KR102638339B1 - System and method for managing bid based on intelligent demand response - Google Patents

Abstract

The present invention relates to a demand response bidding management system and method. In particular, an intelligent demand response bidding management system that intelligently calculates resource capacity that can be reduced for customers wishing to bid and automatically creates a bidding strategy that maximizes profits. and methods thereof.
In addition, according to the present invention, a data storage that stores the customer's past power usage information; an intelligent energy analysis engine that extracts usage patterns using the customer's past power usage information stored in the data storage and performs demand forecasting based on the extracted usage patterns; and an intelligent demand response bidding management system and method including a processing module that generates a bidding strategy using the customer's usage patterns and demand forecasts and performs bidding, so that bid participants can easily respond to demand response bidding. do.

Description

Intelligent demand response bid management system and method {System and method for managing bid based on intelligent demand response}

The present invention relates to a demand response bidding management system and method. In particular, an intelligent demand response bidding management system that intelligently calculates resource capacity that can be reduced for customers wishing to bid and automatically creates a bidding strategy that maximizes profits. and methods thereof.

The demand response management business is divided into reliability demand response and economic demand response, each with differentiated bidding methods and transaction processes.

As of March 2015, the occurrence ratio of reliability demand response (occurring 3 times) and economic demand response (occurring 41 times) is approximately 1:13, so the scale of the economic demand response business is gradually increasing, and competitiveness in the economic demand response business is strengthened. To achieve this, the introduction of differentiated technology is necessary.

Most current economic demand response agency projects follow the manual-based economic demand response agency service system format shown in Figure 1.

The manual-based economical demand response agency service system requires many parts of the operator and bidder's manual work, and must rely entirely on the operator's expertise to formulate an efficient bidding strategy.

For this economical demand response business, KT, Byuksan, Enernoc, etc. recruit individual business owners as members and handle transactions between individuals and the Korea Power Exchange.

Manual work by experts with knowledge of energy management is required to operate the demand response system owned by each company.

In addition, the operator must always continuously monitor when the SMP (electricity market price) is high and configure the optimal bidding strategy to maximize profits based on SMP and CBL for each individual customer.

At this time, since there is no analysis technology that can accurately predict customer usage in advance, the savings can only be calculated by predicting based on CBL.

Because CBL is based on a simple average per hour for each of 6 days, the accuracy of predicting actual customer usage is low.

In addition, the operator groups scattered bidders in various time zones back into groups to enable competitive bidding (sufficient reduction amount) and bids to the final power exchange.

At this time, a strategy to appropriately group customers is needed, taking into account the circumstances of individual customers, but the current system lacks this function.

Meanwhile, in the current system, customers who want to bid must directly enter the time period and amount that can be reduced. As a result, profit maximization cannot be achieved because input is made without any prior knowledge of the time period in which profits can be maximized or the amount that one can reasonably reduce.

Additionally, in order to receive feedback on bidding strategies, you have no choice but to become more dependent on the operator.

Most demand response customers have power generation facilities, but there is a problem in that the company's power generation facility operation schedule according to the customer's input of savings must be entirely coordinated by the customer.

As such, the conventional economic demand response system requires a lot of direct intervention from energy management experts and customers to formulate operation and bidding plans.

Prior Art 1: KR Publication Patent No. 2014-0069737 Prior Art 2: KR Publication Patent No. 2012-0000116 Prior Art 3: KR Published Patent No. 2002-0066654

One aspect of the present invention is a data storage that stores the customer's past power usage information; an intelligent energy analysis engine that extracts usage patterns using the customer's past power usage information stored in the data storage and performs demand forecasting based on the extracted usage patterns; and a processing module that generates a bidding strategy using the customer's usage pattern and demand forecast and performs bidding.

In addition, the data storage in one aspect of the present invention stores past weather information, and the intelligent energy analysis engine analyzes the customer's past power use information stored in the data storage and past weather information to determine usage patterns. Extract .

In addition, the intelligent energy analysis engine of one aspect of the present invention performs demand prediction using Equation 1 below based on the extracted usage pattern.

(Equation 1)

Y _D =β ₁ Y _D-21 -β ₂ Y _D-14 -β ₃ Y _D-7 +…

Here, Y is the energy demand, D is the use date, and β is the weight coefficient.

In addition, the intelligent energy analysis engine of one aspect of the present invention lowers the weight coefficient the farther away the past is from the present, and applies it differently for each business type.

In addition, the intelligent energy analysis engine of one aspect of the present invention calculates the point in time when power usage is below the average through demand forecasting, and the processing module is a facility that suggests to the customer to artificially increase the amount of power usage when power usage is below the average. Includes an operation schedule recommendation module.

In addition, the intelligent energy analysis engine of one aspect of the present invention classifies customers with similar consumption patterns into multiple groups, and classifies customers within each group into customers who bid at a fixed time and customers who can bid flexibly, and the processing module Bidding by group is performed by unifying the bidding timing of customers within the group while coordinating the bidding timing and reduction amount of individual customers within the group using the multiple groups classified by the above intelligent energy analysis engine and the detailed classification of customers within the group. Includes strategy module.

In addition, the group-specific bidding strategy module of one aspect of the present invention performs regrouping with other groups in a different time period if a group does not accumulate a certain amount of savings even after grouping.

In addition, the processing module of one aspect of the present invention includes a customer portfolio creation module that creates a customer portfolio using the customer's past power use information, usage patterns, and demand forecast information; and a customer portfolio management module that updates the customer portfolio.

In addition, the processing module of one aspect of the present invention includes a customer profit prediction module that calculates sales profit using the SMP unit price on the day of bidding and allows the customer to decide whether to make a final bid.

In addition, the intelligent energy distribution engine of one aspect of the present invention recommends a demand response bidding point when excess idle power occurs while the facility is operating.

Meanwhile, another aspect of the present invention includes the steps of (A) an intelligent energy analysis engine extracting a usage pattern using the customer's past power usage information and performing demand forecasting based on the extracted usage pattern; and (B) the processing module generating a bidding strategy using the customer's usage pattern and demand forecast to perform the bidding.

In addition, in step (A) of another aspect of the present invention, the intelligent energy analysis engine extracts a usage pattern by analyzing the customer's past power usage information and past weather information.

In addition, the intelligent energy analysis engine in another aspect of the present invention performs demand prediction using Equation 1 below based on the extracted usage pattern.

(Equation 1)

Y _D =β ₁ Y _D-21 -β ₂ Y _D-14 -β ₃ Y _D-7 +…

Here, Y is the energy demand, D is the use date, and β is the weight coefficient.

In addition, another aspect of the present invention includes (C) the intelligent energy analysis engine calculating a point in time when power consumption is below the average through demand forecasting; and (D) the processing module comprising suggesting to the customer to artificially increase power usage at a time when power usage is below the average.

In addition, another aspect of the present invention is (E) the intelligent energy analysis engine classifies customers with similar consumption patterns into a plurality of groups; (F) the intelligent energy analysis engine classifies customers in each group into customers who can bid at a fixed time and customers who can bid fluidly; (G) the processing module adjusts the bidding timing and reduction amount of individual customers within the group using a plurality of groups classified by the intelligent energy analysis engine and detailed classification of customers within the group; And (H) the processing module further includes a step of performing group bidding by unifying the bidding timing of customers in the group.

In addition, another aspect of the present invention is that, after step (G), (I) if the processing module is a group that does not accumulate a certain amount of savings even after grouping, it further includes the step of regrouping with other groups in a different time zone. Includes.

In addition, another aspect of the present invention includes (J) the processing module generating a customer portfolio using the customer's past power use information, usage patterns, and demand forecast information; and (K) the processing module updating the customer portfolio.

In addition, another aspect of the present invention further includes (L) the step of recommending a demand response bidding point when the intelligent energy distribution engine generates remaining idle power while the facility is operating.

Therefore, in order to solve the above problems, the present invention provides an intelligent demand response bidding management system and method that intelligently calculates the resource capacity that can be reduced for customers wishing to bid and automatically creates a bidding strategy that maximizes profits. is to provide.

According to the present invention, an individually optimized bidding strategy (efficient CBL management, reasonable savings and bidding timing) is created through various analysis algorithms, thereby returning maximum sales revenue to customers and securing higher commissions.

In addition, according to the present invention, bidders with similar power consumption patterns are grouped into groups so that DR bids have sufficient competitiveness to be successful, and the bidding timing is reasonably unified according to the detailed circumstances of individual customers. I can do it for you.

Additionally, according to the present invention, customers experience increased profits and most unnecessary direct interventions are eliminated, thereby increasing convenience and reliability.

Additionally, according to the present invention, labor costs of operations management experts are reduced due to automation of operations management.

In addition, according to the present invention, systematic customer management is possible, such as creating and continuously managing individual portfolios and automatically updating them when customer feedback or changes occur.

In addition, according to the present invention, when the amount of power generation from energy facilities exceeds the predicted demand (more than 30%), it is possible to secure profits by reverse recommending economic DR participation.

In addition, according to the present invention, by automating the operation scheduling of the customer's power generation facility that can make up for the savings according to the customer's target savings amount, it supports the customer to achieve the savings goal with certainty, and accordingly, the bid amount is reduced. The achievement failure rate can be lowered.

1 is a conceptual diagram of a manual-based economic demand response (DR) bidding system according to the prior art.
Figure 2 is a conceptual diagram of an intelligent demand response bidding management system according to the present invention.
Figure 3 is a detailed configuration diagram of an intelligent demand response bidding management system according to a preferred embodiment of the present invention.
Figure 4 is a conceptual diagram of a group bidding management method according to the present invention.
Figure 5 is a flowchart of an intelligent demand response bidding management method according to a preferred embodiment of the present invention.
Figure 6 is a flowchart of the bidding management process for each group in Figure 5.
Figure 7 is a flow chart of the CBL efficiency management method used in the present invention.
Figure 8 is a flow chart of the portfolio creation process used in the present invention.
Figure 9 is a flowchart of the power generation facility operation schedule and DR bidding timing reverse recommendation process used in the present invention.

The present invention can be modified in various ways and can have various embodiments. Hereinafter, specific embodiments will be described in detail based on the accompanying drawings.

In describing the present invention, if it is determined that a detailed description of related known technologies may obscure the gist of the present invention, the detailed description will be omitted.

Terms such as first, second, etc. may be used to describe various components, but the components are not limited by the terms, and the terms are used only for the purpose of distinguishing one component from another component. It is used.

Figure 2 is a conceptual diagram of an intelligent demand response bidding management system according to the present invention.

Referring to FIG. 2, the intelligent demand response bidding management system according to the present invention presents a time zone and reduction amount that can generate maximum profit compared to CBL for each individual customer, schedules power generation facility operation accordingly, improves CBL efficiency, and provides a group-level bidding strategy to increase the success rate. automatically analyzes and manages them.

To this end, the intelligent demand response bidding management system makes sophisticated demand forecasts on an hourly/daily/weekly basis using a data mining algorithm based on the customer's past power usage data, and calculates the time zone when the predicted value differs the most from the actual CBL by the difference. It automatically sets the reduction amount and configures the bidding strategy.

In addition, the intelligent demand response bid management system creates and systematically manages a DR portfolio for each customer, and automatically updates it when changes occur in the customer's real-time power usage status, bidding information, or facilities.

In this way, according to the present invention, operation becomes convenient, costs can be reduced by replacing professional personnel, DR agency fees can be increased by maximizing customer profits, and customer service reliability can be improved.

Figure 3 is a detailed configuration diagram of an intelligent demand response bidding management system according to a preferred embodiment of the present invention.

Referring to Figure 3, the intelligent demand response bidding management system according to a preferred embodiment of the present invention consists of an intelligent energy analysis engine 100 that provides various analysis functions, a processing module 200, and a data storage 400. there is.

In addition, the processing module 200 includes a trusted customer selection module 210, a customer characteristic extraction module 220, a customer portfolio creation module 230, a customer optimal bidding strategy module 240, and a group-specific bidding strategy. Module 250, facility operation schedule recommendation module 260, external linkage module 270, customer feedback processing module 280, customer portfolio management module 290, customer bid production module 300, and , It consists of a customer profit prediction module 310 and a bidding module 320.

Additionally, the data storage 400 consists of a customer portfolio database 410, a weather information database 420, and an electricity market information database 430.

In this configuration, the intelligent energy analysis engine 100 extracts regular and distinct power consumption patterns based on the customer's past power use data and analyzes customer patterns.

In addition, the intelligent energy analysis engine 100 performs demand prediction analysis by predicting precise future power use on an hourly/daily/weekly basis based on extracted customer patterns.

This kind of demand forecasting analysis extracts patterns and makes predictions based on data from the past three weeks.

For example, Monday's power usage forecast is calculated using only Monday data from the past three weeks.

The prediction equation is as shown in Equation 1 below. The farther the past is from the present, the lower the weight coefficient, and this weight coefficient is applied differently for each business type (factory, hotel, hospital, etc.).

The demand forecasting formula in Equation 1 below reflects usage patterns in detail by day of the week, making it more accurate than other companies' CBLs that predict simply using the average of the past 10 days.

(Equation 1)

Y _D =β ₁ Y _D-21 -β ₂ Y _D-14 -β ₃ Y _D-7 +…

Here, Y is the energy demand, D is the use date, and β is the weight coefficient.

Meanwhile, the intelligent energy analysis engine 100 performs weather fusion analysis to analyze related patterns by fusing the customer's past power use data and past weather information.

In addition, since the intelligent energy analysis engine 100 generates revenue based on the difference between CBL and the actual amount of electricity used, if a customer wishes to bid in advance (7 to 5 days in advance), it predicts demand to determine the amount of electricity used in the future. A CBL efficiency analysis is performed to increase overall CBL by proposing to artificially increase usage at points where it falls significantly from the average.

For example, during the CBL efficiency analysis, the intelligent energy analysis engine 100 suggests, "If the point in time when usage drops significantly is predicted, do not turn on the power generation facilities the day before and use the power supplied by KEPCO?" (However, the proposed method at this time Therefore, the cost of artificially consumed power is lower than the expected profit expected from DR participation).

Once the optimal savings amount is determined according to the customer's usage pattern and demand forecast, the intelligent energy analysis engine 100 performs facility operation scheduling to analyze the efficient power generation facility operation schedule to maximize the savings.

In addition, the intelligent energy analysis engine 100 also analyzes the operation schedule of power generation facilities to artificially increase CBL through CBL efficiency analysis.

In addition, the intelligent energy analysis engine 100 automatically recommends economically feasible DR bidding to customers when the amount of power charged to the power generation facility is more than +30% higher than the predicted demand after 1 to 2 days, thereby securing profits. I do it.

In addition, the intelligent energy analysis engine 100 needs to work to ensure that the DR bid is sufficiently competitive to be successful. To this end, bidders are grouped into groups, the timing of each incorrect bid is unified to some extent, and the savings are added up for bidding. Performs customer grouping tasks.

In this regard, referring to FIG. 4, the intelligent energy analysis engine 100 first classifies groups with similar consumption patterns for customer grouping. In this way, when groups with similar consumption patterns are classified, bidding management becomes similar because the bidding patterns are also similar.

In addition, the intelligent energy analysis engine 100 analyzes past bidding patterns for customer grouping and reclassifies customers who can bid only at fixed times and customers who can bid flexibly.

Classification in this way takes into account the customer's situation and ensures maximum reliability in implementing savings.

In addition, the intelligent energy analysis engine 100 calculates various time zone unification standards by repeatedly combining them to be optimal according to the demand situation for each customer for customer grouping work.

Meanwhile, the trust customer selection module 210 uses the customer trust analysis function of the intelligent energy analysis engine 100 to determine whether the usage pattern is sufficiently regular and operates an economic demand response bidding pool based on this.

In addition, the customer characteristic extraction module 220 utilizes the customer pattern analysis, demand forecast analysis, and weather information convergence analysis functions of the intelligent energy analysis engine to derive usage patterns from the customer's past data and based on this on an hourly/daily/weekly basis. Generate demand forecast information.

Next, the customer portfolio creation module 230 collects the customer's past power use information, usage pattern characteristics, hourly/daily/weekly forecast information, real-time CBL information, customer's facility operation schedule information, and past bidding information. Comprehensively create a portfolio.

In addition, the customer optimal bidding strategy module 240 calculates the point when the demand forecast for the next day is measured to be 10% or more lower than the CBL standard in real time every day from the day of the day until the end of the bidding if the customer wishes to bid 5 to 7 days in advance. A savings of 10% to 30% is recommended depending on the customer type (whether or not they have power generation facilities).

In addition, through the CBL efficiency analysis function of the intelligent analysis engine, the overall CBL can be raised by suggesting to the customer to artificially increase the amount of power used at a time when power use is predicted to fall significantly from the cumulative average.

Meanwhile, the group-specific bidding management module 250 provides the optimal bidding strategy for each customer, but in order to increase the final winning rate, a strategy of combining the savings of individual customers and bidding on a group basis is necessary, and the customer grouping function of the intelligent energy analysis engine is required. Using this, bidders with similar power consumption patterns are classified into groups, and the bidding plans of individual customers within each group are coordinated based on individual portfolios and existing optimal bidding strategies to unify the bidding timing to some extent.

In addition, the bid management module 250 for each group later combines the savings of unified customers and bids on a group basis to ensure a high level of success rate.

Next, the facility operation schedule recommendation module 260 uses an intelligent energy analysis engine to secure the proposed reduction amount at the time of bidding in case of a customer who owns power generation facilities (most of KT's demand management business customers have power generation facilities). It recommends the optimal operation time using the facility operation scheduling function.

In addition, when the amount of power charged to the power generation facility is more than +30% higher than the predicted demand 1 to 2 days later, it automatically recommends an economical DR bid to the customer to secure profits.

Meanwhile, the external interconnection module 270 retrieves customer real-time power usage information, SMP unit price, electricity market information such as bidding, and weather information from an external system in real time.

Next, the customer feedback processing module 280 readjusts the bidding strategy and facility operation schedule by reflecting customer feedback on the optimal bidding plan proposed by the system, if any.

Additionally, the customer portfolio management module 290 periodically manages the customer portfolio and automatically updates it when changes occur in real-time customer power usage status, bidding information, or equipment.

Next, the customer bid production module 300 automatically creates a bid reflecting the relevant bidding information when the customer accepts the bidding strategy proposed by the system.

In addition, the customer profit prediction module 310 calculates the sales profit using the SMP unit price on the day of the bidding, and the customer decides whether to make a final bid.

Next, the bidding module 320 groups bidders by time zone through customer grouping analysis of the intelligent energy analysis engine so that the system can be competitive enough to win the DR bid, and groups bidders for which a sufficient amount is not collected even after grouping. regroups with other groups in the closest time zone and makes a final bid to the power exchange.

Meanwhile, the customer portfolio database 410 of the data storage 400 stores customers' real-time power use information, usage pattern and forecast information, bidding information, and power generation facility information.

Additionally, the electricity market information database 420 stores electricity market information such as real-time SMP unit price, past bids, and successful bid status.

The weather information database 430 stores real-time regional weather information.

The operation of the intelligent demand response bidding management system according to an embodiment of the present invention configured as described above will be examined with reference to the flowchart of the intelligent demand response bidding method in FIG. 5 as follows.

First, the trust customer selection module 210 uses the customer trust analysis function of the intelligent energy analysis engine 100 to determine whether the usage pattern is sufficiently regular and selects trust customers based on this to determine economic demand made up of the selected trust customers. Operates a reaction bidding pool (S100).

When a customer in such a demand response bidding pool (poo) requests a bid from the intelligent demand response bidding management system, the customer characteristic extraction module 220 performs customer pattern analysis and demand forecast analysis of the intelligent energy analysis engine 100. and weather information convergence analysis function to derive usage patterns from customers' past data and generate hourly/daily/weekly demand forecast information based on this (S110).

Afterwards, if the customer wishes to bid 5 to 7 days in advance, the customer optimal bidding strategy module 240 determines the bidding time at the point when the demand forecast for the next day is measured to be 10% or more lower than the CBL standard in real time every day from the day of application until the end of bidding. , and 10% to 30% is calculated as savings depending on the customer type (whether or not they have power generation facilities).

In addition, the customer optimal bidding power module 240, through the CBL efficiency analysis function of the intelligent energy analysis engine 100, proposes to the customer to artificially increase the amount of power used at a time when power use is predicted to fall significantly from the cumulative average, thereby reducing the overall CBL. Allow it to be raised.

Meanwhile, the customer profit prediction module 310 calculates the expected profit based on the SMP at the time of bidding.

The customer optimal bidding power module 240 uses the calculated bidding timing and savings amount and the expected profit of the customer profit prediction module 310 to generate bidding power for each customer and proposes it to the customer. If the customer approves, the strategy module Confirm with

In other words, the customer optimal bidding power module 240 creates the best bidding strategy for each customer by calculating the bidding timing, savings amount, and expected profit with demand forecast information based on the customer's usage pattern characteristics (S120).

Next, the facility operation schedule recommendation module 260 uses an intelligent energy analysis engine to secure the amount of savings proposed at the time of bidding for customers who own power generation facilities (most of KT's demand management business customers own power generation facilities). It recommends the optimal operation time using the facility operation scheduling function.

At this time, the customer feedback processing module 280 readjusts the bidding strategy and facility operation schedule by reflecting the customer's feedback on the bidding plan proposed by the customer optimal bidding power module 240 (S130).

And, the group-specific bidding strategy module 250 performs group-specific bidding management (S140).

This group-specific bid management process is explained using the group-specific bid management flowchart of FIG. 6 as follows.

First, when the bidding power of an individual customer is configured, the intelligent energy analysis engine 100 analyzes customer demand patterns (S200), clusters similar customers based on the last 1 to 2 weeks, and classifies them into multiple groups (S210). The reason for this classification is that customers with similar consumption patterns are believed to have similar bidding strategies.

In addition, the intelligent energy analysis engine 100 performs a detailed classification of customers within the group into customers who must bid only at a fixed point in time, even if they have similar consumption patterns, and customers who can bid fluidly because they own power generation facilities (S220). .

At this time, the intelligent energy analysis engine 100 is related to the customer portfolio created by the customer portfolio production module 220, including customer power generation facility status, customer strategy use history, customer response history, bidding history, etc. in the customer portfolio database 410. Use customer demand forecast information, etc.

In this way, when the detailed classification of customers within the group is completed, the group bidding power module 250 coordinates the bidding timing and reduction amount of individual customers within each group (S230).

The reason why the function to coordinate the bidding timing and savings amount of existing individual customers is necessary is because to be competitive only when bidding by collecting sufficient savings amount for each group, and the coordination timing and savings amount should not significantly impede the existing expected bidding profits of individual customers. do.

When the group bidding strategy module 250 selects the optimal bidding time, various standard factors are reflected, and the reflected standard factors include customers' portfolios and external information (SMP information by time, success rate by past time period, etc.).

That is, the following criteria elements 1) to 6) are considered.

1. High SMP time zone

2. Times with high bid success rate and minimum savings amount accordingly

3. Time zone selected by high-quality customers within the group: The reliability and best customers of each customer are determined based on past bidding performance and bidding savings, and the bidding times of these customers can be used as a standard.

4. Time zone selected by multiple customers within the group: The timings at which multiple customers within the group bid can be used as the standard.

5. For customers within the group who can only bid at a fixed time, coordination is only possible for +-1 hour at the fixed time, and once the time is coordinated, the final target savings amount may vary based on existing demand forecasting. Taking this into account, there should not be much change even when adjusted compared to the existing expected profit.

6. Customers in the group who can bid fluidly by owning power generation facilities comprehensively judge 1, 2, 3, and 4 and adjust to the point where the winning rate is high and maximum profit is possible. If these customers are adjusted compared to the existing expected profits, there should not be much change.

Meanwhile, the group bidding strategy module 250 repeatedly combines the standard elements (1, 2, 3, 4, 5, 6) to determine the points in time when the probability of successful bid is high and the amount of savings generated at these points. By comprehensively considering the adequacy of customers' profits, the bidding timing of customers within the group is unified to some extent (3 to 5) (S240).

If the group-specific bidding strategy module 250 is a group that does not accumulate sufficient savings even after grouping, it performs regrouping by combining it with other clusters (groups) in the closest time zone (S250).

In addition, the group bidding power module 250 notifies customers who have experienced changes in the existing bidding plan of the coordinated strategy plan and receives final approval (S260). It shows why coordination was done along with detailed information.

The optimal bidding times for each group are also divided into ranks 1 to 3, allowing customers to intuitively check the probability of winning each bid (based on the success rate of past winning bid data) and the change in expected profit according to the bidding time, so that they can determine the bidding time they want. Let them choose.

For customers who can only bid at a fixed time, only 1st or 2nd ranked coordinated bidding times are presented, but for customers who can bid fluidly, all coordinated bidding times can be presented in a wide range of ways.

Through this process, the group-specific bidding strategy module 250 returns to the stage of coordinating the bidding timing of individual customers within each group by canceling the bid for customers who do not like all of the rankings or entering their own bids. Once the customer gives final approval, bidding is completed for each group (S150).

That is, a bid is prepared using the customer bid creation module 300, and a power exchange bidding is conducted through the bid module 310.

Meanwhile, as shown in FIG. 7, when a customer's bid request occurs, the facility operation schedule recommendation module 260 starts CBL efficiency management up to 7 days after the request (S300).

The facility operation schedule recommendation module 260 suggests a method to artificially increase the CBL when the demand forecast at a specific point in time is measured to be 30% lower than the 7-day forecast average (S310).

For example, for customer companies with power generation facilities, the facility operation schedule recommendation module 260 provides measures such as “Do not operate the power generation facilities the day before the relevant time, but use KEPCO’s power to raise the CBL at the relevant time.” (However, the cost of power consumption artificially required by the proposed plan is lower than the expected profit expected from DR participation).

In this way, if the customer follows the suggestions of the facility operation schedule recommendation module 260, the CBL value increases and the actual DR compensation increases.

Next, the customer portfolio creation module 230 creates a customer portfolio. Referring to FIG. 8, the customer portfolio creation module 230 creates a portfolio for each customer (S400).

Additionally, the customer portfolio creation module 230 enters the customer's power generation facilities, business size, type, location, etc. into the portfolio (S410).

Next, the customer portfolio creation module 240 writes the customer's power generation facility operation schedule into the portfolio (S420).

In addition, the customer portfolio creation module 240 retrieves the customer's power usage data in real time from KEPCO's ISmart system through the external interconnection module 270 and records it in the portfolio (S430).

Next, the customer portfolio creation module 240 records consumption patterns extracted from the customer characteristic extraction module 220 based on the customer's past power use data in the portfolio (S440).

Additionally, the customer portfolio creation module 240 records the CBL calculated by the customer characteristic extraction module 220 at each time based on past power usage in the portfolio (S450).

In addition, the customer portfolio creation module 240 records each time a customer makes a bid in the portfolio and records success and secured profits in the portfolio (S460).

Next, the customer portfolio management module 290 updates the customer portfolio whenever relevant information changes (S470).

Meanwhile, the customer optimal bidding strategy module 240 performs reverse recommendation of the power generation facility operation schedule and DR bidding timing. As explained with reference to FIG. 9, the customer's bidding timing and savings amount are first determined (S500).

Then, the customer optimal bidding strategy module 240 reads the existing facility schedule from the customer's portfolio (S510).

Next, the customer optimal bidding strategy module 240 adjusts the operation schedule of the power generation facility from -1 day to +1 day at the customer's bidding time (S520).

By adjusting the operation schedule like this, customers can secure savings by producing their own electricity through the operation of power generation facilities.

Meanwhile, the intelligent energy analysis engine 100 automatically recommends to the customer whether to bid for economical demand response (DR) when the amount of power charged to the power generation facility is more than +30% higher than the predicted demand after 1 to 2 days, thereby generating profits. It allows you to secure it (S530).

According to the present invention, an individually optimized bidding strategy (efficient CBL management, reasonable savings and bidding timing) is created through various analysis algorithms, thereby returning maximum sales revenue to customers and securing higher commissions.

In addition, according to the present invention, bidders with similar power consumption patterns are grouped into groups so that DR bids have sufficient competitiveness to be successful, and the bidding timing is reasonably unified according to the detailed circumstances of individual customers. I can do it for you.

Additionally, according to the present invention, customers experience increased profits and most unnecessary direct interventions are eliminated, thereby increasing convenience and reliability.

Additionally, according to the present invention, labor costs of operations management experts are reduced due to automation of operations management.

In addition, according to the present invention, systematic customer management is possible, such as creating and continuously managing individual portfolios and automatically updating them when customer feedback or changes occur.

In addition, according to the present invention, when the amount of power generation from energy facilities exceeds the predicted demand (more than 30%), it is possible to secure profits by reverse recommending economic DR participation.

In addition, according to the present invention, by automating the operation scheduling of the customer's power generation facility that can make up for the savings according to the customer's target savings amount, it supports the customer to achieve the savings goal with certainty, and accordingly, the bid amount is reduced. The achievement failure rate can be lowered.

Although the present invention has been described above with reference to preferred embodiments, those skilled in the art can modify the present invention in various ways without departing from the spirit and scope of the present invention as set forth in the following patent claims. and that it can be changed.

100: Intelligent energy analysis engine 200: Processing module
210: Trusted customer selection module 220: Customer characteristic extraction module
230: Customer portfolio creation module 240: Customer optimal bidding strategy module
250: Group-specific bidding strategy module 260: Facility operation schedule recommendation module
270: External linkage module 280: Customer feedback processing module
290: Customer portfolio management module 300: Customer bid production module
310: Customer revenue prediction module 320: Bidding module
400: Data storage 410: Customer portfolio database
420: Weather information database 430: Electricity market information database

Claims (18)

A data storage that stores customer's past power usage information;
an intelligent energy analysis engine that extracts usage patterns using the customer's past power usage information stored in the data storage and performs demand forecasting based on the extracted usage patterns; and
It includes a processing module that performs bidding by creating a bidding strategy using the customer's usage pattern and demand forecast,
The intelligent energy analysis engine performs customer grouping to classify and group the customers based on the usage patterns,
The processing module is
a customer optimal bidding strategy module that generates a bidding strategy for each customer using the bidding timing and savings amount calculated from the demand forecast according to the customer's usage pattern and the resulting expected profit; and
By coordinating the bidding timing and savings amount of individual customers belonging to the group according to the customer-specific bidding strategy, it includes a group-specific bidding strategy module that unifies the bidding timing of customers in the group and makes group bids,
The bidding strategy module for each group is
An intelligent demand response bidding management system that unifies the bidding timing of the group bidding according to the profit adequacy of the individual customer generated by the points in which the probability of winning the group bid is high and the amount of savings at each point. In claim 1,
The data storage stores past weather information,
The intelligent energy analysis engine is an intelligent demand response bid management system that extracts usage patterns by analyzing the customer's past power use information and past weather information stored in the data storage. The method of claim 1, wherein the intelligent energy analysis engine
Based on the extracted usage pattern, past energy demands on the same day of the week as the usage day are obtained, and different weight coefficients are given to each of the past energy demands to perform demand forecasting for the energy demand on the usage day. , an intelligent demand response bid management system. In claim 3,
The intelligent energy analysis engine lowers the weight coefficient the further away the past is from the present, and applies it differently for each business type. An intelligent demand response bid management system. In claim 1,
The intelligent energy analysis engine calculates the point in time when power usage is below the average through demand forecasting,
The processing module is an intelligent demand response bidding management system that includes a facility operation schedule recommendation module that suggests to the customer to artificially increase power usage at a time when power usage is below the average. In claim 1,
The intelligent energy analysis engine classifies customers with similar consumption patterns into multiple groups, and classifies customers within each group into fixed-point bidding customers and dynamic bidding customers.
The group-specific bidding strategy module uses the multiple groups classified by the intelligent energy analysis engine and the detailed classification of customers within the group to coordinate the bidding timing and savings amount of individual customers within the group and unifies the bidding timing of customers within the group. An intelligent demand response bid management system that conducts bidding. In claim 6,
The group-specific bidding strategy module is an intelligent demand response bidding management system that performs regrouping with other groups at different times if a group does not accumulate a certain amount of savings even after grouping. In claim 6,
The processing module is
A customer portfolio creation module that creates a customer portfolio using the customer's past power use information, usage patterns, and demand forecast information; and
An intelligent demand response bid management system comprising a customer portfolio management module that updates the customer portfolio. In claim 6,
The processing module is
It includes a customer profit prediction module that calculates sales profit using the SMP unit price on the day of bidding and allows the customer to decide whether to make a final bid, or
Using the bidding strategy module for each group, the bid timing of the individual customers is reflected at least one of SMP information by time, success rate by past time, a time zone selected by a good customer in the group, and a time zone selected by multiple customers in the group. and an intelligent demand response bid management system that coordinates savings. In claim 1,
The intelligent energy analysis engine is an intelligent demand response bidding management system that recommends the timing of demand response bidding when remaining idle power occurs while the facility is operating. (A) The intelligent energy analysis engine extracts usage patterns using the customer's past power usage information and performs demand forecasting based on the extracted usage patterns; and
(B) wherein the processing module generates a bidding strategy using customer usage patterns and demand forecasts to perform bidding,
The intelligent energy analysis engine performs customer grouping to classify and group the customers based on the usage patterns,
The processing module bids by unifying the bidding timing of customers belonging to the group,
The steps for performing the bidding are
A customer optimal bidding strategy module generating a bidding strategy for each customer using the bidding timing and reduction amount calculated from demand prediction according to the customer's usage pattern and the resulting expected profit; and
The group bidding strategy module further includes the step of group bidding by unifying the bidding timing of customers within the group by coordinating the bidding timing and reduction amount according to the bidding strategy for each customer of individual customers belonging to the group,
The group bidding step is
Intelligent, where the group bidding strategy module unifies the bidding timing of the group bidding according to the profit adequacy of the individual customer generated by the time points with a high probability of winning in the group bid and the amount of savings at each time point. How to manage demand response bidding. In claim 11,
In step (A), the intelligent energy analysis engine extracts a usage pattern by analyzing the customer's past power usage information and past weather information. The method of claim 11, wherein the intelligent energy analysis engine
Based on the extracted usage pattern, past energy demands on the same day of the week as the usage day are obtained, and different weight coefficients are given to each of the past energy demands to perform demand forecasting for the energy demand on the usage day. , an intelligent demand response bid management method. In claim 11,
(C) the intelligent energy analysis engine calculates the point in time when power consumption is below the average through demand forecasting; and
(D) The intelligent demand response bidding management method comprising the step of the processing module suggesting to the customer to artificially increase power usage at a time when power usage is below the average. In claim 11,
(E) the intelligent energy analysis engine classifies customers with similar consumption patterns into multiple groups;
(F) the intelligent energy analysis engine classifies customers in each group into fixed-point bidding customers and dynamic bidding customers;
(G) the processing module adjusts the bidding timing and savings amount of individual customers within the group using a plurality of groups classified by the intelligent energy analysis engine and detailed classification of customers within the group; and
(H) The intelligent demand response bidding management method further includes the step of the processing module performing group bidding by unifying the bidding timing of customers in the group. The method of claim 15,
After step (G) above,
(I) An intelligent demand response bidding management method further comprising the step of performing regrouping with other groups in a different time zone if the processing module is a group that does not accumulate a certain amount of savings even after grouping. The method of claim 15,
(J) the processing module generating a customer portfolio using the customer's past power use information, usage patterns, and demand forecast information; and
(K) the intelligent demand response bid management method further comprising the step of the processing module updating a customer portfolio. In claim 11,
(L) An intelligent demand response bid management method further comprising recommending a demand response bid timing when the intelligent energy division engine generates remaining idle power while the facility is operating.

Images