Information Processing Apparatus, Information Processing Method, and Program

Abstract

Provided are an information processing apparatus, an information processing method, and a program that improve the prediction accuracy of electricity prices while achieving consistency among a plurality of markets. 【Solution means】An information processing apparatus 1 includes an acquisition unit 10 that acquires information regarding operating power sources by applying a power demand prediction to a power source operation model, and a prediction unit 20 that predicts the electricity price in a first system by applying the information regarding the operating power sources to a first individual model, and predicts the electricity price in a second system different from the first system by applying the information regarding the operating power sources to a second individual model.

Description

Technical Field

[0001] The present invention relates to an information processing apparatus, an information processing method, and a program for predicting electricity prices.

Background Art

[0002] Conventionally, attempts have been made to predict electricity prices in each market. For example, in Patent Document 1, the amount of electricity procured from each of a plurality of power procurement means is used as an operation variable, and a value including the total cost required to procure the amount of electricity corresponding to the predicted demand using the plurality of power procurement means is used as an objective function, and a model generation means for generating an optimization model having a constraint condition regarding the amount of electricity that can be procured from the power procurement means, and a determination means for determining a power procurement plan indicating the amount of electricity procured from each of the plurality of power procurement means by solving the optimization model by a mathematical optimization method have been proposed. It is disclosed that this information processing apparatus has a power price prediction means for predicting a power price based on the history of past power prices.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] In the conventional aspect as shown in Patent Document 1, it is possible to predict the electricity price in a single market such as the spot market, but it is not possible to sufficiently achieve consistency among the markets, and as a result, the prediction accuracy is not sufficient.

[0005] The present invention has been made in view of such points, and provides an information processing apparatus, an information processing method, and a program capable of improving the prediction accuracy of electricity prices while achieving consistency among systems such as a plurality of markets.

Means for Solving the Problem

[0006] [Concept 1] An information processing apparatus according to a first aspect of the present invention includes: an acquisition unit that acquires information regarding an operating power source by applying a power demand prediction to a power source operation model; a prediction unit that predicts the electricity price in a first system by applying the information regarding the operating power source to a first individual model, and predicts the electricity price in a second system different from the first system by applying the information regarding the operating power source to a second individual model; and may be provided.

[0007] [Concept 2] In the information processing apparatus according to Concept 1, the electricity price in the first system is any one of a contracted price in the spot market, a contracted price in the ΔkW demand-supply adjustment market, a registered price in the kWh demand-supply adjustment market, an area-specific future contracted price in the time-ahead market, and an imbalance price, and the electricity price in the second system is any one of a contracted price in the spot market, a contracted price in the ΔkW demand-supply adjustment market, a registered price in the kWh demand-supply adjustment market, an area-specific future contracted price in the time-ahead market, and an imbalance price, and may be the electricity price in a second system different from the first system.

[0008] [Concept 3] In the information processing apparatus according to Concept 2, the electricity price in the first system is a contracted price in the spot market or a contracted price in the ΔkW demand-supply adjustment market, the electricity price in the second system is a registered price in the kWh demand-supply adjustment market, an area-specific future contracted price in the time-ahead market, or an imbalance price, The prediction unit may predict the power generation plan and adjustable power that can be supplied for each power source from the concluded results of prices in the first system, or predict the power sources that are operating at the actual supply and demand time point for the spot market, and predict the power sources that are waiting to operate for the ΔkW supply and demand adjustment market at the actual supply and demand time point, and use the results of the prediction to predict the electricity price in the second system.

[0009] [Concept 4] In the information processing apparatus according to any one of Concepts 1 to 3, The prediction unit determines whether a continuous operation power source that operates continuously for a predetermined time from the concluded results of prices is operating. If it is determined that the continuous operation power source is operating, the electricity price in the first system or the second system may be predicted on the premise that the continuous operation power source operates for a predetermined time.

[0010] [Concept 5] The information processing apparatus according to the second aspect of the present invention is An acquisition unit that acquires information on operating power sources by applying demand prediction to a power source operation model, Predict the agreed price in the spot market by applying the information on the operating power source to the spot market model, predict the agreed price in the ΔkW supply and demand adjustment market by applying the information on the operating power source to the ΔkW supply and demand adjustment market model, predict the registered price in the kWh supply and demand adjustment market by applying the information on the operating power source to the kWh supply and demand adjustment market model, predict the area-specific future agreed price in the time-ahead market by applying the information on the operating power source to the time-ahead market model, and predict the imbalance price by applying the information on the operating power source to the imbalance price model. A prediction unit, May be provided.

[0011] [Concept 6] In the information processing apparatus according to Concept 5, The prediction unit predicts the power generation plans and adjustable power that can be supplied by each power source from the concluded price results in the spot market and the supply-demand adjustment market, or predicts the power sources operating at the actual supply-demand time point from the concluded price results in the spot market and predicts the power sources waiting to operate at the actual supply-demand time point from the concluded price results in the ΔkW supply-demand adjustment market, and may predict the registered price in the kWh supply-demand adjustment market, the area-specific future concluded price in the pre-time market, and the imbalance price using the results of the prediction.

[0012] [Concept 7] In the information processing apparatus according to Concept 5 or 6, the prediction unit determines whether a continuously operating power source that operates continuously for a predetermined time is operating from the concluded price results, and if it is determined that the continuously operating power source is operating, on the premise that the continuously operating power source operates for a predetermined time, it may predict the concluded price in the spot market, the concluded price in the ΔkW supply-demand adjustment market, the registered price in the kWh supply-demand adjustment market, the area-specific future concluded price in the pre-time market, and the imbalance price.

[0013] [Concept 8] The information processing method according to the present invention is a step of obtaining information on operating power sources by applying a demand prediction to a power source operation model by an acquisition unit; a step of predicting the concluded price in the spot market by applying the information on the operating power sources to a spot market model by a prediction unit, predicting the concluded price in the ΔkW supply-demand adjustment market by applying the information on the operating power sources to a ΔkW supply-demand adjustment market model, predicting the registered price in the kWh supply-demand adjustment market by applying the information on the operating power sources to a kWh supply-demand adjustment market model, predicting the area-specific future concluded price in the pre-time market by applying the information on the operating power sources to a pre-time market model, and predicting the imbalance price by applying the information on the operating power sources to an imbalance price model; may be provided.

[0014] [Concept 9] The program according to the present invention is A program for installation in an information processing apparatus, in the information processing apparatus in which the program is installed, a function of obtaining information regarding the operating power supply by applying demand prediction to a power operation model; a function of predicting the agreed price in the spot market by applying the information regarding the operating power supply to a spot market model, predicting the agreed price in the ΔkW supply-demand adjustment market by applying the information regarding the operating power supply to a ΔkW supply-demand adjustment market model, predicting the registered price in the kWh supply-demand adjustment market by applying the information regarding the operating power supply to a kWh supply-demand adjustment market model, predicting the area-specific future agreed price in the time-ahead market by applying the information regarding the operating power supply to a time-ahead market model, and predicting the imbalance price by applying the information regarding the operating power supply to an imbalance price model; may be realized.

Effect of the Invention

[0015] In the present invention, by applying power demand prediction to a power operation model, information regarding the operating power sources is obtained. By applying the information regarding the operating power sources to a first individual model, the electricity price in a first system is predicted. When adopting a mode of predicting the electricity price in a second system different from the first system by applying the information regarding the operating power sources to a second individual model, it is possible to improve the prediction accuracy of the price related to electricity while achieving consistency among multiple systems. Also, in the present invention, when a prediction unit predicts the agreed price in the spot market by applying information regarding the operating power sources to a spot market model, predicts the agreed price in the ΔkW supply-demand adjustment market by applying the information regarding the operating power sources to a ΔkW supply-demand adjustment market model, predicts the registered price in the kWh supply-demand adjustment market by applying the information regarding the operating power sources to a kWh supply-demand adjustment market model, predicts the area-specific future agreed price in the pre-time market by applying the information regarding the operating power sources to a pre-time market model, and predicts the imbalance price by applying the information regarding the operating power sources to an imbalance price model, it is possible to improve the prediction accuracy of the electricity prices in the spot market, the ΔkW supply-demand adjustment market, the kWh supply-demand adjustment market, and the pre-time market, as well as the imbalance price.

Brief Description of the Drawings

[0016]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Embodiments for Carrying Out the Invention

[0017] Embodiment Hereinafter, embodiments of an information processing apparatus and an information processing method according to the present invention will be described. In this embodiment, a program that can be installed in a computer such as a personal computer so that the computer can execute the information processing method of this embodiment, and a recording medium on which the program is recorded are also provided.

[0018] The information processing apparatus of this embodiment may be installed anywhere, may be a server, or may be used in a cloud environment. The information processing apparatus of this embodiment may be composed of one device or a plurality of devices. Also, when the information processing apparatus is composed of a plurality of devices, it is not necessary for each device to be provided in the same space such as the same room, and they may be provided in different rooms, different buildings, different regions, etc. Further, when the information processing apparatus is composed of a plurality of devices, a part of them may be owned and / or managed by one organization, and the rest may be owned and / or managed by another organization.

[0019] The information processing apparatus according to this embodiment is generated, for example, by installing a program. This program may be distributed by e-mail, may be obtained by accessing a predetermined URL and then logging in, or may be recorded on a recording medium. The program according to this embodiment is used to generate the information processing apparatus shown below, and the recording medium according to this embodiment is used to record the program. Further, the information processing method according to this embodiment is executed by an information processing apparatus in which the above program is installed. The information processing apparatus may execute the information processing method according to this embodiment by executing an application installed in the information processing apparatus.

[0020] As shown in FIG. 8, the information processing apparatus 1 according to this embodiment applies the power demand prediction to the power supply operation model, and at the supply and demand time point, the operating power supply (this "operating power supply" includes "the power supply waiting for operation"), the non-operating power supply, and the acquisition unit 10 that acquires information on the operating power supply including the equipment utilization rate of the operating power supply, and applies the information on the operating power supply to a certain individual model (the first individual model) to predict the electricity price in the first system, and applies the information on the operating power supply to another individual model (the second individual model) to predict the electricity price in the second system different from the first system. And a prediction unit 20, and an output unit 90 that outputs the prediction result by the prediction unit 20.

[0021] The information processing apparatus 1 may be capable of communicating with a plurality of user terminals 100 and an administrator terminal 150. The user terminal 100 is a terminal such as a personal computer, a smartphone, a tablet terminal, a wearable computer such as a smartwatch or smart glasses, and typically a portable terminal such as a smartphone or a tablet terminal. The user terminal 100 may include an operation unit 120 for the user to input and a display unit 110 for displaying information. In a smartphone or a tablet terminal, the operation display unit has the functions of the operation unit 120 and the display unit 110. The administrator terminal 150 is also, as an example, a terminal such as a personal computer, a smartphone, a tablet terminal, a wearable computer such as a smartwatch or smart glasses. The administrator terminal 150 may include an operation unit 170 for the administrator to input and a display unit 160 for displaying information. In a smartphone or a tablet terminal, the operation display unit has the functions of the operation unit 170 and the display unit 160. Inputs from the user and the administrator may be made by text input or by voice input. The content output by the output unit 90 can be viewed on the display unit 110 of the user terminal 100 or the display unit 160 of the administrator terminal 150. Further, the information input from the operation unit 120 of the user terminal 100 or the operation unit 170 of the administrator terminal 150 may be stored in the storage unit 80 in association with user identification information or administrator identification information.

[0022] "System" means the electricity market such as the spot market, the supply-demand adjustment market, and the pre-time market, and the pricing system in the imbalance price. According to this embodiment, the electricity prices in a plurality of systems can be estimated using information on operating power sources (this "operating power source" includes "power sources waiting for operation"), non-operating power sources, and the equipment utilization rate of operating power sources. It is beneficial to obtain information on operating power sources using a single power source operation model and use this information on operating power sources to predict the prices of electricity in a plurality of systems. Regarding the price for each system, a plurality of individual models are stored in the storage unit 80, and the electricity price in each system may be predicted by applying information on operating power sources to the individual models. Each individual model may be a model generated or corrected using machine learning based on past performance. In predicting the electricity price in each system, the start-stop characteristics and fuel cost characteristics of the power source are particularly important. These parameters may be appropriately updated using the actual performance information of each electricity price. When correcting each individual model using machine learning, for example, a model for each system may be generated in advance for each individual model based on electrical engineering, etc., and from the perspective of improving accuracy, an aspect of correcting each model by machine learning using past performance can be cited.

[0023] Existing algorithms such as dynamic programming and branch-and-bound methods can be used to determine the start-stop state of each power source using the power source operation model, and algorithms such as the equal incremental fuel cost method can be used to calculate the marginal cost based on the fuel cost characteristics (see Figure 2). The equal incremental fuel cost method is a method in which the fuel cost of each generator is expressed as a predetermined function of the generator output, and if each generator is operated so that the incremental fuel cost is equal in each time period, the generator output that minimizes the fuel cost for the entire generator group can be obtained.

[0024] The following can be used as input data when constructing the power source operation model and the individual models (see Figure 1). For the individual models, they may be generated or corrected by machine learning using past performance regarding the following information and stored in the storage unit 80. · Spot bidding · Transaction record: supply-demand curve, block bidding · Quantity agreed, occurrence status of market segmentation, etc. · Power source: Capacity, thermal efficiency, start-up cost, stop information (HJKS), etc. · Supply and demand: Demand by time zone, renewable energy · Nuclear power generation amount, pumped storage operation, etc. · Tie line: Power flow record, operating capacity, etc. · Fuel cost: Coal, LNG, petroleum, etc. · Former Kanden (former general electric utility): Company's own demand, company's own power source, demand · Power generation forecast, etc. · Imbalance: Price · Generation amount record, merit order of adjustment power kWh, etc.

[0025] In the market and system related to the kWh price (other than the imbalance price), it is considered that power generation companies will take actions to bid or register the surplus from the power generation plan at each point in time at the marginal cost. Therefore, based on the power source startup state calculated from the demand forecast such as demand assumption (see Figure 2), it is possible to calculate the marginal cost curve (such as the merit order of adjustment power kWh shown in Figures 4 and 6) using the equal incremental fuel cost method, etc., and to assume bids that incorporate startup costs. Regarding the demand forecast of electric power, a technique that is already known can be used.

[0026] The power source operation model may be a pre-prepared model, or it may be operated in order from the power source with the lowest price so as to fill the demand forecast of electric power (see Figure 2). When such a mode is adopted, among the multiple available power sources, the power source with the lowest price becomes the utilization state, and it is possible to grasp the power source with an equipment utilization rate of 〇〇% in operation or the non-operated power source, and the amount of electric power in the non-operated power source will be derived as the non-operated power amount. In addition, for power sources with similar prices, the model may be created so that they are operated evenly. The similar price means that, for example, the price difference is within a range of 5% of the price, and the administrator may appropriately define it via the administrator terminal 150. The power source operation model may be constructed based on the generator information registered in the power generation information disclosure system, the published materials of power generation companies, the power generation record by power source, etc. The bidding curve may be generated by inputting input data into the power source operation model.

[0027] The demand - supply adjustment market to be adjusted may include the ΔkW demand - supply adjustment market and the kWh demand - supply adjustment market. In the demand - supply adjustment market, There are products such as JPEG0007697093000002.jpg11160.

[0028] The electricity price in the first system is any one of the agreed price in the spot market, the agreed price in the ΔkW demand - supply adjustment market - the agreement probability (which means the agreed price and the agreement probability for each agreed price), the registered price in the kWh demand - supply adjustment market - the response probability (which means the registered price and the response probability for each registered price), the area - specific future agreed price in the pre - time market, and the imbalance price. The electricity price in the second system is any one of the agreed price in the spot market, the agreed price in the ΔkW demand - supply adjustment market - the agreement probability, the registered price in the kWh demand - supply adjustment market - the response probability, the area - specific future agreed price in the pre - time market, and the imbalance price, and it may be an electricity price of a system different from the first system.

[0029] The electricity price in the first system is the agreed price in the spot market or the agreed price - agreed probability in the ΔkW demand - supply adjustment market, and the electricity price in the second system may be the registered price - response probability in the kWh demand - supply adjustment market, the area - specific future agreed price in the pre - time market, or the imbalance price. Since the agreed price in the spot market and the agreed price in the ΔkW demand - supply adjustment market are agreed and determined by the day before the day when actual demand - supply is carried out, the prediction unit 20 predicts the power generation plan and the adjustable power that can be supplied of each power source from the agreed performance of the agreed price in the spot market included in the first system or the agreed price in the ΔkW demand - supply adjustment market, or predicts the power sources operating at the actual demand - supply time for the spot market and predicts the power sources waiting to operate for the ΔkW demand - supply adjustment market at the actual demand - supply time, and uses the results of the prediction to predict the electricity price in the second system such as the registered price - response probability in the kWh demand - supply adjustment market, the area - specific future agreed price in the pre - time market, or the imbalance price. When such a mode is adopted, since the electricity price in the second system is predicted using the actually determined information (agreed price, etc.), the electricity price can be predicted with higher accuracy. At this time, not only the agreed results of the spot market and the demand - supply adjustment market (ΔkW agreed price - agreed probability), but also related public information may be used so that the prediction unit 20 predicts the electricity price in the second system.

[0030] As in this embodiment, by using the results obtained using a commonly used power operation model and predicting the electricity prices of each system with an individual model, it is beneficial in that discrepancies between the results are reduced in each electricity price prediction and a consistent explanation becomes possible. More specifically, as a mechanism for forming an electricity market, there are various price indicators such as a spot market and a forward market where electricity is bought and sold to achieve the planned value simultaneously and in the same quantity, a supply-demand adjustment market where adjustment power is procured to compensate for the deviation from the planned value, and an imbalance price for clearing the deviation between the final planned value and the actual value. On the other hand, since the power operation plan, which is important in the formation of each electricity price, is sequentially updated in consideration of the constraints imposed by the power generation plan at that time at the bidding and contracting timings of each electricity market, all of the above electricity prices are correlated with each other. In this regard, by using a commonly used power operation model, discrepancies between the results in each electricity price prediction can be reduced and a consistent explanation becomes possible.

[0031] Note that on pages 7 and 11 of "Issues in the Electricity Market System from the Perspective of Thermal Power Generators" dated December 28, 2021 (https: / / www.meti.go.jp / shingikai / energy_environment / oroshi_jukyu / pdf / 001_09_00.pdf), an example of each market and its contracting timing, as well as the purpose and contracting method of each market, are shown, so please refer to that as well. Also, regarding the imbalance price, since it is also explained in "Regarding the Imbalance Tariff System, etc." dated January 28, 2022 (https: / / www.emsc.meti.go.jp / info / public / pdf / 20220117001b.pdf), please refer to that as well.

[0032] In the spot market and the ΔkW supply-demand adjustment market, it is necessary to simulate the power operation before the day before the actual supply-demand, which is determined by the startup and shutdown of power sources. Since the bidding volume in the spot market also increases or decreases to secure the ΔkW supplied to the supply-demand adjustment market, when simulating the operation of power generators aiming for the most economical operation under the market rules, the supply volume to each market can be calculated in consideration of such correlations. In the imbalance price, the supply-demand adjustment market (kWh price - instruction probability), and the pre-time market, it is important to assume the required adjustment power and the adjustment power kWh price merit order. By generating predictive values linked by a common power operation model, for example, if the imbalance prediction price is high in a certain prediction target period, the contractable price in the pre-time market will also be high and the probability of receiving an instruction in the supply-demand adjustment market will also be high. Thus, an operation plan that refers to multiple electricity prices based on reasonable prediction results becomes possible.

[0033] The prediction unit 20 may predict, using an operation prediction model, the power generation plan of each power source, the adjustable power that can be supplied, and the power sources operating at the actual supply and demand time and / or the power sources waiting to operate at the actual supply and demand time, from the actual transaction results of prices. Since the price for each power source is generally determined, it is beneficial to use the agreed price. However, predictions may be made using not only the actual transaction results of prices but also information such as the agreed quantity and the actual usage results of tie lines. As an example, the operation prediction model is a model that associates the actual transaction results of prices, the agreed quantity, the actual usage results of tie lines, etc. with the power sources presumed to be operating, and is stored in the storage unit 80. The operation prediction model may be a model generated using the agreed price and the power sources operating at the agreed price or waiting to operate. As an example, it may be a model generated using the agreed price in the target area and the power sources operating at the agreed price or waiting to operate in that area. Also, the operation prediction model may be a model generated or corrected by machine learning using the agreed price in the target area and the past performance regarding the power sources operating at the agreed price or waiting to operate. For the power sources operating or waiting to operate associated with the agreed price, published information may be used, or it may be information appropriately input by the administrator from the administrator terminal 150. The size of the area may be an area for each prefecture, or it may be divided into predetermined regions such as Hokkaido, Tohoku, Kanto, Chubu, Kinki, Chugoku, Shikoku, and Kyushu (including Okinawa), or it may be for each jurisdiction area of general power transmission and distribution operators as shown in the division in FIG. 5 (Hokkaido, Tohoku, Tokyo, Chubu, Hokuriku, Kansai, Chugoku, Shikoku, Kyushu). When correcting the operation prediction model using machine learning, for example, an operation prediction model is generated in a reasonable form based on electrical engineering, etc. using the agreed price and the power sources operating at the agreed price, and then correction by machine learning is performed using past performance from the perspective of improving accuracy.

[0034] Note that power sources using large turbines, etc. are generally operated continuously for a certain period of time, such as several hours or more or one day or more once they are started, and are not suddenly stopped. For this reason, from the contract performance of prices, the contracted quantity, the usage performance of tie lines, etc. in the spot market, the ΔkW supply-demand adjustment market, the kWh supply-demand adjustment market, and the forward market, at a certain point in time (the actual supply-demand point in time), when the prediction unit 20 determines (predicts) that a predetermined power source (continuous operation power source) using a large turbine, etc. is operating, on the premise that the continuous operation power source continues to operate (using the condition that it continues to operate at a predetermined time such as 4 hours, 12 hours, 1 day, 2 days from the start time of actual supply-demand for which the continuous operation power source is the target), it may be possible to predict the contract price in the spot market, the contract price - contract probability in the ΔkW supply-demand adjustment market, the registered price - response probability in the kWh supply-demand adjustment market, the area-specific future contract price in the forward market, and the imbalance price at another future point in time (the actual supply-demand point in time). By adopting such an aspect, the prediction accuracy of electricity prices can be further enhanced.

[0035] Speaking of a typical example of this embodiment, the prediction unit 20 predicts the contract price in the spot market by applying the information on the operating power source (see Figure 2) to the spot market model, predicts the contract price - contract probability in the ΔkW supply-demand adjustment market by applying the information on the operating power source to the ΔkW supply-demand adjustment market model, predicts the registered price - response probability in the kWh supply-demand adjustment market by applying the information on the operating power source to the kWh supply-demand adjustment market model, predicts the area-specific future contract price in the forward market by applying the information on the operating power source to the forward market model, and predicts the imbalance price by applying the information on the operating power source to the imbalance price model, and predicts the electricity prices in five different systems (see Figure 1). When such an aspect is adopted, it is possible to obtain information on the operating power source using a single power source operation model, and with this information on the operating power source, it is possible to predict the electricity prices in five different major systems with consistency, realizing extremely beneficial effects.

[0036] However, it is not limited to such a mode, and the electricity prices in two, three, or four systems may be predicted from these five systems. Each individual model of the spot market model, ΔkW demand and supply adjustment market model, kWh demand and supply adjustment market model, forward market model, and imbalance price model is stored in the storage unit 80.

[0037] In the demand and supply adjustment market, the electricity price is determined by a multi-price auction. In the spot market, the electricity price is determined by a single-price auction. In the forward market, the electricity price is determined by over-the-counter trading. That is, the electricity price is determined by different methods depending on the system. In order to make the prediction explanatory and calculate phenomena such as price hikes logically, a prediction model based on the trading rules of each market may be used in the individual model.

[0038] The spot market model may be generated using information on past operating power sources and data on electricity prices including past block bids and regular bids in the spot market, and stored in the storage unit 80. Similarly, the ΔkW demand and supply adjustment market model may be generated using information on past operating power sources and data on electricity prices in the past ΔkW demand and supply adjustment market, and stored in the storage unit 80. The kWh demand and supply adjustment market model may be generated using information on past operating power sources and data on electricity prices in the past kWh demand and supply adjustment market, and stored in the storage unit 80. The forward market model may be generated using information on past operating power sources and data on electricity prices in the past forward market, and stored in the storage unit 80. The imbalance price model may be generated using information on past operating power sources and data on past imbalance prices, and stored in the storage unit 80.

[0039] Machine learning may also be used. The spot market model may be generated by performing machine learning using learning data including information on past operating power sources and electricity prices including block bids and normal bids in the past spot market, and may be stored in the storage unit 80. Similarly, the ΔkW demand-supply adjustment market model may be generated by performing machine learning using learning data including information on past operating power sources and electricity prices in the past ΔkW demand-supply adjustment market, and may be stored in the storage unit 80. The kWh demand-supply adjustment market model may be generated by performing machine learning using learning data including information on past operating power sources and electricity prices in the past kWh demand-supply adjustment market, and may be stored in the storage unit 80. The time-ahead market model may be generated by performing machine learning using learning data including information on past operating power sources and electricity prices in the past time-ahead market, and may be stored in the storage unit 80. The imbalance price model may be generated by performing machine learning using learning data including information on past operating power sources and past imbalance prices, and may be stored in the storage unit 80.

[0040] In the spot market model, market rules that are price spike factors such as block contracts and market segmentation may be simulated (see Figure 1). The refined power source information based on actual results and the like can also be used for other predictions. In the imbalance price model, demand and renewable energy predictions provided by the former Ichiden (former general electric power companies) may be sequentially incorporated. The price may be calculated from the merit order of the adjustment power kWh based on market rules. The ΔkW demand-supply adjustment market model may be a model that predicts the contract probability with respect to the bid unit price. The wide-area procurement of the adjustment power considering the tie-line upper limit may be calculated to simulate the contract process. The kWh demand-supply adjustment market model may be a model that predicts the command probability with respect to the registered kWh unit price. The merit order of the adjustment power kWh assumed in the imbalance price prediction may be used. In the pre-market model, the movement of the transaction price (nationwide) may be separated by area using information such as bid price information. When adopting such a mode, it becomes possible to place bids considering price increases and decreases several hours later by predicting the price movement by area.

[0041] [Prediction of Spot Market Price] In the prediction of block bids in the spot market, the power operation model can be used to simulate the power startup and shutdown plans of major power sources, especially large power generation companies that are required to supply all of their surplus power, using dynamic programming, branch and bound methods, etc., to assume the block bid volume and price (see Figure 3 "Power Operation Model"). The block bid for a power source in a stopped state may include the startup cost, and the block bid with a step constraint (constraint related to the output change rate of the generator) may assume the bid price without the startup cost. Since the block bid is determined by the weighted average price of the bidding time period, a block bid equivalent to the daily average price of the spot market does not necessarily result in a contract, and as a result, the selling short volume may be insufficient and the price may increase. However, when adopting this mode, such a phenomenon can be simulated.

[0042] In the prediction of normal bids in the spot market, based on the calculation results of the startup and shutdown plan, a marginal cost curve of power sources with surplus supply capacity may be created using the equal incremental fuel cost method, etc., to assume the normal bid volume and price (see Figure 3 "Power Operation Model").

[0043] In the spot market model, in order to predict other bid information, it may be possible to predict buy bids, tie line capacity available in the spot market, zero-yen sell bid volume (renewable energy), etc. (see the "AI Model" in Figure 3). The AI model in Figure 3 is a model included in the spot market model and is generated by machine learning using past performance as learning data.

[0044] Using the assumptions related to the bidding information, after simulating block agreements and tie-line flexibility / market segmentation processing in accordance with market rules, supply and demand curves may be created for each segmented area, and the final area-specific price may be calculated based on the intersection points thereof (see the "Agreement Processing Model" in FIG. 3). When such an aspect is adopted, it becomes possible to simulate predictions such as price hikes due to sellouts caused by block agreements, market segmentation, etc. The agreement processing model may be a model generated or corrected by machine learning using the past block bidding volume and price, normal bidding volume and price, buy curve and zero-yen sell bids, and the operating capacity of tie-lines, as well as the past agreed prices in the spot market, as learning data.

[0045] The area-specific price may be a probabilistic prediction by probabilistically assuming the input bidding information. When such an aspect is adopted, the price elasticity of each can be simulated for the prediction assuming supply and demand curves, and a left-right asymmetric probability distribution reflecting the distribution of agreement results can be generated such that the occurrence probability gradually decays as the agreed price increases when the input bidding information is probabilistically assumed.

[0046] When aiming for further improvement in prediction accuracy, trend correction may be performed on the price prediction generated by AI or the like (spot market model) using various explanatory variables (see FIG. 3).

[0047] In the aspect shown in FIG. 3, by applying the input data to the power operation model and the spot market model, the block bidding volume and price, as well as the normal bidding volume and price, will be predicted. Also, by applying the input data to the AI model included in the spot market model, the buy curve and zero-yen sell bids, the operating capacity of tie-lines, and the price trend will be predicted. Then, by applying the predicted block bidding volume and price, normal bidding volume and price, buy curve and zero-yen sell bids, and the operating capacity of tie-lines to the agreement processing model, the agreed price in the spot market will be predicted.

[0048] In the aspect shown in FIG. 3, since the supply curve generated from the power operation model is refined using the relevant public information in the spot market, it can also contribute to improving the accuracy of the kWh price assumption in other electricity price predictions.

[0049] [Prediction of imbalance price] In the imbalance price model, based on the power demand prediction, the amount of imbalance generated in each area is predicted by AI or the like, and the adjustment power kWh merit order (see the dashed arrow in FIG. 4) is created based on the kWh price limit cost curve derived from the power operation model and the public information related to the adjustment power kWh (see the dotted arrow in FIG. 4). The limit adjustment power kWh price for the predicted amount of imbalance generated may be used as the predicted value. The predicted amount of imbalance generated in each area may be appropriately updated as the weather forecast and area supply and demand forecast are updated. When creating the adjustment power kWh price merit order from the power operation model or the like, the equal incremental fuel cost method may be used as in the spot market, or it may be appropriately corrected using information on supply and demand sequentially disclosed by general power transmission and distribution operators or the like. This also applies when predicting the agreed price in the ΔkW supply and demand adjustment market, the area-specific future agreed price in the time-ahead market, and the imbalance price. Under normal circumstances, the intersection point shifted by the predicted value of the amount of imbalance generated from the total in-area power generation plan (see the dotted line extending vertically in FIG. 4) may be used as the prediction result.

[0050] When adopting the mode of calculating the predicted value based on the calculation formula determined in the system design by decomposing it into the amount of imbalance generated and the adjustment power kWh price merit order, it is beneficial in that it can explain the cause, for example, whether the amount of imbalance generated was large when the price soared or whether the slope of the adjustment power kWh price merit order was large.

[0051] In case of supply-demand tightness, the correction tariff calculation index may be calculated from the calculation results of the power operation model for the demand prediction by area predicted by AI, or information on supply-demand successively disclosed by general power transmission and distribution operators, etc., and the predicted value may be calculated based on the "corrected imbalance tariff during supply-demand tightness" defined in the imbalance tariff system using the calculated value.

[0052] Regarding the time of suppressing solar and wind power output, the time of suppressing Power Source III, and the time of blackout occurrence, it may be calculated by the calculation formula defined in the imbalance tariff system from the supply-demand prediction and grid information updated successively.

[0053] Since it is uncertain which calculation method will be applied at the time of prediction, an ensemble prediction may be made using the probabilities of application of each calculation method. For example, if the probability of normal time is 50% and the probability of occurrence of output suppression is 50% at the time of prediction, the final predicted value may be obtained by taking the average of the predicted values for normal time and output suppression time. Also, similar to the spot market, trend correction may be performed based on price prediction by AI or the like.

[0054] Since the amount of imbalance occurrence is being predicted, when using tie-line information, the effect of imbalance netting (reduction of the amount of imbalance occurrence) by wide-area operation can be simulated.

[0055] The adjustment power kWh price merit order may be assumed by successively incorporating the spot market supply-demand curve information published at noon the previous day. When adopting such a mode, the recent power operation status can be reflected and the prediction accuracy and explainability can be improved.

[0056] [Prediction in the ΔkW Demand Adjustment Market] In the prediction in the ΔkW demand adjustment market, the following input data may be used (see Figure 5). · Past performance such as contract price, solicitation volume, bid volume, tie-line limit, etc. · Spot market contract results (uncontracted portion by area and its price range, etc.) · Forecast values related to renewable energy as of the day before yesterday Regarding past performance, it may be used as input information when generating or correcting the ΔkW demand - supply adjustment market model by machine learning.

[0057] Regarding the solicitation volume by area shown in Figure 5, it may be predicted by AI (specifically, the ΔkW demand - supply adjustment market model) based on the characteristics of each product from the information publicly disclosed by general power transmission and distribution operators, etc. For example If it is JPEG0007697093000003.jpg1254, for the adjustment power corresponding to the renewable energy prediction error, meteorological predictions such as solar radiation and wind speed can be used as explanatory variables. Regarding the bidding curve by area shown in Figure 5, based on the power operation model and the contract performance of each electricity market price, etc., the bidding curve in each area may be created based on the demand - supply adjustment market guidelines. The bidding curve appropriately created from the contract performance of each product may be corrected by AI (specifically, the ΔkW demand - supply adjustment market model). Regarding the tie - line upper limit shown in Figure 5, the available tie - line upper limit for each product may be predicted from the tie - line guaranteed quantity determined by the system design and past performance, etc. The tie - line upper limit may also be predicted by AI (specifically, the ΔkW demand - supply adjustment market model).

[0058] By simulating the optimization calculation of wide - area procurement such that the adjustment power cost is minimized considering the tie - line upper limit, the maximum contract price for each area (power source location) may be calculated (refer to the "table" shown at the bottom of Figure 5). Note that for primary (primary adjustment power) and JPEG0007697093000004.jpg1153, when wide - area procurement is difficult due to DC tie - lines, this calculation may be performed for each group of wide - area procurement - capable areas.

[0059] In the ΔkW demand - supply adjustment market, a multi - price auction is adopted. Therefore, it can be predicted that a bid will be concluded if it is below the highest price. The prediction using the ΔkW demand - supply adjustment market model may be carried out probabilistically (see the "graph" shown in the lower right of Figure 5), or the relationship between ΔkW and the conclusion probability may be quantitatively modeled by calculating the highest conclusion price for each area (power source jurisdiction) in a large number of scenarios. When such an approach is adopted, information useful for considering bidding strategies can be provided, such as the conclusion probability when bidding at a certain price, or conversely, what the bidding price should be set below when the conclusion probability is, for example, 80% or more. Also, the optimal bidding price in terms of the expected value can be automatically calculated by calculating "ΔkW bidding price × conclusion probability".

[0060] In the manner shown in Figure 5, since information related to bids in the demand - supply adjustment market is predicted respectively and the actual conclusion rules are simulated to calculate the ΔkW conclusion price, a model that can explain the reasons can be provided, such as whether the recruitment volume was large when the price soared, whether the number of bids was small, or whether efficient wide - area procurement was not achieved due to connection line restrictions.

[0061] In the demand - supply adjustment market, changes in the system design, such as the division of the bidding unit into 30 - minute intervals and the introduction of composite conclusions, are expected in the future. However, when adopting the approach of simulating and predicting the conclusion rules after assuming the recruitment volume and the number of bids for each product, it is beneficial in that it can be immediately responded to by changing the processing on the conclusion rule side even if the system is changed.

[0062] [Prediction in the kWh demand - supply adjustment market] The required adjustment power of the primary (primary adjustment power) and the tertiary (tertiary adjustment power) is the adjustment power for the 30 - minute average value of the prediction error and is in the area of EDC (Economic load Dispatching Control). Therefore, similar to the imbalance generation amount, appropriate prediction may be made by AI or the like as the weather prediction, area demand - supply prediction, etc. are updated. Primary (primary adjustment power) and The required adjustment power of JPEG0007697093000006.jpg1154 is the fine adjustment power within 30 minutes. Although it is difficult to make a prediction based on public information for the areas of LFC (Load Frequency Control) and GF (Governor Free), it is possible to assume, for example, after narrowing down a certain range from the required adjustment power of the above EDC.

[0063] Regarding the adjustment power kWh price merit order, while using a model assumed by the imbalance price model for the entire adjustment power kWh (refer to the dashed arrow of the "graph" on the left side of Figure 6), it may be possible to assume by commodity by aggregating the marginal costs of the power source groups corresponding to each adjustment power from the power operation model. When adopting a model similar to the imbalance price model for the prediction of the required adjustment power and the assumption of the adjustment power kWh price merit order, etc., it is beneficial in that the discrepancy in the prediction results between the imbalance price model and the kWh supply-demand adjustment market model can be eliminated.

[0064] Since the prediction of the required adjustment power for each commodity is accompanied by a certain degree of uncertainty, it can be a probabilistic prediction based on past performance of the prediction error, etc. It may also be possible to aggregate the probability that a command will come when a certain adjustment power kWh price is registered (refer to the "adjustment power generation probability" in Figure 6). In this case, the relationship between the adjustment power kWh price and the command probability may be modeled (refer to the "graph" on the right side of Figure 6). Especially in pumped-storage power generation and batteries with constraints on continuous output, the uncertainty of commands in the supply-demand adjustment market is a major issue in formulating operation plans. However, the fact that the future supply surplus and shortage can be predicted based on the command probability is beneficial in that it is possible to realize an operation that does not sacrifice business profitability and minimizes penalties by utilizing the time-ahead market.

[0065] Since operators participating in the adjustment power kWh market can grasp whether a command has come for the registered kWh price, these data may be used for learning by AI, etc. to correct the kWh supply-demand adjustment market model.

[0066] The prediction may be made in such a way that it is decomposed into the prediction of the necessary adjustment power and the adjustment power kWh price merit order, and the actual command by the general power transmission and distribution company is simulated. In this case, for example, when a command comes in despite registering a high kWh unit price, it is possible to create a model that can explain whether the necessary adjustment power was large or the slope of the adjustment power kWh merit order was large.

[0067] [Prediction in the pre-hour market] In the pre-hour market model, it is also possible to predict the future contract transition by area. As described above, the size of the area here may be an area for each prefecture, or it may be divided into predetermined regions such as Hokkaido, Tohoku, Kanto, Chubu, Kinki, Chugoku, Shikoku, and Kyushu (including Okinawa), or it may be for each jurisdiction area of the general power transmission and distribution company as shown in the division in Fig. 5 (Hokkaido, Tohoku, Tokyo, Chubu, Hokuriku, Kansai, Chugoku, Shikoku, Kyushu). Regarding the assumption of the contract performance by area, although participants in the pre-hour market can obtain the bid price information (bid volume and bid price), since the area information is unknown, due to market segmentation, for example, even if an ask price lower than the lowest price of the bid price comes in, it may not be contracted (see the "graph" in the upper left of Fig. 7). In this regard, in the pre-hour market model, by assuming the transition of the contract performance by area using the bid price information, the same-day spot market-related information, the tie-line information, etc. (see the "graph" in the lower left of Fig. 7), such a situation (a situation where it is not contracted) can be prevented from occurring. Considering that the pre-hour market is a market for achieving the planned value at the same time and the same quantity to reduce the imbalance, it is expected that there is a certain correlation between the prediction of the imbalance price using the imbalance price model and the prediction results using related information such as that. Therefore, while constructing an area-by-area contract price prediction model using AI or the like, by appropriately adjusting the parameters of the area-by-area contract performance assumption logic (by machine learning) so that the correlation (prediction accuracy) between the prediction result and the actual performance becomes high, it is also possible to generate reliable area-by-area contract performance data (the correct value in this prediction).

[0068] At present, the concluded performance by area has not been made public. However, due to the regional imbalance in power generation and demand, the impact of market segmentation in the pre-time market transactions is significant. Therefore, by making probable assumptions and predictions of the concluded performance by area based on the high correlation with information such as the prediction results of the imbalance price, important information for the pre-time market transactions can be provided.

[0069] Market participants can obtain the indicative price information and the concluded results each time. However, since the call market method is adopted in the pre-time market model, there is also a possibility of selling at a higher price and buying at a lower price in the future. By predicting the future changes in the concluded transactions by area, it is possible to lead to an improvement in profitability.

[0070] The acquisition unit 10, prediction unit 20, output unit 80, etc. shown in FIG. 8 may be realized by one unit (control unit), or may be realized by different units. The functions by a plurality of "units" may be integrated. Further, the acquisition unit 10, prediction unit 20, output unit 80, etc. may be realized by a circuit configuration. The information processing apparatus 1 has a processor, and various functions by the acquisition unit 10, prediction unit 20, output unit 80, etc. of the information processing apparatus 1 of the present embodiment may be realized when the processor executes a program.

[0071] The description of the above-described embodiment and the disclosure of the drawings are merely examples for explaining the invention described in the claims, and the invention described in the claims is not limited by the description of the above-described embodiment or the disclosure of the drawings.

Explanation of Reference Numerals

[0072] 10 Acquisition unit 20 Prediction unit 80 Storage unit 90 Output unit

Claims

1. an acquisition unit that acquires information about an operating power source, a non-operating power source, or an operating power source including a facility utilization rate of the operating power source by applying the power demand forecast to a power source operation model; a prediction unit that predicts an electricity price in a first system, which includes any one of an agreement price in the spot market, an agreement price in the ΔkW balancing market, a registered price in the kWh balancing market, a future agreement price by area in the time-ahead market, or an imbalance price, by applying information about the operating power source to a first individual model, and predicts an electricity price in a second system, which is different from the first system, which includes any one of an agreement price in the spot market, an agreement price in the ΔkW balancing market, a registered price in the kWh balancing market, a future agreement price by area in the time-ahead market, or an imbalance price, by applying information about the operating power source to a second individual model; An information processing device comprising:

2. An information processing device as described in claim 1, wherein the operating power sources included in the information regarding the operating power sources include power sources that are waiting to be operated.

3. The electricity price in the first system is a contract price in a spot market or a contract price in a ΔkW supply and demand adjustment market, The electricity price in the second system is a registered price in the kWh supply and demand adjustment market, a future contract price by area in the hourly market, or an imbalance price, The information processing device of claim 1, wherein the prediction unit predicts the power generation plan and available adjustment capacity of a power source from the actual price agreements in the spot market or the actual price agreements in the ΔkW supply and demand adjustment market, and predicts the registered price in the kWh supply and demand adjustment market, the future agreement price by area in the time-ahead market, or the imbalance price by applying information about the operating power source including the results of the prediction to a second individual model.

4. The electricity price in the first system is a contract price in the spot market and a contract price in the ΔkW supply and demand adjustment market, The electricity price in the second system is a registered price in the kWh supply and demand adjustment market, a future contract price by area in the hourly market, or an imbalance price, 2. The information processing device of claim 1, wherein the prediction unit predicts power sources operating at the time of actual supply and demand in the spot market based on the price agreement history in the spot market and the price agreement history in the ΔkW supply and demand adjustment market, and predicts power sources waiting to operate at the time of actual supply and demand in the ΔkW supply and demand adjustment market, and predicts registered prices in the kWh supply and demand adjustment market, future agreement prices by area in the time-ahead market, by applying information about the operating power sources including the results of the prediction to a second individual model.

5. The information processing device according to any one of claims 1 to 4, wherein the prediction unit uses a model that associates price contract history with power sources that are assumed to be in operation to determine whether a continuously operating power source that operates continuously for a specified period of time is in operation from the price contract history, and if it determines that the continuously operating power source is in operation, predicts the electricity price in the first system or the second system on the assumption that the continuously operating power source will be operated for the specified period of time.

6. An acquisition unit that acquires information about an operating power source, a non-operating power source, or an operating power source including a facility utilization rate of the operating power source by applying the demand forecast to a power source operation model; a prediction unit that predicts electricity prices in a spot market by applying information about the operating power source to a spot market model, predicts an agreement price in the ΔkW supply and demand adjustment market by applying information about the operating power source to a ΔkW supply and demand adjustment market model, predicts a registered price in the kWh supply and demand adjustment market by applying information about the operating power source to a kWh supply and demand adjustment market model, predicts future agreement prices by area in the time-ahead market by applying information about the operating power source to an hour-ahead market model, and predicts an imbalance price by applying information about the operating power source to an imbalance price model; An information processing device comprising:

7. The information processing device of claim 6, wherein the prediction unit predicts the power generation plan and available adjustment capacity of a power source from the actual price agreements in the spot market or the actual price agreements in the ΔkW supply and demand adjustment market, predicts the registered price in the kWh supply and demand adjustment market by applying information about the operating power source including the results of the prediction to a kWh supply and demand adjustment market model, predicts future agreement prices by area in the hour-ahead market by applying information about the operating power source including the results of the prediction to an hour-ahead market model, or predicts the imbalance price by applying information about the operating power source including the results of the prediction to an imbalance price model.

8. 7. The information processing device of claim 6, wherein the prediction unit predicts power sources operating at the time of actual supply and demand based on price agreement records in the spot market, and predicts power sources waiting to operate at the time of actual supply and demand based on price agreement records in the ΔkW supply and demand adjustment market, predicts registered prices in the kWh supply and demand adjustment market by applying information about operating power sources including the results of the prediction to a kWh supply and demand adjustment market model, predicts future agreement prices by area in the time-ahead market by applying information about operating power sources including the results of the prediction to an imbalance price model, or predicts imbalance prices by applying information about operating power sources including the results of the prediction to an imbalance price model.

9. 9. The information processing device according to claim 6, wherein the prediction unit uses a model that associates price agreement records with power sources that are assumed to be in operation to determine whether a continuously operating power source that operates continuously for a predetermined period of time is in operation from the price agreement records, and if it determines that the continuously operating power source is in operation, predicts the agreement price in the spot market, the agreement price in the ΔkW supply and demand adjustment market, the registered price in the kWh supply and demand adjustment market, and the future agreement price and imbalance price by area in the time-ahead market, on the assumption that the continuously operating power source will be operated for the predetermined period of time.

10. acquiring information about an operating power source, a non-operating power source, or an operating power source including a facility utilization rate of the operating power source by applying the demand forecast to a power source operation model by an acquisition unit; A step of predicting, by a prediction unit, an electricity price in a first system including a contract price in a spot market, a contract price in a ΔkW supply-demand balancing market, a registered price in a kWh supply-demand balancing market, a future contract price by area in a time-ahead market, or an imbalance price by applying information about the operating power source to a first individual model; A step of applying information about the operating power source to a second individual model by a prediction unit to predict an electricity price in a second system different from the first system, the second system including a contract price in a spot market, a contract price in a ΔkW supply and demand adjustment market, a registered price in a kWh supply and demand adjustment market, a future contract price by area in a time-ahead market, or an imbalance price; An information processing method comprising:

11. A program to be installed on an information processing device, In the information processing device in which the program is installed, A function for obtaining information about operational resources, including operational resources, non-operational resources, or utilization rates of operational resources by applying the demand forecast to a resource operation model; a function of predicting electricity prices in a first system, including any one of the contract price in the spot market, the contract price in the ΔkW supply and demand adjustment market, the registered price in the kWh supply and demand adjustment market, the future contract price by area in the time-ahead market, or the imbalance price by applying information about the operating power source to a first individual model, and predicting electricity prices in a second system, which is different from the first system, including any one of the contract price in the spot market, the contract price in the ΔkW supply and demand adjustment market, the registered price in the kWh supply and demand adjustment market, the future contract price by area in the time-ahead market, or the imbalance price by applying information about the operating power source to a second individual model; A program to achieve this.

Images