Information processing device, control method for information processing device, and program

Abstract

This system outputs support information for conducting electricity trading in an electricity trading market where the winning bid price is determined by a multi-price auction system. [Solution] An information processing device that outputs support information for conducting electricity trading in an electricity trading market in which the winning bid price is determined by a multi-price auction method, comprising: a prediction scenario generation unit that generates a prediction scenario that predicts the highest winning bid price for each trading time frame on a trading day using historical information including past winning bid prices in the electricity trading market; a bidding scenario acquisition unit that acquires a bidding scenario that specifies the bidding price and trading volume for a bidding time frame which is a trading time frame in which bidding is conducted; and a support information generation unit that generates evaluation information of the bidding scenario, including information on the probability of winning a bid and the expected amount of revenue in the bidding time frame, as support information, based on the prediction scenario and the bidding scenario.

Description

Technical Field

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

Background Art

[0002] In recent years, markets for conducting power transactions have been developed, and power transactions are being carried out in various markets according to the purpose and application.

[0003] And various technologies for supporting power transactions in such power trading markets have also been developed (see, for example, Patent Document 1).

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0005] In recent years, it has also become commercialized to seek profit by selling the power stored in the grid-connected battery in the power trading market.

[0006] In this case, it is important how to purchase power at a low price and sell it at a high price. However, in the case of a power trading market that adopts a multi-price auction method such as the demand-supply adjustment market, if you bid at a low price, you will conclude (win the bid) at that price, and there is a high possibility that you will only obtain a small profit. On the other hand, if you bid at a high price, although you can obtain a greater profit if you win the bid, the possibility of not winning the bid is high. Therefore, it is difficult to judge the bidding price.

[0007] Therefore, there is a need for a technology that can provide support information for use when conducting power transactions in a power trading market where the winning bid price is determined by such a multi-price auction method.

[0008] This invention has been made in view of the above problems, and aims to provide an information processing device, a control method for the information processing device, and a program that can output support information for conducting electricity trading in an electricity trading market where the winning bid price is determined by a multi-price auction method. [Means for solving the problem]

[0009] An information processing device according to one embodiment of the present invention is an information processing device that outputs support information for conducting electricity trading in an electricity trading market in which the successful bid price is determined by a multi-price auction method, and comprises: a prediction scenario generation unit that generates a prediction scenario that predicts the highest successful bid price for each trading time frame on a trading day using historical information including past successful bid prices in the electricity trading market; a bidding scenario acquisition unit that acquires a bidding scenario that specifies the bidding price and trading volume for a bidding time frame which is a trading time frame in which bidding is conducted; and a support information generation unit that generates evaluation information of the bidding scenario, including information on the possibility of successful bids and the estimated amount of revenue in the bidding time frame, as support information, based on the prediction scenario and the bidding scenario.

[0010] Furthermore, the problems disclosed in this application and their solutions will be made clear from the description in the section on embodiments for carrying out the invention and from the drawings. [Effects of the Invention]

[0011] This will enable the output of support information for conducting electricity trading in the electricity trading market, where the winning bid price is determined by a multi-price auction system. [Brief explanation of the drawing]

[0012] [Figure 1] This is an overall system diagram including the information processing system. [Figure 2] This diagram shows the hardware configuration of an information processing device. [Figure 3] This is a diagram showing the memory device of an information processing device. [Figure 4] This is a diagram showing an initial data table. [Figure 5] This is a diagram showing an equipment data table. [Figure 6] This is a diagram showing the functional configuration of an information processing device. [Figure 7] This is a diagram showing a highest bid price scenario. [Figure 8] This is a diagram showing a bid price scenario. [Figure 9] This is a diagram showing a bidding scenario. [Figure 10] This is a diagram showing an example of support information. [Figure 11] This is a diagram showing an example of an input screen of an input scenario. [Figure 12] This is a diagram showing an example of support information. [Figure 13] This is a diagram showing a flowchart showing the process flow of an information processing device.

Mode for Carrying Out the Invention

[0013] From the descriptions in this specification and the attached drawings, at least the following matters become clear. Hereinafter, the present invention will be described based on one embodiment thereof with reference to the attached drawings.

[0014] ==Overall Configuration== FIG. 1 shows an example of the overall configuration of an information processing system 1000 according to one embodiment of the present invention.

[0015] The information processing system 1000 includes a storage battery 200 and an information processing device 100.

[0016] The storage battery 200 is a system storage battery that can be connected to a power system (not shown), and can be charged with power from the power system and discharged to the power system. Charging and discharging of the storage battery 200 can be autonomously performed by the storage battery 200 according to a pre-created charge-discharge plan, or can be performed according to a command from the information processing device 100 or other computers (not shown). Further, the storage battery 200 may be charged not only from the power system but also with power generated by a renewable energy power generation device such as a solar power generation device or a wind power generation device (not shown).

[0017] The information processing device 100 is a computer used by an operator who conducts a power transaction in a power trading market where the winning price (contract price) is determined by the multi-price auction method.

[0018] The multi-price auction method is a method in which winning (contracting) is performed at each bid price in ascending order of the bid price until a pre-agreed quantity of transactions (for example, in the case of the supply-demand adjustment market, the quantity of power regulation capacity solicited) is reached. Therefore, when selling power in a power trading market that adopts the multi-price auction method, it is desirable to be able to win at the highest possible winning price. However, if the bid price exceeds the highest winning price, the bid cannot be won and no profit can be obtained (however, this is based on the premise that the market is active (that is, in the case of "quantity of bids > quantity solicited"). In the case of "quantity of bids < quantity solicited", any bid price below the upper limit price will result in winning the bid.). Further, the highest winning price fluctuates depending on factors such as the quantity of transactions (quantity solicited) at that time, the bidding situation of other bidders (the number of bidders, the bid price and quantity of each bidder), and meteorological conditions such as weather and temperature.

[0019] As an example, the information processing device 100 outputs support information for use by an operator who intends to sell (sell power) the power stored in the storage battery 200 as regulation capacity in the supply-demand adjustment market.

[0020] The supply-demand adjustment market treats various regulation capacities as commodities. In this embodiment, as an example, the tertiary regulation capacity (2) is the subject of the transaction.

[0021] As will be explained in more detail later, the support information is, for example, evaluation information for the bidding scenario, which is created based on a "prediction scenario" (also referred to as the highest successful bid price scenario) that predicts the highest successful bid price for each trading time slot on a day when electricity is traded (for example, the next day) (for example, in the case of tertiary adjustment capacity (2), 48 trading time slots (frames) divided into 30-minute intervals from 0:00 to 24:00), and a "bidding scenario" that identifies the bid price and trading volume for the bidding time slot (also referred to as the bidding frame) which is the trading time slot (frame) in which the operator plans to bid.

[0022] An example of a "prediction scenario" is shown in Figure 7. Figure 7 shows three prediction scenarios, B1 to B3 (these prediction scenarios have different predicted highest bid prices for each bid slot). An example of a "bidding scenario" is shown in Figure 9. Figure 9 shows three bidding scenarios, A1 to A3 (these bidding scenarios have different bid prices, bid slots, and transaction volumes for each bid slot).

[0023] The information processing device 100 then generates evaluation information that includes information on the probability of winning the bid and the estimated revenue for each bidding time slot (bid frame). This allows the business operator to select the optimal bidding scenario or revise the bidding scenario (for example, changing to a bidding scenario that is expected to generate higher revenue, changing to a lower-risk bidding scenario that is more likely to win the bid, or canceling the bid).

[0024] ==Information Processing Unit== Next, the information processing device 100 will be described with reference to Figure 2.

[0025] As shown in Figure 2, the information processing device 100 is configured to include a CPU 110, memory 120, communication device 130, storage device 140, input device 150, output device 160, and recording medium reading device 170.

[0026] The CPU 110 is an arithmetic unit that controls the entire information processing device 100. It reads the information processing device control program 700, which consists of codes for performing various operations according to this embodiment and various data stored in the storage device 140, into the memory 120 and executes or processes them to realize various functions of the information processing device 100.

[0027] For example, the CPU 110 executes or processes the information processing device control program 700 and various data, and in cooperation with hardware devices such as the memory 120, communication device 130, and storage device 140, functions such as the prediction scenario generation unit 101, the bidding scenario acquisition unit 102, and the support information generation unit 103, which will be described later, are realized.

[0028] The storage device 140 is a non-volatile, non-temporary storage device such as a hard disk drive or an SSD (Solid State Drive). As shown in Figure 3, the storage device 140 stores various types of data, including the information processing device control program 700 executed by the information processing device 100, the initial data table 600, the prediction model 610, and the equipment data table 620 (details will be described later).

[0029] The recording medium reader 170 reads the information processing device control program 700 and various data recorded on the recording medium 800, such as a CD-ROM, DVD, or USB® memory, and stores them in the storage device 140.

[0030] The communication device 130 exchanges various data and programs with other computers (not shown) via a network 500 such as the Internet or a telephone network. For example, the information processing device control program 700 and various tables mentioned above can be stored on another computer (not shown), and the information processing device 100 can download this data and the information processing device control program 700 from this computer. Alternatively, the various tables mentioned above may be constructed on another computer.

[0031] Furthermore, the communication device 130 also functions as an interface for receiving various input data from computer terminals (not shown) such as personal computers, smartphones, and tablets operated by users, and for transmitting support information generated by the information processing device 100 to these computer terminals.

[0032] The input device 150 is an interface for inputting data, such as from a keyboard or mouse. The output device 160 is an interface for outputting data, such as from a display or printer.

[0033] Furthermore, the information processing device 100 may be implemented by a single computer, or by multiple computers connected to each other in a manner that enables communication. In addition, the information processing device 100 may be a computer built using software such as a virtual machine or a cloud computer.

[0034] Next, the initial data table 600, the prediction model 610, and the equipment data table 620 will be explained with reference to Figures 4 and 5, etc.

[0035] The initial data table 600 is a table that stores historical information (for example, past successful bid prices and bid prices in the supply and demand adjustment market, the amount of electricity (adjustment capacity) to be requested, the successful bid price in the spot market, the amount of electricity demanded and supplied, weather information, etc.) that the information processing device 100 uses to generate "prediction scenarios" and "bidding scenarios". As shown in Figure 4, the initial data table 600 has columns for "Date", "Time", "Successful bid price in the supply and demand adjustment market", "Bid price in the supply and demand adjustment market", "Amount to be requested", "Successful bid price in the spot market", "Supply and demand", and "Weather information".

[0036] The "Date" column records the date on which each piece of historical information was obtained.

[0037] The "Time Slot" column records numbers that identify the 48 trading time slots (time slots) that divide the day into 30-minute intervals. For example, "0" indicates the time slot from 0:00 AM to 0:30 AM, and "1" indicates the time slot from 0:30 AM to 1:00 AM.

[0038] The "Winning Bid Price in the Supply and Demand Adjustment Market" column records the past winning bid prices for each time slot in the supply and demand adjustment market. Because the supply and demand adjustment market employs a multi-price auction system, multiple winning bid prices exist within the same time slot. Therefore, the "Winning Bid Price" column records these multiple winning bid prices, as well as the highest winning bid price, lowest winning bid price, and average winning bid price calculated from these winning bid prices. The information processing device 100 obtains the daily winning bid prices for each time slot from a computer operated by the power supply and demand adjustment exchange and calculates the highest winning bid price, lowest winning bid price, and average winning bid price.

[0039] The "Bid Price in the Supply and Demand Adjustment Market" column records the past bid prices for each time slot in the supply and demand adjustment market. Although multiple bid prices exist within the same time slot, the information processing device 100 retrieves all daily bid prices for each time slot from the computer operated by the Electric Power Supply and Demand Adjustment Power Exchange.

[0040] The "Amount Requested" column records the past amount of adjustment capacity requested in each time slot of the supply and demand adjustment market. The information processing device 100 obtains the daily amount of requested capacity for each time slot from a computer operated by the Electricity Supply and Demand Adjustment Capacity Exchange.

[0041] The "Spot Market Winning Bid Price" column records the past winning bid prices for each time slot in the spot market (the day-ahead market). Since the spot market employs a single-price auction system, there is only one winning bid price for each time slot. The information processing device 100 obtains the daily winning bid prices for each time slot from a computer operated by the Japan Electric Power Exchange.

[0042] The "Supply and Demand" column records the power demand and power supply amounts for each section within the area where the battery storage 200 is installed. The power demand and power supply amounts are published by the general power transmission and distribution company that has jurisdiction over the area. The information processing device 100 obtains the daily power demand and power supply amounts for each section from a computer operated by the general power transmission and distribution company.

[0043] The "Weather Information" section records weather data for the area within each frame. In the example shown in Figure 4, temperature, solar radiation, wind speed, and weather conditions (sunny, rainy, cloudy, snowy, etc.) are recorded. The information processing device 100 acquires this weather data from computers operated by the Japan Meteorological Agency or businesses providing weather data services. Alternatively, weather data measuring devices may be installed at the location where the battery 200 is installed, and the information processing device 100 may acquire weather data from these measuring devices.

[0044] Furthermore, in addition to what is shown in Figure 4, the initial data table 600 may also record, for example, the past winning bid prices for each time slot in the pre-market (today's market) as historical information. In the pre-market, the winning bid price (transaction price) is determined using a continuous trading method, so there are multiple winning bid prices within the same time slot. The information processing device 100 obtains the daily winning bid prices for each time slot from a computer operated by the Japan Electric Power Exchange.

[0045] The information processing device 100 then uses this historical information to generate a prediction model 610 and stores it in the storage device 140. For example, the information processing device 100 uses the correlation between each data point of historical information over a predetermined period in the past (approximately one to three months) to generate a prediction model 610 for calculating the highest successful bid price in the supply and demand adjustment market. Similarly, the information processing device 100 generates various prediction models 610, such as a prediction model 610 for calculating the lowest successful bid price in the supply and demand adjustment market, and a prediction model 610 for calculating the average successful bid price in the supply and demand adjustment market.

[0046] In this embodiment, the information processing device 100 generates a first prediction model 610A that calculates a predicted value for the highest successful bid price in the supply and demand adjustment market from the temperature, a second prediction model 610B that calculates a predicted value for the lowest successful bid price in the supply and demand adjustment market from the temperature, a third prediction model 610C that calculates a predicted value for the average successful bid price in the supply and demand adjustment market from the temperature, and a fourth prediction model 610D that calculates a predicted value for the successful bid price in the spot market from the temperature.

[0047] The equipment data table 620 is a table that stores the specifications and performance of the equipment owned by the information processing system 1000. In the example shown in Figure 5, the specifications and performance of the battery 200 are recorded. When the information processing device 100 creates a trading plan, as described later, it takes into account the specifications and performance of the equipment recorded in the equipment data table 620. If the information processing system 1000 is equipped with renewable energy power generation equipment such as solar power generation equipment or wind power generation equipment, and the battery 200 is operated to charge with electricity generated by these renewable energy power generation equipment and trade electricity in the supply and demand adjustment market or spot market, then the specifications and performance of these pieces of equipment are also recorded in the equipment data table 620.

[0048] ==Functional Configuration of Information Processing Device== Next, the functional configuration of the information processing device 100 will be explained with reference to the functional configuration diagram shown in Figure 6.

[0049] As described above, the information processing device 100 realizes its various functions when the information processing device control program 700 and various data are read into the memory 120 and executed or processed by the CPU 110.

[0050] Specifically, the information processing device 100 implements various functions such as a prediction scenario generation unit 101, a bidding scenario acquisition unit 102, and a support information generation unit 103.

[0051] <Prediction Scenario Generation Unit> The prediction scenario generation unit 101 uses historical information, including past successful bid prices, in the electricity trading market where the successful bid price is determined by a multi-price auction method, to generate a prediction scenario (highest successful bid price scenario) that predicts the highest successful bid price for each trading time slot on the trading day.

[0052] For example, the prediction scenario generation unit 101 creates a highest bid price scenario (prediction scenario) that predicts the highest bid price for each time slot (time slots 0 to 47) of the trading day (e.g., the next day; the same applies hereafter) as follows. In the following example, for the sake of simplicity, we will explain how to calculate the highest bid price scenario based on the predicted temperature for each time slot of the trading day. However, in addition to the predicted temperature, the highest bid price scenario may also be calculated using the amount offered, electricity demand, electricity supply, weather, or a combination of these.

[0053] The prediction scenario generation unit 101 first obtains the predicted temperature (Ti) for frame i (where i is an integer from 0 to 47) on the trading day. Then, the prediction scenario generation unit 101 refers to the "temperature" column and "highest successful bid price" column for frame i in the initial data table 600 for a predetermined past period (for example, the past 3 months), and extracts the highest successful bid price when the difference between the temperature at that time and the predicted temperature (Ti) was within a predetermined range (for example, within ±1℃). In this way, multiple past highest successful bid prices for frame i are extracted.

[0054] Next, the prediction scenario generation unit 101 calculates the mean and standard deviation of these multiple past highest bid prices by statistically processing them. For example, the mean of the highest bid prices can be calculated as mi, and the standard deviation as σi. Then, the prediction scenario generation unit 101 sets the z-value (z-score) to a predetermined value K (for example, K=1) and calculates the predicted value of the highest bid price for frame i on the trading day using mi + K × σi.

[0055] The prediction scenario generation unit 101 performs the above processing for all frames (i=0 to 47) to generate a highest bid price scenario consisting of predicted values ​​for each frame where the value of K is the same.

[0056] Furthermore, the prediction scenario generation unit 101 may vary the z-value in various ways (for example, K=0, ±0.5, ±1, ±1.5, ±2, ±2.5, ±3) to calculate the predicted value of the highest bid price for frame i using mi + K × σi. The prediction scenario generation unit 101 can then group the predicted values ​​for each frame with the same K value and combine them into a single highest bid price scenario, thereby generating multiple highest bid price scenarios. For example, Figure 7 shows an example where three highest bid price scenarios have been created. The number of highest bid price scenarios created by the prediction scenario generation unit 101 is not particularly limited, but the more scenarios created, the more detailed the evaluation information that the support information generation unit 103 (described later) can generate when generating evaluation information can be.

[0057] <Bidding Scenario Acquisition Section> Next, the bidding scenario acquisition unit 102 acquires a bidding scenario that specifies the bid price and transaction volume for the bidding time slot (bid frame) on the trading day. The bidding scenario acquisition unit 102 may acquire a bidding scenario by accepting input from the user of the bid frame, transaction volume, and bid price in which the user intends to bid, through a user interface as shown in Figure 11 (if custom input is selected in Figure 11), or, if automatic creation is selected in Figure 11, it may create a bidding scenario using historical information as described below. An example of a bidding scenario is shown in Figure 9.

[0058] When automatically creating a bidding scenario, the bidding scenario acquisition unit 102 first creates a "bidding price scenario" consisting of the bid prices for each time slot (time slot 0 to time slot 47) on the trading day (the next day) (see Figure 8), and then creates a "bidding scenario" by determining the trading volume for each bidding time slot (see Figure 9).

[0059] First, let's explain how to create a "bid price scenario."

[0060] For the sake of simplicity, the following example explains how to calculate a bidding price scenario based on the predicted temperature for each time slot on the trading day. However, in addition to predicted temperature, you may also calculate the bidding price scenario using the amount to be solicited, electricity demand, supply, weather, or a combination of these.

[0061] The bidding scenario acquisition unit 102 first acquires the predicted temperature (Ti) for time slot i on the trading day. Then, the bidding scenario acquisition unit 102 refers to the "temperature" column and the "bid price in the supply and demand adjustment market" column for time slot i over a predetermined past period (e.g., 3 months) in the initial data table 600, and extracts the bid price when the difference between the temperature at that time and the predicted temperature (Ti) was within a predetermined range (e.g., within ±1℃). In this way, multiple past bid prices for time slot i are extracted.

[0062] Next, the bidding scenario acquisition unit 102 calculates the mean and standard deviation by statistically processing these multiple past bid prices. For example, the mean of these bid prices can be calculated as mi, and the standard deviation as σi. Then, the bidding scenario acquisition unit 102 sets the z value to a predetermined value K (for example, K=1) and calculates the bid price for time slot i on the trading day using mi + K × σi.

[0063] The bidding scenario acquisition unit 102 performs the above processing for all frames (i=0 to 47) to calculate the bidding price scenario for each frame with the same value of K.

[0064] Alternatively, the bidding scenario acquisition unit 102 may create an "input price scenario" by the following method.

[0065] First, the bidding scenario acquisition unit 102 acquires the predicted temperature (Ti) for frame i on the trading day. Then, the bidding scenario acquisition unit 102 inputs the predicted temperature (Ti) for frame i into the first prediction model 610A described above to calculate the predicted value of the highest successful bid price for frame i.

[0066] The bidding scenario acquisition unit 102 then uses the predicted highest bid price calculated for all frames (i=0 to 47) as a single bidding price scenario.

[0067] Next, we will explain how the bidding scenario acquisition unit 102 creates a "bidding scenario." The bidding scenario acquisition unit 102 creates an "input scenario" that identifies the bidding slot, bidding price, and trading volume in the supply and demand adjustment market by creating a power trading plan (bidding slot, trading volume, bidding price) that will yield greater profits when selling electricity purchased in the spot market as adjustment power in the supply and demand adjustment market.

[0068] This trading plan can be formulated as a mixed-integer linear programming problem for profit maximization and can be solved using methods such as the branch and bound method, optimization solvers, and metaheuristics. For example, the bidding scenario acquisition unit 102 creates a trading plan by calculating the operation of the battery 200 in each time slot (charging, discharging, doing nothing) and the trading volume in each time slot that maximizes the profit from electricity trading on the trading day, while considering the specifications and performance of the battery 200 recorded in the equipment data table 620, as well as the trading rules (minimum trading volume, etc.) in the spot market and supply and demand adjustment market as constraints. In this case, the amount of profit can be calculated using the expected value of the selling price, which can be calculated by multiplying the distribution (probability distribution) of the predicted highest successful bid price for each time slot on the trading day, the bid price, and the probability of successful bid obtained from this distribution.

[0069] Furthermore, when calculating the profitability of the trading plan, the purchase price of electricity in the spot market can be the predicted value obtained by inputting the predicted temperature (Ti) for each time slot of the trading day into the fourth forecast model 610D.

[0070] Furthermore, the selling price of electricity (adjustment capacity) in the supply and demand adjustment market can be calculated using the bidding price scenario created above.

[0071] Furthermore, when creating a trading plan, instead of setting the selling price to 0 yen if a trade cannot be executed in the supply and demand adjustment market, the expected winning bid price when selling electricity in the pre-hour market (same-day market) may be used as the selling price. In that case, the selling price may be calculated as an expected value by summing the product of the probability of execution in the supply and demand adjustment market and the expected execution price (bid price), and the product of the probability of not execution in the supply and demand adjustment market and the expected selling price in the pre-hour market. In this case, the information processing device 100 may calculate the expected execution price in the pre-hour market using historical information of past execution prices in the pre-hour market and a prediction model 610 that predicts this execution price, similar to the prediction of the highest winning bid price, etc., as described above. Furthermore, the bidding scenario acquisition unit 102 creates a trading plan considering that when conducting electricity sales transactions in the pre-hour market, it is necessary to trade in a time slot that has not been closed (a time slot more than one hour in the future).

[0072] As described above, the bidding scenario acquisition unit 102 creates a "bidding scenario" that specifies the bid price and transaction volume for the bidding time frame, which is the transaction time frame in which the bids are made.

[0073] The bidding scenario acquisition unit 102 can also create multiple "bidding scenarios" in which at least one of the bidding time frame, bid price, and transaction volume is different.

[0074] In the example user input screen shown in Figure 11, if the user inputs multiple bidding scenarios, the bidding scenario acquisition unit 102 can acquire the multiple bidding scenarios created by the user by providing multiple input fields for the bidding scenarios.

[0075] On the other hand, if automatic creation of bidding scenarios is selected in Figure 11, the bidding scenario acquisition unit 102 first creates multiple "bid price scenarios," and then creates a transaction plan for each input price scenario, thereby creating multiple "bidding scenarios."

[0076] To explain in more detail, first the bidding scenario acquisition unit 102 acquires the predicted temperature (Ti) for time slot i on the trading day. Then, the bidding scenario acquisition unit 102 refers to the "temperature" column and the "bid price in the supply and demand adjustment market" column for time slot i over a predetermined past period (e.g., 3 months) in the initial data table 600, and extracts the bid price when the difference between the temperature at that time and the predicted temperature (Ti) was within a predetermined range (e.g., within ±1℃). In this way, multiple past bid prices for time slot i are extracted.

[0077] Next, the bidding scenario acquisition unit 102 calculates the mean and standard deviation by statistically processing these multiple past bid prices. For example, the mean of these past bid prices can be calculated as mi, and the standard deviation as σi. The bidding scenario acquisition unit 102 then varies the z value in various ways (for example, predetermined values ​​K = 0, ±1, ±2, ±3) and calculates the bid price for time slot i on the trading day using mi + K × σi.

[0078] The bidding scenario acquisition unit 102 performs the above processing for all frames (i=0 to 47), then groups the bid prices of each frame with the same K value and combines them into a single bid price scenario to calculate multiple bid price scenarios. Figure 8 shows an example of when three bid price scenarios a1 to a3 are generated.

[0079] Alternatively, the bidding scenario acquisition unit 102 may create multiple input price scenarios using the following method.

[0080] First, the bidding scenario acquisition unit 102 acquires the predicted temperature (Ti) for frame i on the trading day. Then, the bidding scenario acquisition unit 102 inputs the predicted temperature (Ti) for frame i into the first prediction model 610A described above to calculate the predicted highest bid price for frame i. The bidding scenario acquisition unit 102 then combines the predicted highest bid prices calculated for all frames (i=0 to 47) into a single bidding price scenario (first bidding price scenario).

[0081] Similarly, the bidding scenario acquisition unit 102 inputs the predicted temperature (Ti) for frame i into the second prediction model 610B described above to calculate the predicted minimum bid price for frame i. The bidding scenario acquisition unit 102 then combines the predicted minimum bid prices calculated for all frames (i=0 to 47) to create a single bidding price scenario (second bidding price scenario).

[0082] Furthermore, the bidding scenario acquisition unit 102 inputs the predicted temperature (Ti) of frame i into the third prediction model 610C described above to calculate the predicted average winning bid price for frame i. The bidding scenario acquisition unit 102 then combines the predicted average winning bid prices calculated for all frames (i=0 to 47) to create a single bidding price scenario (third bidding price scenario).

[0083] Once multiple "bid price scenarios" have been created as described above, the bid scenario acquisition unit 102 creates a transaction plan for each input price scenario that maximizes the revenue from electricity trading.

[0084] The bidding scenario acquisition unit 102 then creates the "first bidding scenario" for the "first bidding price scenario," the "second bidding scenario" for the "second bidding price scenario," and the "third bidding scenario" for the "third bidding price scenario."

[0085] In this way, the bidding scenario acquisition unit 102 can create multiple "bidding scenarios". Figure 9 shows how three bidding scenarios A1 to A3 have been created.

[0086] Furthermore, when creating multiple bidding scenarios, the bidding scenario acquisition unit 102 may set the bid price for each bidding time frame using random numbers. For example, the bidding scenario acquisition unit 102 first creates one bidding scenario using one of the methods described above, and then creates multiple bidding scenarios with different bid prices by varying the bid price for each bidding time frame of this bidding scenario using random numbers. This configuration makes it possible to create bidding scenarios quickly and in a simple manner.

[0087] <Support information generation section> Next, the support information generation unit 103 generates support information, which includes evaluation information of the bidding scenario, based on the "prediction scenario" (highest successful bid price scenario) and the "bidding scenario," and includes information on the probability of winning the bid and the estimated revenue within the bidding time frame.

[0088] The evaluation information regarding the probability of winning the bid and the estimated revenue will be explained with reference to the example shown in Figure 10. Figure 10 is an example of creating multiple prediction scenarios (highest bid price scenarios), and the predicted values ​​of the highest bid price for each frame are displayed as a distribution with a range. In Figure 10, the distribution of the predicted values ​​of the highest bid price may appear to be uniform, but the distribution of the predicted values ​​of the highest bid price is a distribution in which the density differs depending on the price (for example, a Gaussian distribution in this embodiment). In the example shown in Figure 10, the range within ±2σ of the mean of the predicted values ​​is displayed as a region with a range. In the case of a Gaussian distribution, the distribution function can be determined from the mean and standard deviation of the predicted values, so if the position of the bid price within the distribution of the predicted values ​​of the highest bid price is known, the probability of winning the bid at that bid price can be calculated. For example, if the position of the bid price is at the "mean + 1σ" position within the distribution of the predicted values ​​of the highest bid price, the probability of winning the bid is approximately 32%, and if it is at the "mean + 2σ" position, the probability of winning the bid is approximately 5%.

[0089] Figure 10 also shows bidding scenario A1 as an example of a bidding scenario. Bidding scenario A1 is a plan to sell electricity in the supply and demand adjustment market in frame 1 and frame 2, so the bidding price and transaction volume for frame 1 and frame 2 are shown.

[0090] First, looking at frame 1, we can see that the bid price is above the upper limit of the predicted range for the highest successful bid price. Therefore, in bidding scenario A1, while a successful bid on frame 1 could yield high returns, the actual probability of success is not high (less than 5%). Thus, trading frame 1 in bidding scenario A1 can be considered high-risk, high-return.

[0091] On the other hand, in the case of frame 2, the bid price is lower than the bid price for frame 1, and it is located near the middle of the distribution range of the predicted highest successful bid price. Therefore, in bidding scenario A1, the profit if successful in bidding on frame 2 is moderate. Also, since the probability of winning the bid on frame 2 is near the middle of the distribution of predicted values, it cannot be said that a successful bid is guaranteed, but the probability of winning is higher than that of frame 1 (around 50%). Therefore, the transaction in frame 2 in bidding scenario A1 can be said to be a medium-risk, medium-return transaction.

[0092] Based on these findings, the support information generation unit 103 can generate support information for bidding scenario A1 that includes an analysis result indicating that the estimated revenue if the bid is won is somewhat high, but the probability of winning the bid is somewhat low.

[0093] By performing such analysis, the support information generation unit 103 generates support information that evaluates the bidding scenario, including information on the probability of winning the bid and the estimated revenue within the bidding time frame, based on the "prediction scenario" and the "bidding scenario".

[0094] Figure 12 shows examples of output screens for the support information generated by the support information generation unit 103 for the three bidding scenarios A1, A2, and A3.

[0095] The graph displayed near the center of Figure 12 plots the return (expected value of revenue) and risk for three bidding scenarios A1, A2, and A3 on a two-dimensional coordinate plane. This method allows for a clear and visual presentation of the characteristics of multiple bidding scenarios to the user.

[0096] When a user clicks on one of the icons representing the three bidding scenarios plotted on the graph (black circles in the example shown in Figure 12), a separate screen opens displaying the optimal trading plan (the trading plan that is expected to maximize profits) for bidding in that scenario. In the trading plan for each bidding scenario, the bar graph labeled "Purchase" shows the timing (times) and quantity of electricity purchased in the spot market, while the bar graph labeled "Sell" shows the timing (times) and quantity of electricity sold as adjustment in the supply and demand adjustment market. (The electricity sales in "Plan A1" and "Plan A3" target the supply and demand adjustment market, but the electricity sales in "Plan A2" target the spot market. When buying and selling in the spot market, it is not necessary to submit a bid price (this is why the bid price is not shown for "Plan A2" in Figure 12). The spot market uses a single-price system, so there is no risk of not winning a bid like in a multi-price system, and it tends to be relatively low-risk. For this reason, in this example, it is ranked first in the recommendation order for "low-risk emphasis.") Although not shown in Figure 12, the expected execution price for each transaction and the estimated revenue if electricity trading were conducted according to the bidding scenario can also be displayed.

[0097] Furthermore, since the priority given to risk or return varies depending on the circumstances and the user's policies, it is also possible to display tables of each bidding scenario sorted by return (highest first) and by risk (lowest first), as shown in Figure 12 as the recommended ranking. This approach makes it easier for users to select a bidding scenario that better suits their perspective on risk and return.

[0098] In this way, the support information generation unit 103 uses one or more prediction scenarios to calculate the probability of winning the bid and the estimated revenue for each of the one or more bidding scenarios within the bidding time frame, and generates support information based on these calculation results. For example, if the bidding scenario acquisition unit 102 generates the first bidding scenario, the second bidding scenario, and the third bidding scenario described above, the support information generation unit 103 generates support information for each of the first bidding scenario, the second bidding scenario, and the third bidding scenario based on the probability of winning the bid and the estimated revenue for each of the bidding time frames calculated for the combination of the prediction scenario and the first bidding scenario, the combination of the prediction scenario and the second bidding scenario, and the combination of the prediction scenario and the third bidding scenario.

[0099] This configuration allows users to select the optimal bidding scenario and revise their bidding scenarios (such as changing to a bidding scenario that is expected to generate higher profits, changing to a lower-risk bidding scenario that is more likely to win the bid, or canceling the bid altogether).

[0100] ==Processing Flow== Next, the processing flow of the information processing device 100 according to this embodiment will be explained with reference to the flowchart shown in Figure 13.

[0101] First, the information processing device 100 uses historical information, including past successful bid prices in an electricity trading market where the successful bid price is determined by a multi-price auction method such as a supply and demand adjustment market, to generate a maximum successful bid price scenario (prediction scenario) that predicts the highest successful bid price for each trading time frame on the trading day (S1000). In this embodiment, the historical information is stored in the initial data table 600. The information processing device 100 may generate one maximum successful bid price scenario, or it may generate multiple maximum successful bid price scenarios as shown in Figure 7.

[0102] The information processing device 100 then obtains a bidding scenario specifying the bid price and transaction volume for the bidding time slot (bid frame) on the trading day (S1010). The information processing device 100 may obtain a bidding scenario by receiving input from the user regarding the bid frame, transaction volume, and bid price through a user interface as shown in Figure 11, or it may create a bidding scenario using historical information and a prediction model 610. In this case, the information processing device 100 may create one bidding scenario, or it may create multiple scenarios as shown in Figure 9.

[0103] Next, the information processing device 100 generates evaluation information of the bidding scenario as supporting information, based on the highest successful bid price scenario and the bidding scenario, including information on the probability of winning the bid and the estimated amount of revenue within the bidding time frame (S1020).

[0104] This approach can support users who want to earn more profits by trading electricity in the electricity trading market, where the winning bid price is determined by a multi-price auction. In particular, since the support information includes information on the likelihood of winning a bid and the estimated profit amount, it is possible to provide support information that helps each user create the optimal bidding scenario, depending on their policy, such as whether they want to aim for higher profits while accepting the risk of not winning a bid, or whether they want to ensure they win a bid rather than aiming for high profits.

[0105] As described above, the information processing device 100, the control method for the information processing device 100, and the information processing device control program 700 according to this embodiment make it possible to output support information for conducting electricity trading in an electricity trading market where the winning bid price is determined by a multi-price auction method.

[0106] The embodiments described above are provided to facilitate understanding of the present invention and are not intended to limit its interpretation. The present invention may be modified or improved without departing from its spirit, and equivalents thereof are also included. [Explanation of symbols]

[0107] 100 Information Processing Devices 101 Prediction Scenario Generation Unit 102 Bidding Scenario Acquisition Section 103 Support information generation section 110 CPU 120 memory 130 Communication equipment 140 Storage device 150 Input Devices 160 Output device 170 Recording medium reading device 200 Battery 500 Networks 600 Initial Data Table 610 Predictive Models 610A First Prediction Model 610B Second Prediction Model 610C Third Prediction Model 610D 4th Prediction Model 620 Equipment Data Table 700 Information Processing Device Control Program 800 recording media 1000 Information Processing Systems

Claims

1. An information processing device that outputs support information for conducting electricity trading in an electricity trading market where the winning bid price is determined by a multi-price auction method, A prediction scenario generation unit generates a prediction scenario that predicts the highest winning bid price for each trading time slot on the trading day, using historical information including past winning bid prices in the aforementioned electricity trading market. A bidding scenario acquisition unit acquires a bidding scenario that specifies the bid price and trading volume for the bidding time frame, which is the trading time frame in which bids are made. A support information generation unit generates support information, which includes evaluation information of the bidding scenario, including information on the probability of winning the bid and the estimated revenue within the bidding time frame, based on the prediction scenario and the bidding scenario. An information processing device equipped with the following features.

2. An information processing apparatus according to claim 1, The prediction scenario generation unit uses the historical information to generate multiple prediction scenarios in which the predicted value of the highest successful bid price for at least one of the trading timeframes differs. The support information generation unit generates the support information based on the probability of winning the bid within the bidding time frame and the estimated revenue amount calculated for the bidding scenario using a plurality of prediction scenarios. Information processing device.

3. An information processing apparatus according to claim 1, The aforementioned bidding scenario acquisition unit acquires multiple bidding scenarios in which at least one of the bidding time frame, bid price, and transaction volume is different. The support information generation unit generates the support information based on the probability of winning the bid and the estimated revenue calculated for each of the bidding scenarios using the prediction scenario. Information processing device.

4. An information processing apparatus according to claim 3, The prediction scenario generation unit uses the historical information to generate multiple prediction scenarios in which the predicted value of the highest successful bid price for at least one of the trading timeframes differs. The support information generation unit generates the support information based on the probability of winning the bid and the estimated revenue calculated for each of the bidding scenarios using a plurality of prediction scenarios. Information processing device.

5. An information processing apparatus according to claim 1, The bidding scenario acquisition unit acquires a first bidding scenario in which the bid price for the bidding time frame is the same as the predicted highest successful bid price for that bidding time frame calculated using the historical information, a second bidding scenario in which the bid price for the bidding time frame is the same as the predicted lowest successful bid price for that bidding time frame calculated using the historical information, and a third bidding scenario in which the bid price for the bidding time frame is the same as the predicted average successful bid price for that bidding time frame calculated using the historical information. The support information generation unit generates the support information for each of the first, second, and third bidding scenarios based on the probability of winning the bid and the estimated revenue calculated for each of the combinations of the prediction scenario and the first bidding scenario, the prediction scenario and the second bidding scenario, and the prediction scenario and the third bidding scenario. Information processing device.

6. An information processing apparatus according to claim 3, The bidding scenario acquisition unit acquires multiple bidding scenarios by setting the bid price for the bidding time frame using random numbers. Information processing device.

7. A control method for an information processing device that outputs support information for conducting electricity trading in an electricity trading market where the winning bid price is determined by a multi-price auction method, The aforementioned information processing device Using historical information including past successful bid prices in the aforementioned electricity trading market, a prediction scenario is generated that predicts the highest successful bid price for each trading time slot on the trading day. Obtain a bidding scenario that specifies the bid price and trading volume for the bidding time frame, which is the trading time frame in which bids are made. Based on the aforementioned prediction scenario and the aforementioned bidding scenario, evaluation information of the bidding scenario, including information on the probability of winning the bid and the estimated revenue within the bidding time frame, is generated as the aforementioned support information. A method for controlling an information processing device.

8. A program that causes a computer to output support information for conducting electricity trading in an electricity trading market where the winning bid price is determined by a multi-price auction method, To the aforementioned computer, A procedure for generating a prediction scenario that predicts the highest winning bid price for each trading time slot on a trading day, using historical information including past winning bid prices in the aforementioned electricity trading market, The procedure for obtaining a bidding scenario that specifies the bid price and trading volume for the bidding time frame, which is the trading time frame in which bids are made, A procedure for generating evaluation information of the bidding scenario as supporting information, based on the prediction scenario and the bidding scenario, including information on the probability of winning the bid and the estimated revenue within the bidding time frame, A program that executes the command.

Images