stable levels of the unit and to guarantee a given capacity factor are a significant
cause of inflexibility. In theory, additional contracts for flexibility services could
be added on top of the physical PPAs. However, this would imply an additional
payment which may not be the most cost-effective way to integrate VRE.
The operating characteristics of power plants (for example: start-up time, stable
minimum operating level, ramp rates) can be a barrier especially for older thermal
assets. These assets can be retrofitted to ensure, for example, lower stable
minimum levels and higher ramp rates, which in turn would make these assets
more valuable in a clean energy transition to allow integration of variable
PAGE | 70
renewables. It is noted that the coal fleet of Indonesia is relatively young compared
to the average age of coal-fired power plants globally.
In Thailand, it was found that retrofitting power plants to enable lower minimum
stable levels and faster ramping, while valuable, was less cost-effective compared
to the relaxation of contractual constraints. The costs and benefits of all options
must therefore be considered.
The long-term contracts for gas supply, which include ToP clauses, are another
cause of inflexibility. Due to these clauses, the gas supply is a sunk cost, altering
the merit order of the units in the dispatch.
Strategies to introduce new contractual structures

In order to identify pathways to increase contractual flexibility, it is important to
understand what level of flexibility is needed and what the barriers are for the
Indonesian system, specifically. Detailed information regarding the payment
structures and/or minimum take obligations in the PPAs with the coal generators
could support a more detailed assessment of the contractual barriers. This could
be the scope of a later study.
When contractual structures limit flexibility, two main strategies exist:
• let contracts run out and make changes only in new contracts, including renewals
of old contracts
• seek to renegotiate current contracts as well as make changes in new contracts
including renewals of old contracts.
Irrespective of which of the two strategies is deployed, it is important to start
preparing new contractual structures, such that any new contracts between PLN
and generation assets are more flexible and better suited for higher shares of
VRE.
Key factors that need to be considered in order to understand which of the two
main strategies is most applicable include the duration and extent of the impact of
inflexible contracts. In a situation where the contractual structures are about to
naturally run out, it is preferable not to renegotiate contracts.
Flexible contractual features

Best practices for new contractual structures can be explored in future work.
These would build on the principles provided here. For thermal assets, the
following principles can be applied to contracts to increase flexibility:
• separating physical production guarantees from budget stability

• lowering minimum take obligations
PAGE | 71
• lowering minimum run rates to match the technical capabilities

• implementing differentiated price incentivising flexibility
• implementing budget security instruments, such as floors on settlement
irrespective of generation
• providing financial incentives for retrofits of older plants to increase flexibility
• separating contracts for system services from the supply of energy.
For renewable energy, the following contractual features support flexibility:
• clear procedures for curtailment

• clear settlement rules for curtailment (for example, with compensation)
• mechanisms that incentivise system friendly deployment, 25 including system
services
• forecasting requirements.
Approach to renegotiating existing contracts

In cases like Indonesia, where contractual structures will continue to significantly
limit the system flexibility for many years, renegotiation should be considered.
Renegotiation needs to be done with extreme care in order not to negatively
impact the investment climate since it can increase the required return of
investment on future assets, and thereby increase the cost of the clean energy
transition. This should be co-ordinated with efforts towards coal phase-out.
When renegotiating, it is important for governments to understand that the main

goal is to optimise the overall cost of the system in the long run rather than trying
to save as much money as possible on each individual contract. This is to avoid
long-term consequences on investor trust, which can make the transitions less
affordable. Transparent renegotiation will build investor confidence even though
existing contractual structures may be changed or to some extent bought out.
Stakeholder consultation is a key component of contract negotiation, one in which

capacity building among asset owners, system operators, financial institutions and
policy makers can play a key role. It is important to consult both asset owners and
financial institutions in order to understand the most important factors for them in
the process. For example, asset owners may be willing to accept lower guaranteed
take obligations if they are compensated with differentiated prices, but this may
not be acceptable for the financial institutions.
25
System-friendly deployment: promoters are incentivised to invest first in resources that have the highest value for the
system, for example, those which minimise the need for additional grid assets or which produce at times when the system
needs it the most.
PAGE | 72
It is important that all stakeholders understand the system needs, and also which
options can contribute to flexibility from contractual, technical, operational/process
and policy perspectives. For example, financial institutions may need capacity
building in order to be able to evaluate how reduced minimum take obligations will
affect the risk profile of the investment.
One approach that can be helpful when renegotiating contracts is auctions. If the
needed flexibility is well defined, then auctions can be held with the new
contractual structures, and assets can bid in the needed compensation to change
contractual structures to increase system flexibility. In the auction, the needed
amount of flexibility can be defined such that not all contracts have to be
renegotiated. Additionally, the competition element of the auction will ensure that
the renegotiation will be done most efficiently, meaning that the contracts that
require the lowest compensation are the ones that will be changed. There can be
differences across assets’ willingness to restructure due to different factors like
investor risk appetite, ratio of debt versus equity, technical capability, etc. The
auction design determines the success of the auction. If the design and process
are not carefully considered and aligned to the desired outcome, auctions can
prove unsuccessful.
Governments have different financial abilities to renegotiate contracts. In the case

where financial resources are limited, it is even more important to prioritise which
contracts to renegotiate in order to maximise system flexibility. In order to do this,
a detailed understanding of the need for flexibility, as well as a thorough
understanding of the elements of the contracts that limit the utilisation of existing
flexibility are required. Additionally, governments should seek to collaborate with
philanthropy, development banks and other governments in order to increase the
financial ability to restructure contracts where needed. An example of this is ADB’s
Energy Transition Mechanism which includes efforts for early retirement of coal
plants in Southeast Asia.
In-depth assessment of needs can support

Indonesia’s own smart grids strategy
The study in Chapter 3 explored the flexibility requirements in 2025. Given the
need for long-term contracts for investors in new renewable capacities, a study of
the flexibility requirements over a longer period, at least until 2040, would support
the efficient renegotiation of contracts. Beyond the current decade, a number of
other enhancements would add to the current flexibility resources. These include
smart technologies and enhanced practices for planning and operations.
PAGE | 73
Digitalisation opens up a wealth of opportunities in the

power system
Digitalisation contributes to unlocking technical flexibility from a wide variety of
resources. Digitalisation is the application of powerful information and
communications technologies in equipment, analytics and user interfaces.
Examples of digital enabling technologies in today’s energy systems are smart
meters, digital sensors in generation and on grids, drones (robotics) and “digital
twins” (a digital simulation of a real-life asset to enable the virtual testing of
features, feasibility and durability). Digitalisation is at the heart of the power system
transformation since it offers opportunities in all the segments of the value chain,
from the generation of electricity all the way down to the final consumer, who can
be empowered to play a bigger role. In a “smart grid”, information is exchanged
between all interested stakeholders, enabling them to make optimal decisions.
Smart grid applications and benefits in a nutshell

Utilities have been at the forefront of digitalisation through data and automation,
and corrective control means. Potential benefits span a wide range of applications.
Generation: Digital upgrades to power plants can improve efficiency and
resilience: a key application is assets performance monitoring. Predictive
maintenance uses advanced analytics to closely monitor and analyse equipment
so that potential problems can be identified at an early stage and repairs can be
carried out before failures happen. This significantly reduces unplanned outages
and downtime. The IEA estimates that over the period to 2040, digital equipment
could deliver an average 5% reduction in power operations and maintenance
(O&M) costs, and an average 5% improvement in performance.
These gains in operation optimisation would also bring savings in new
investments: if the lifetime of all the power assets in the world were to be extended
by five years, (considering all assets, including grids and generation) close to USD
1.3 trillion of cumulative investment could be deferred over 2016-40 globally.
In the case of variable renewable plants generating power from solar and wind, a
layer is added to forecast generation from weather data and to optimise the
contribution from VRE plants to the system. Often connected to the grid through
power electronic interfaces, these plants are flexible and can contribute to grid
management, for example supporting local voltage and modulating down output
power to avoid congestions. Since 2003, Spain’s Iberdrola’s CORE (Renewable
Energy Operation Centre) has been displaying the benefits of technology to
monitor renewable generation remotely.
PAGE | 74
Grids: Power system flexibility is a cornerstone of electricity security in modern

power systems. The grid plays a central role in unlocking flexibility from power
plants, energy storage and demand-side resources. Digitalising electricity
networks is essential to pool all available flexibility sources.
Digital substations are making operations more efficient and the system more
resilient. New generation supervisory control and data acquisition (SCADA) and
EMS are incorporating advanced functions facilitating grid management. Finally,
more flexible planning practices and close-to-real time operation take advantage
of short-term weather forecasts to minimise RE curtailment and maximise the
generation from RE sources. Red Electrica de Espana’s CECRE (Control Centre
of Renewable Energies) monitors all renewable units above 5 MW in Spain and
supports the seamless penetration of renewables. Dynamic line rating (DLR), for
example, helps make better use of existing transmission corridors thanks to more
accurate assessment of their real-time capacity.
Distributed generation and consumers: Smart meters and digital investments
provide better visibility of the distribution grid and consumer uses. This reduces
non-technical losses and interruption times.
Digitalisation also allows harnessing the flexibility potential from final consumers
located at the low and medium voltage levels. While only a minor part (15%) is
used today, the IEA estimates the demand response potential in 2040 at 6400
TWh – which is 20% of the total energy consumed.
Real-time communication through smart meters can lower peak demand by
affecting buildings consumption and shifting EV charging times. Auto-consumption
can also contribute to lower RE curtailment where the DG hosting capacity of local
grids is a constraint.
Reinforcing the institutional framework around smart

grids
To support Indonesia’s sustainability goals, the MEMR mandated PLN to develop
a smart grid programme through Presidential Decree 18/2020 with the target to
develop five “smart” distribution networks per year in Java-Bali from 2020 to 2024.
PLN’s smart grid roadmap runs in two phases. In 2021-2025, the focus is on
reliability, efficiency, customer experience and grid productivity. From 2026 and
beyond, the focus is on resilience, customer engagement, sustainability and self-
healing. This programme includes a dozen pilot projects and a roadmap for
deployment of advanced metering infrastructure over the 2021-2025 period and
beyond. PLN has already achieved some success in deploying digital technology
in two substations and advanced analytics in dispatching centres.
PAGE | 75
The Indonesian institutions and MEMR can take a stronger role in smart grids.
Although utilities have a major role in building the digital infrastructure and leading
the associated developments, the authorities are in charge of deploying policies
and the right incentives to ensure that the benefits of smart grids are shared along
the value chain. To fill this role, the MEMR should lead – in collaboration with
stakeholders – the development of a vision and roadmap, initiate programmes,
perform monitoring and adopt the needed regulations and standards.
India provides an example. The Indian Ministry of Power has established a

national smart grid mission (NSGM) with a leading role across the country
(Government of India, 2015). NSGM's implementation framework has four pillars:
(1) Develop a vision and an institutional structure: define the governance, set the
goals and control the operational budget; (2) Establish standards and the policy
framework: the necessary standards required for information technology and
operational technology for smart grids; (3) Develop business models and catalyse
investments; and (4) Monitor progress in achieving the goals and report to the
Ministry. India's know-how in smart grids has enabled it to become recognised
today as one of the leading parties in the international smart grid network ISGAN,
and the country hosts a Smart Grid Forum.
Smart grids are also an opportunity for PLN to expand their value offering by
providing services to consumers to reduce their consumption (and thus bills)
and benefit the system while aiming to achieve the sustainability goals. These
services could help consumers buy energy-efficient appliances (supported by
long-term contracts with manufacturers) or implement automation services, such
as energy management systems or smart appliances. These services can be
provided by separate businesses called energy service companies (ESCOs).
Energy performance contracts (EPCs) between the customer and the supplier
may include the replacement or deployment of equipment, and the supplier is often
rewarded through capturing a part of the customers’ energy savings.
These services may seem difficult to combine with the traditional business model
of the utility: the sale of electricity as a commodity may decrease, thereby
decreasing the corresponding revenue streams. On the other hand, this new value
offering is in line with the growing expectations of consumers for an energy supply
that is of higher quality and more environmentally friendly while remaining
affordable. Furthermore, the supply of electricity to consumers is currently
subsidised. Appropriate regulatory incentives may help the state-owned utility to
see benefits in these services and to deliver them. Given PLN’s footprint in the
electricity sector, incentives can be of macroeconomic nature (related to targets
of power consumed per unit of economic output) or related to targets of improved
power efficiency in selected sectors and industries.
PAGE | 76
A long-term smart grid strategy building on processes

and people to ensure the successful use of deployed
hardware
There are many benefits to digitalisation in power systems. Indonesia should,
however, identify the applications that deliver the best value in its own context,
based on the power system framework and its priorities for the next decades. For
example, as the current pricing scheme does not incentivise consumer flexibility,
investment in smart meters may deliver little benefits.
While driven by technology, the deployment of smart grids should be seen as a

holistic effort of the power sector towards digitalisation that builds on solid
institutions and enhanced processes. Smart grids are not only about assets,
methodologies and algorithms. These need to be supported by staff with the right
skills and operational procedures that make use of the benefits of the technology.
Operational procedures may need to be updated or developed to ensure proper
use of the new and existing technology. Staff also need to be trained in new skills,
such as data management and data interpretation. For example, as data flow in
and insights are provided on the health of assets on a continuous basis (instead
of during periodic revisions), procedures should be in place to initiate an
intervention on a deteriorating asset in a reasonable timeframe and to quickly
adopt the required operational changes.
Indicators are needed to measure the successful use of deployed hardware.

While PLN has targets for the deployment of digital infrastructure and equipment,
it is equally important to ensure that the deployed equipment is used efficiently.
Sharing information with interested stakeholders will

benefit the system, but cybersecurity grows in
importance
Digitalisation will deliver new and more abundant information, thanks to remote
monitoring and analytics, that utilities could use to improve the quality of their
service and efficiency. Access to transparent and up-to-date information about the
system also allows stakeholders to make optimal economic decisions, such as
investing in the needed resources that help the system. The information shared
can span a wide range: raising awareness of the current projects and achieved
benefits (such as the public smart metering implementation dashboard from India),
providing information on support schemes and the way risks are managed,
explaining how data security and privacy are ensured, listing channels for
implementation support, and providing real-life data to support business cases.
On the other hand, digitalisation requires utilities to deploy new skills in

cybersecurity and a legal framework to ensure data privacy.
PAGE | 77
Enhanced operating practices can help

realise the full benefits of technical and
contractual flexibility
The removal of barriers from contractual structures and the deployment of smart
grids will have clear benefits for system security and flexibility. Substantial
enhancements could be achieved without major investment, through improved
operational practices. Increasing the share of VRE would require not only
improved operational planning, but also modern operational practices, which
would also lead to revision of the existing grid code. Forecasting and flexible
resource requirements would change the traditional schedule of operations
especially for countries that rely heavily on baseload generation, as does
Indonesia.
Indonesia has already made some steps to modernise its system operations. It
previously had different grid codes for the different power systems (Regulation
03/2007 applied to Java-Madura-Bali and Regulation 37/2008 applied to
Sumatra). They were consolidated in 2020 under MEMR Regulation 20/2020, 26
along with laying out provisions for variable renewable energy. The Connection
Code is a key instrument to harness flexibility from new power plants of all sizes
including VRE plants. To accommodate the increasing VRE penetration, the
connection codes could be further improved with specific, forward-looking
requirements in order to maintain the security of the system. Key technical aspects
to enhance system flexibility include ramping, voltage control, frequency
response, power quality, frequency regulations and fault ride through. Reasonable
requirements in the Connection Code (differentiating size and connection voltage)
can provide large flexibility at low cost and which can be harnessed later.
26
Regulation of the Minister of Energy and Mineral Resources number 20 of year 2020, on network rules for electric power
systems (Grid Code).
PAGE | 78
Schedule of power system operations
Annual Monthly Weekly Daily Real-time
- updated every 6 - revision of - estimates of - determination of - balancing

months with a previous unserved energy total active and - re-dispatch for
planning horizon forecasts and reserve reactive power differences
of 2 years - half-hour margins generation beyond 5%
- allocation from interval modelling - designation of - determination of
minimum off- of load and reserves load reduction
take in contracts optimum whenever
- monthly economic necessary
forecast data dispatch
from VRE and
hydro
- maintenance
planning
- power flow and
transmission
constraints
Source: IEA analysis of MEMR (2020), Indonesian Grid Code.

Short-term forecasts and intraday scheduling to enable

better use of the existing resources
PLN dispatches generating resources according to their marginal cost and
schedules submitted by 10:00 for the day ahead. VRE operators also submit real-
time data and a daily forecast with a resolution of 15 minutes, updated every 6
hours.
Moving scheduling and dispatch closer to real time would allow for a more
accurate representation of variations of net demand (demand minus non-
dispatchable generation). In addition to the existing day-ahead process, an
intraday unit commitment would result in a more efficient use of reserves because
a larger portion of the VRE variability is absorbed by the updated schedules and
does not need to be balanced by reserves. A new operational function focused on
intraday planning and scheduling can be considered to assist with forecasts and
manage intraday generation schedules. Centralised system-level forecasting of
VRE generation can improve system operation by enabling the system operator
to account for overall variability of VRE outputs across the whole system and
accurately predict the amount of VRE generation available. This system-wide
forecast would complement the plant-level forecast and allow design incentives to
improve the quality of the forecasts submitted by VRE operators.
PAGE | 79
Improving reserves procurement to incentivise flexibility

Currently, reserve requirements are deterministic, based on the size of credible
outages. PLN enters into mid- to long-term contracts with conventional generation
and interruptible load to ensure their availability.
An increased granularity of reserve products and close-to-real-time procurement
can improve cost efficiency. Timing refers to both the balancing reserves interval
and the forward time of procurement (time between gate closure and delivery). A
shorter interval decreases the volume of reserves required due to schedule
changes, but also complicates procurement. For balancing markets to tap into all
of the relevant sources of flexibility, the rules set for participation need to carefully
address the access requirements including minimum resource size, technical
capabilities and resource type. Appropriate requirements – for example
asymmetrical products (separate products for upwards and downwards reserves
with different requirements) – can open participation by VRE but also demand
response (including EV charging) and distributed energy resources (at least,
through aggregators). Appropriate steps are needed to monitor these new
products as the quality of the reserves may decrease (typically, if a service is
provided by units not equipped with the detailed telemetry and metering
capabilities to meet the current requirements). The pricing structure would also
need to be adapted.
Monitoring and forecasting dispersed resources to bring

light to variability
Sufficient observability of resources meant to grow in the future will deliver benefits
in the future. This applies to VRE (both dispersed and utility-scale) and to
electrification of new end-uses such as EVs, air conditioning and electric stoves.
There is no immediate need for a massive investment in digital infrastructure to
enable VRE penetration. It is possible to derive substantial value through
deploying measurement devices at selected units and extrapolation of available
data. Similarly, VRE forecasting tools use sensing technologies, together with
mathematical models, to accurately predict wind speed and solar irradiance, and
subsequently forecast outputs from VRE plants on a sub-hourly basis.
New reliability standards to support VRE

Following the 2019 large-scale blackout in Java-Bali, which left millions of citizens
without electricity for hours and up to a day in some areas, the Indonesian
reliability framework has been improved. The grid code and the distribution code
are now the key references for system operations. In transmission, the N-1
PAGE | 80
criterion is the standard. A spinning reserve must be kept matching the size of the
largest online generator. A defence plan is in place with manual and automatic
shedding to prevent frequency instability. Interruptible contracts represent 20-25%
of total consumption. At least 30% of the load can be shed automatically.
On the other hand, reliability and reserves requirements are static and do not take
VRE into account. PLN is required to maintain a 35% reserve margin. This criterion
contributed to the overcapacity in thermal generation and additional challenges for
VRE. As the transition proceeds, the static reserve margin on generation should
leave room to new metrics, eventually taking into account probabilistic aspects
when VRE sources become significant. The most common metrics are the
expected energy not served (EENS), expressed in MWh per year, the loss of load
expectation (LOLE) and loss of load probability (LOLP) for a specified ENS
volume, both expressed in hours per year. No single reliability metric and standard
can capture all types of events, from situations where customer shedding is to be
performed preventively in order to shave peak load during a rare instance of very
high demand, to large-scale outages affecting customers for several hours to
days.
The proposed framework to establish or revise reliability metrics is as follows: The

competent authority (in the case of Indonesia, the MEMR) needs to decide what
are the events that have to be prevented; as a corollary, what events/damages
are acceptable for the society and how often. This results from consultations with
field technical experts. From this decision, it is then possible to set the policy
objectives and update reliability standards and metrics accordingly. There is a
balance to be found between the security standards and their cost. Therefore, the
set standards are to be monitored continuously and reviewed periodically.
Enhanced planning practices to support the power

system transformation
Adapting reliability standards to take into account the nature of VRE will affect
operations and planning. But VRE is not the only major change. End-use
electrification such as EVs and electric cooking increase the demand for power
but the nature of these new uses also offers opportunities for flexibility. Therefore,
system planning practices could be improved so that a long-term flexibility strategy
is developed and implemented. Various instruments can be combined to unlock
the needed flexibility such as grid codes (in particular, grid connection
requirements), financial incentives and enhanced operational practices.
Co-ordinated and integrated planning, such as the Australian Energy Market

Operator’s (AEMO) bi-yearly Integrated System Plan (ISP), which serves as
Australia’s whole-of-system plan for the next 20 years, provides collaborative
frameworks that bring together stakeholders to design collectively the energy
PAGE | 81
systems of the future. Even though, AEMO’s ISP is set in a market-based

environment, many good practices can be derived for Indonesia’s power sector.
Incentives to deploy the needed grid infrastructure

Grid development delays and curtailment are a major risk for investors in
renewable resources. A higher level of investment in grids is needed to enable
better access to the available resources and to support Indonesia’s sustainability
objectives. In the meantime, enhanced network management practices such as
dynamic line rating (DLR) 27 can be implemented to help maximise the use of the
existing grid, defer investments in new assets and reduce congestions when
weather conditions are favourable. From 2028 onwards, interconnections with
other systems may increase benefits, through sharing reserves and further
optimising the use of resources. This was already evidenced by the IEA’s study
supporting multilateral power trading in ASEAN and is further elaborated in IEA’s
upcoming Energy Sector Roadmap to Net Zero Emissions in Indonesia. Multi-
value approaches help accelerate grid expansion as they account for all benefits
of grid projects: increased reliability; sharing existing resources and reducing
reserves; and decarbonisation through better use of low-carbon resources.
The creation of renewable energy development zones (REDZ) can also be useful
in accelerating the growth of renewables. REDZ are locations with abundant
renewable potential which are remote from the existing grid. To harness the
potential from these areas in the most cost-effective way, the grid is designed and
deployed pro-actively to accommodate many renewable plants.
To accelerate investment in grids, the Indonesian institutions can create an

environment to attract private sector capital. Many business models exist for
privately-financed transmission as elaborated by the IEA and UN ESCAP. As there
is a legal requirement that the Indonesian grid be operated by PLN, two models
stand out: the financial ownership model and the BLT model. Regardless of the
chosen model, policy and regulatory changes may be needed.
Financial ownership means that a private company provides (part of) the financing
for a new transmission line, but that the line is built and operated by the public
utility. The investment prospect is provided by the public utility. The private
investor receives a return and pays operation costs according to its ownership
share. As the revenue is typically linked to the line usage, the main risk is related
to the trust in the utility to build and operate the line as planned. This model was
27
Dynamic line rating (DLR) refers to the active varying of presumed thermal capacity for overhead power lines by taking
into account actual operating and ambient conditions instead of assuming a fixed capacity. In particular, colder and windier
days allow a higher physical flow than the fixed or seasonal rating.
PAGE | 82
applied to one cross-border line between Germany and Denmark, where Vattenfall
owns a third of the line.
The BLT is an alternative to build, own, operate and transfer (BOOT), which is one
of the most used models worldwide. In the BLT model, a private company finances
and builds a new transmission line on behalf of the public utility and then leases
the project back to the utility for a predetermined period (the concession period –
typically 25 years or longer). In this model, the private company assumes the risk
to build and commission the line by the contractual deadline.
Consideration of all cost-effective options over the long

term, including alternatives to new supply side assets
While new assets are important, policy and regulations must encourage and
reward risk-taking towards adoption of innovative, non-capital related solutions,
especially when they benefit the end user. Even in growing economies like
Indonesia, a significant share of the increasing demand can be met without
deploying new power plants. Measures incentivising energy efficiency and smart
consumption, and shifting demand away from peak times, will become necessary
as new electric load grows. The various options should be compared through a
cost benefit analysis that takes into account costs and a variety of benefit
categories which include reliability and environmental impacts. The ranking of the
projects can then be made based on criteria aligned with policy objectives.
Stakeholder engagement and transparency in planning

to attract investment in grid-friendly resources
Significant improvements to the processes under PNL's RUPTL are to mandate
stakeholder consultations and to make more transparent the allocation of projects
approved under the RUPTL.
Planning is essential to assess the system needs and provide signals for
investment in the resources. However, planning the system for the next decades
is a complex task, as there are many uncertainties such as technologies and their
costs, the global environment and the behaviour of consumers. Therefore,
planning should ensure robustness with respect to these uncertainties. This can
be achieved through the use of stochastic approaches and multiple scenarios
complemented with sensitivities with respect to major assumptions. Scenarios
should span a large set of possible futures, such as the penetration of distributed
resources and technology choices. Major assumptions may include the speed of
phasing in/out of technologies (phase-out of coal generation or phase in of new
low-carbon technologies) and the cost of capital. As climate adaptation is a
growing concern, planning should also consider stress tests with respect to
extreme weather events.
PAGE | 83
Transparency and stakeholders’ consultation across the planning cycle bring

many benefits. On the one hand, stakeholders can provide inputs to reduce the
range of uncertainties and select appropriate scenarios. On the other hand, the
sharing of plans provides stakeholders (investors and industrial consumers) with
transparent information about the locational value of new resources, supporting
deployment at the most efficient locations.
PAGE | 84
Enhancing Indonesia’s Power System Annex
Annex
Abbreviations and acronyms
AEMO Australian Energy Market Operator
ASEAN Association of Southeast Asian Nations
BLT build-lease-transfer
BOOT Build-own-operate-transfer
BPP average cost of production (biaya pokok penyediaan)
CAPEX capital expenditure
CF capacity factor
CCGT combined-cycle gas turbines;
CCUS carbon capture, usage and storage
DGE Directorate General of Electricity (Direktorat Jenderal Ketenagalistrikan,
DJK)
DMO Domestic Market Obligation
ETS emissions trading scheme
EV electric vehicle
ICE internal combustion engine
IEA International Energy Agency
IPP independent power producer
JVB Java-Bali (system)
KEN national energy strategy (Kebijakan Energi Nasional)
LCOE levelised cost of energy
LOLP loss of load probability
MEMR Ministry of Energy and Mineral Resources (Kementerian Energi dan
Sumber Daya Mineral, ESDM)
NDC Nationally Determined Contribution
NRE new and renewable energy (Energi Baru dan Terbarukan. EBT)
OCGT open-cycle gas turbine.
O&M operations and maintenance
OPEX operational expenditure
PLN Perusahaan Listrik Negara (State Electricity Company)
PPA power purchase agreement
PV photovoltaic
RUEN National Energy Master Plan (Rencana Umum Energi Nasional)
RUKN National Electricity General Plan (Rencana Umum Ketenagalistrikan
Nasional)
RUPTL Electricity Supply Business Plan (Rencana Usaha Penyediaan Tenaga
Listrik)
SAIDI system average interruption duration index
SAIFI system average interruption frequency index
SUM Sumatra (system)
ToP take-or-pay
PAGE | 85
Enhancing Indonesia’s Power System Annex
VRE variable renewable energy

WACC weighted average cost of capital
Units of measure
bcm billion cubic metres
Btu British thermal unit
EJ exajoules (1 EJ = 23.88 million tonnes oil equivalent)
GW gigawatt
GWh gigawatt hour
km kilometre
kV kilovolt
kW kilowatt
mBtu million British thermal units
mtpa million tonnes per annum
MW megawatt
MWh megawatt hour
TWh terawatt hour
PAGE | 86
Enhancing Indonesia’s Power System: Pathways to Meet the Renewables Targets in 2025 and Beyond
This publication was produced with the financial assistance of the European Union
as part of the Clean Energy Transitions in Emerging Economies programme which
received funding from the European Union’s Horizon 2020 research and
innovation programme under grant agreement No. 952363.
This publication reflects the views of the IEA Secretariat but does not necessarily
reflect those of individual IEA member countries or the European Union (EU).
Neither the IEA nor the EU make any representation or warranty, express or
implied, in respect to the publication’s contents (including its completeness or
accuracy) and shall not be responsible for any use of, or reliance on, the
publication.
Comments and questions on this report are welcome and can be addressed to
jacques.warichet@iea.org.
This publication and any map included herein are without prejudice to the status
of or sovereignty over any territory, to the delimitation of international frontiers
and boundaries and to the name of any territory, city or area.

IEA Publications
International Energy Agency
Website: www.iea.org
Contact information: www.iea.org/about/contact
Typeset in France by IEA – August 2022

Cover design: IEA
Photo credits: © Getty Images
PAGE | 87

Enhancing Indonesia S Power System

Uploaded by

Copyright:

Available Formats

Enhancing Indonesia S Power System

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Enhancing Indonesia S Power System

Uploaded by

Copyright:

Available Formats

Enhancing Indonesia’s

Source: IEA. All rights

IEA. All rights reserved.

Elspeth Thomson edited this report. Colleagues in the Communications and

A key barrier to accommodating variable renewables in the Indonesian power

The overall conclusion is that, from a system integration perspective, Indonesia

IEA. All rights reserved.

In 2021, Indonesia announced a commitment to achieve net zero emissions by

populated island with a coal-dominated electricity system while Sumatra is a more

Power system flexibility to support clean

In particular, solar PV produces electricity according to diffuse solar radiation,

The building blocks of power system flexibility

IEA. All rights reserved.

Report scope and structure

residential cooking – separate but parallel goals of the Indonesian government –

IEA. All rights reserved.

Chapter 2. Indonesia’s power

Key findings – Power sector in Indonesia

Indonesia’s power sector is a major source of

Source: IEA (2021), World Energy Statistics and Balances.

… but Indonesia has ambitious targets for

Selected targets under the 2014 National Energy Policy

Unit 2020 2025 2050

In its updated Nationally Determined Contribution (NDC), the country aims to

Indonesia’s abundant resources for power generation enabled it to increase

Solar and wind

Source: IEA (2021), World Energy Statistics and Balances.

The transmission system is mostly made up of 150kV transmission lines, with a

Apart from Java-Madura-Bali, the islands currently have separate systems.

Transmission lines in the Java-Bali and Sumatra system

The power sector is structured around state-

The 2009 Electricity Law designates PLN, a state-owned vertically-integrated

Several independent power producers (IPPs) are allowed to participate by selling

In contrast, transmission and distribution infrastructure remains within the

Overly optimistic demand forecasts and

550 140 RUKN 2008

Historical reserve margins in the Java-Bali power system

60% Western Java

SAIDI and SAIFI indicators in Indonesia, 2011-2020

The operating practices in Indonesia are deterministic, in line with traditional

Indonesia has substantial energy resources

PLN relies on biomass co-firing to achieve

Growth in renewables generation in the RUPTL optimal and low-carbon scenarios,

Sources: IEA analysis of PLN (2021), RUPTL 2021.

To achieve its renewable energy targets, Indonesia is aiming to leverage biomass

Source: IEA Analysis of PLN (2021), RUPTL 2021.

Power pricing regulation limits attractiveness

The total BPP is made up of three components: BPP-pembangkit (generation),

In order to ensure electricity affordability, the MEMR controls the generation

As renewables purchase prices are also regulated, it introduces stiff competition

Historical electricity subsidies, average subsidy rate and BPP, 2010-2020

There is untapped flexibility potential in the

Prominent technical flexibility resources include power plants (both conventional

Contractual/institutional flexibility is the flexibility provided by underlying

In addition to technical and contractual flexibility, modern system operational

Understanding flexibility requirements, from short-term to long-term, can support

Source: IEA analysis of PLN Data.

Typical ramping requirements in Java-Bali and Sumatra systems