Estimation of Inertia Constant of Iran Power Grid
Estimation of Inertia Constant of Iran Power Grid
Estimation of Inertia Constant of Iran Power Grid
Abstract— Inertia constant is an effective parameter for power normally installed in some power system substations. In this
system analysis such as the load frequency control, dynamic behavior paper, we use the PMU data at a very effective event in Iran
study and it is necessary for system operators to do decision making network for estimation of Iran Grid inertia constant.
during contingency conditions. This paper uses the largest simulation Comparing the results of measurement and simulation based
model of Iran power grid (containing integrated transmission and
sub-transmission network model) and Phasor Measurement Unit
method show the validity of the calculations.
(PMU) data to analysis the generator outage of Bushehr power plant II. INERTIA OF POWER SYSTEM
as a very effective event to estimate the inertia constant of Iran power
system. The approximate matching between simulation and Immediately following a disturbance, the missing/excess
measurement based calculation results supports the validity of power is delivered from the kinetic energy stored in the
resulted value. rotating mass of the turbines. This leads to a
deceleration/acceleration and thus to a decrease/increase in the
Keywords- Inertia constant; Phasor Measurement Unit (PMU); system frequency. The contribution of each generator towards
The Largest Simulation model of Iran Grid.
the total additional power required is proportional to its
inertia. Individual contributions to the balance are proportional
I. INTRODUCTION to the inertia/acceleration time constant of each generator.
Some parameters are very important for power system This relation can be mathematically described as follows:
evaluation and control such as inertia constant which is
especially used for load frequency control and dynamic ΔPi = Pi − Pi disp = K i .Δf (1)
behavior analysis of power network. On the Other hand, the where,
inertia is a very effective parameter for system operators to do
is the modified active power of generator ݅, is the
decision making during contingency conditions. Therefore,
some papers such as [1]–[4] have presented discussions and initial active power dispatch of generator ݅ and is the
methods for calculation and using this parameter in various active power change in generator ݅. is the total frequency
issues. deviation and is the inertia gain parameter of generator i,
In this paper, we use the largest simulation model of Iran which can be calculated as:
containing transmission and sub-transmission network (with K i = J i .ωn .2π (2)
about 60000 busbar/terminal, detailed model of 500 Where is the rated angular velocity of generator i and
generators, 4400 lines and 4000 transformers, etc.) which is as the moment of inertia of the generator can be calculated as:
developed and updated by workgroups of regional electric Ta 2H i .Pni
companies (RECs) with cooperation of Iran Grid Management J i = Pni . i2 = (3)
ωn ωn 2
Company(IGMC)[5]. Simulations are done using DIgSILENT
PowerFactory[6]. Where is inertia time constant of generator i and
On the other hand Wide Area Measurement System is the acceleration time constant of the generator rated to its
(WAMS) has been installed and operated in Iran network and nominal active power ( ). Where, having nominal apparent
used for system analysis and validation of simulation power ( ) and nominal power factor ( ) of the
models[7], [8]. One of the main goals of WAMS is monitoring generator, it is clear that:
of system parameters in order to estimate the power system Pni = S ni .cos θ n (4)
state such as stability margin and enhancement of system
control. The main property of WAMS is synchronized On the other hand, for all power network, swing equation
measurement in Phasor Measurement Units (PMUs) which states that[3]:
158
2016 24th Iranian Conference on Electrical Engineering (ICEE)
ΔP
Pn
T a = 2H = (5)
1 df
f n dt
Figure 1 illustrates the different type of active power control
based on frequency deviation after a system event containing a
load-generation unbalance. The Rate of Change of Frequency
(ROCOF) which is available using PMU can be used for
calculation of the system inertia (H). Using equation (5) the
value of H network is calculated for large-scale networks is
usually between 2 and 9 seconds[1].
159
2016 24th Iranian Conference on Electrical Engineering (ICEE)
The difference between resulted values for inertia from two
approaches (HIranGrid* and HIranGrid†) is less than %2 which can
be neglected according to large scale data in modeling and
measuring procedure. Both approaches have obtained
approximately same values for the important parameter of Iran
network and it can be stated that:
HIranGrid § 5.2 sec. (8)
V. CONCLUSION
In this paper, the inertia constant of Iran power grid has
been obtained using two methods: simulation approach using
the largest simulation model of Iran power grid and PMU
based calculation. Outage of the generator of Bushehr power
plant as an effective event has been studied for obtaining the
inertia constant. Both approaches have obtained approximately
same values for the important parameter of Iran network. On
Figure 4: frequency data at first moments after the event the other hand, comparison of the results of two methods can
be useful for validation of the model because of approximate
B. Inertia Calculation matching between simulation and actual measurement data.
At the time of the outage of the power plant, a few
generators of network had been out of service. The initial REFERENCES
injected power from Bushehr generator to the grid before [1] P. Tielens and D. Van Hertem, “Grid Inertia and Frequency
event was about 900MW and it is considered as ǻP in eq. Control in Power Systems with High Penetration of
Renewables,” no. 2, pp. 1–6.
(5). Changes of frequency can be represented in pu using [2] P. Wall, F. González-Longatt, and V. Terzija,
nominal frequency fn=50 Hz as the base value. In Table.1 the “Demonstration of an inertia constant estimation method
rate of change of frequency (df/dt) called ROCOF in through simulation,” Univ. Power Eng. Conf. (UPEC), 2010
Choghadak substation (close to the power plant) is 45th Int., pp. 1–6, 2010.
calculated. For elimination of unwanted oscillations from the [3] D. P. Chassin, Z. Huang, M. K. Donnelly, C. Hassler, E.
Ramirez, and C. Ray, “Estimation of WECC system inertia
data, the average of rate of change of the network frequency using observed frequency transients,” IEEE Trans. Power
for 10 points in range of 400 ms after the event is considered. Syst., vol. 20, no. 2, pp. 1190–1192, 2005.
[4] T. Inoue, H. Taniguchi, Y. Ikeguchi, and K. Yoshida,
Table 1: Measurement data of bus frequency recorded in PMU and “Estimation of power system inertia constant and capacity of
ROCOF results spinning-reserve support generators using measured
Sample Number Time(sec) Frequency (Hz) df/dt (pu/sec) frequency transients,” IEEE Trans. Power Syst., vol. 12, no.
1, pp. 136–143, 1997.
1 33.522 49.90996 -
[5] Iran Grid Management Company and Regional Electric
2 33.565 49.90446 0.002530 Companies of Iran, “Nationwide Transmission and
3 33.609 49.90025 0.001937 Subtransmission Network Modeling,” 2015. [Online].
Available: http://igmc.ir/tabid/135/Default.aspx.
4 33.652 49.89681 0.001583 [6] DIgSILENT GmbH., “DIgSILENT PowerFactory (Version
5 33.696 49.89375 0.001409 15) User Manual.” DIgSILENT GmbH, Gomaringen,
6 33.739 49.89117 0.001186 Germany, 2014.
[7] A. a. Hajnoroozi, F. Aminifar, and H. Ayoubzadeh,
7 33.783 49.88947 0.000783 “Generating Unit Model Validation and Calibration Through
8 33.826 49.88764 0.000841 Synchrophasor Measurements,” IEEE Trans. Smart Grid,
vol. 6, no. 1, pp. 441–449, 2015.
9 33.870 49.88525 0.001097
[8] S. Rabiee, H. Ayoubzadeh, D. Farrokhzad, and F. Aminifar,
10 33.913 49.88264 0.001202 “Practical aspects of phasor measurement unit (PMU)
11 33.957 49.87912 0.001621 installation in power grids,” in Smart Grid Conference
(SGC), 2013, 2013, pp. 20–25.
Average 0.001419
160