Information processing device, information processing method, and program

Abstract

To find an optimal bid amount for a transaction target with greater precision.SOLUTION: An information processing apparatus includes a processing unit. The processing unit creates a plurality of power generation amount scenarios including a plurality of third power generation amount prediction values based on a plurality of pieces of power generation amount performance information and a second power generation prediction value. The processing unit calculates a bid range indicating a range of a power generation amount to be bid into a first market based on the plurality of power generation amount scenarios. The processing unit creates a plurality of price scenarios including a plurality of third price prediction values of a price based on the plurality of pieces of price performance information and a second price prediction value. The processing unit calculates a bid amount representing the power generation amount to be bid into the first market based on the bid range and the plurality of price scenarios.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] Embodiments of the present invention relate to an information processing device, an information processing method, and a program. [Background technology]

[0002] Electricity trading markets are known for the trading of electricity generated by power generators and other operators. For example, in Japan, electricity is traded in multiple markets operated by the Japan Electric Power Exchange (JEPX). These multiple markets include, for example, the spot market and the day-ahead market. The spot market is a market where bids are placed and transactions are completed by the day before for electricity to be sold or purchased the following day. The day-ahead market (same-day market) is a market that can be used right up until the actual supply and demand.

[0003] Furthermore, a technology has been proposed to calculate the optimal ratio in which power generators should divide their generated electricity and auction it in multiple markets (spot market, auction-ahead market, etc.) to maximize their profits. [Prior art documents] [Patent Documents]

[0004] [Patent Document 1] Japanese Patent Publication No. 2022-079368 [Non-patent literature]

[0005] [Non-Patent Document 1] Norio Hibiki, "Introduction to Portfolio Optimization," Journal of Operations Research, Vol. 61, June 2016, pp. 335-340. [Overview of the Initiative] [Problems that the invention aims to solve]

[0006] An object of the present invention is to provide an information processing apparatus, an information processing method, and a program that can more accurately obtain an optimal bidding amount for a transaction target in a market such as electric power.

Means for Solving the Problems

[0007] The information processing apparatus according to the embodiment includes a processing unit. The processing unit creates a plurality of power generation amount scenarios including a plurality of third power generation amount prediction values based on a plurality of power generation amount performance information and a second power generation amount prediction value. The processing unit calculates a bidding range indicating a range of the power generation amount to be bid for the first market based on the plurality of power generation amount scenarios. The processing unit creates a plurality of price scenarios including a plurality of third price prediction values of prices based on a plurality of price performance information and a second price prediction value. The processing unit calculates a bidding amount representing the power generation amount to be bid for the first market based on the bidding range and the plurality of price scenarios.

Brief Description of the Drawings

[0008] ​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​A block diagram of an information processing device according to the second embodiment. [Figure 13] Flowchart of the bid amount calculation process in the second embodiment. [Figure 14] Hardware configuration diagram of an information processing device according to an embodiment. [Modes for carrying out the invention]

[0009] A preferred embodiment of the information processing device according to the present invention will be described in detail below with reference to the attached drawings. In the following description, the market and the object of trade in the market will be described as an electricity trading market and electricity, but the market and the object of trade are not limited to these. In this example, the trading quantity, which represents the amount of the object of trade (electricity) traded in the market, corresponds to the amount of electricity generated.

[0010] (First embodiment) JEPX stipulates that electricity generated should be traded in the spot market as much as possible, and any surplus or deficit in the amount generated on a given day should be bought and sold in the pre-hour market. Conventional technologies are not designed to take these regulations into account, which may lead to situations where the regulations are not followed. In the following, the spot market may be referred to as the main trading market (an example of the first market). Markets other than the spot market, including the pre-hour market, may be referred to as non-main trading markets.

[0011] This embodiment assumes that there are two or more electricity trading markets, such as JEPX, and that the amount of electricity generated is bid on in the main trading market in a way that minimizes surplus or deficit (in accordance with JEPX regulations). Based on this premise, the information processing device of this embodiment determines the optimal bid amount for the main trading market.

[0012] The following explanation uses the example of multiple markets consisting of two markets: a primary trading market and a secondary trading market. The secondary trading market is, for example, a pre-sale market, but may also include other markets. Furthermore, the following explanation uses the example of a power generator selling the electricity it generates on multiple markets. The bid amount for electricity purchased by electricity consumers can be determined using a similar procedure.

[0013] The information processing device according to this embodiment has, for example, the following functions and determines an appropriate bid amount for major trading markets. (F1) A function that calculates the range of power generation (bidding range) considering prediction error (risk) using multiple past power generation performance data and predicted power generation values ​​(hereinafter referred to as predicted power generation value PEb). (F2) A function that uses historical price data from multiple markets (major and minor trading markets) and price forecasts (hereinafter referred to as price forecasts PCb) to determine the bid volume for the major trading market within the bidding range, in order to achieve stable and high revenue as much as possible.

[0014] The actual power generation data includes, for example, past power generation forecast values ​​PEa (first trading volume forecast value) and the actual amount of power generated that was traded relative to the power generation forecast value PEa. The power generation forecast value PEb (second trading volume forecast value) is, for example, a forecast of the amount of electricity that will actually be traded the following day, and is the amount of power generated that is predicted from weather forecasts, etc.

[0015] Historical price information includes, for example, past price forecasts PCa (first price forecast) and actual trading prices against price forecast PCa. Price forecast PCb (second price forecast) is, for example, a forecast of the market price that will actually be traded the following day. Note that price (including price forecast and market price) includes not only prices in major trading markets but also prices in non-major trading markets.

[0016] Function F1, for example, creates multiple power generation scenarios, each containing multiple power generation forecast values ​​PEc (third trading volume forecast value). Based on these multiple power generation scenarios, a power generation distribution based on empirical distributions is created, and the bidding range is calculated using this power generation distribution. The power generation scenarios represent information that shows possible power generation amounts, taking into account the prediction errors that may be included in the power generation forecast value PEb.

[0017] Function F2, for example, creates multiple price scenarios, each containing multiple price prediction values ​​PCc (third price prediction values), and calculates the bid volume for the primary trading market within the bidding range based on these multiple price scenarios. The price scenarios represent information that shows candidate pairs of prices that are likely to occur in the primary trading market (primary trading market price) and prices in non-primary trading markets (non-primary trading market price).

[0018] Figure 1 is a block diagram showing an example of the configuration of an information processing device 100 according to the first embodiment. As shown in Figure 1, the information processing device 100 includes a storage unit 121, a reception unit 101, a power generation scenario creation unit 102 (an example of a transaction volume scenario creation unit), a price scenario creation unit 103, a range calculation unit 104, a bid amount calculation unit 105, and an output control unit 106.

[0019] The storage unit 121 stores various types of data used by the information processing device 100. For example, the storage unit 121 stores data received by the reception unit 101, as well as processing results from other units.

[0020] The storage unit 121 can be composed of any commonly used storage medium, such as flash memory, memory card, RAM (Random Access Memory), HDD (Hard Disk Drive), and optical disc.

[0021] The reception unit 101 accepts input of various data to be used in the information processing device 100. For example, the reception unit 101 accepts multiple power generation performance data, power generation forecast value PEb, multiple price performance data, and price forecast value PCb. The reception unit 101 may also accept information used to calculate the bidding range. This information includes, for example, a lower percentile value indicating the lower percentile of the bidding range and an upper percentile value indicating the upper percentile of the bidding range. The reception unit 101 may further accept the number of power generation scenarios to be created (number of power generation scenarios) and the number of price scenarios to be created (number of price scenarios).

[0022] The power generation scenario creation unit 102 creates multiple power generation scenarios, each containing multiple predicted power generation values ​​PEc, based on multiple actual power generation data and a predicted power generation value PEb. If the number of power generation scenarios is entered, the power generation scenario creation unit 102 creates power generation scenarios equal to the number of scenarios.

[0023] The price scenario creation unit 103 creates multiple price scenarios, each containing multiple price forecast values ​​PCc, based on multiple price history data and price forecast values ​​PCb. If the number of price scenarios is entered, the price scenario creation unit 103 creates price scenarios equal to the number of price scenarios.

[0024] The range calculation unit 104 calculates a bidding range that indicates the range of power generation to be bid on to the main trading market, based on multiple power generation scenarios. For example, the range calculation unit 104 calculates a bidding range that uses the lower percentile value of multiple power generation forecast values ​​PEc included in the multiple power generation scenarios as the lower limit and the upper percentile value as the upper limit.

[0025] The bid amount calculation unit 105 calculates a bid amount representing the amount of power generation to be bid on to the main trading market, based on the bid range and multiple price scenarios. For example, the bid amount calculation unit 105 calculates a bid amount that is included in the bid range and is common to multiple price scenarios that maximize the valuation profit. The definition of valuation profit includes the expected value of revenue, revenue based on the CVaR (Conditional Value at Risk) concept, and revenue based on the VaR (Value at Risk) concept.

[0026] The output control unit 106 controls the output of various data used by the information processing device 100. For example, the output control unit 106 outputs the bid amount calculated by the bid amount calculation unit 105. The output method by the output control unit 106 can be any method, but for example, it can be used to send data to an external device (server, other information processing device, etc.), to display it on a display device such as a liquid crystal display, or to output it to a recording medium using an image forming device such as a printer.

[0027] At least a portion of each of the above components (reception unit 101, power generation scenario creation unit 102, price scenario creation unit 103, range calculation unit 104, bid amount calculation unit 105, and output control unit 106) may be implemented by a single processing unit. Each of the above components may be implemented by, for example, one or more processors. For example, each of the above components may be implemented by having a processor such as a CPU (Central Processing Unit) execute a program, i.e., by software. Each of the above components may be implemented by a processor such as a dedicated IC (Integrated Circuit), i.e., by hardware. Each of the above components may be implemented by using both software and hardware. When multiple processors are used, each processor may implement one of the above components, or two or more of the above components.

[0028] The following provides further details on the processing of each of the above parts. Note that each of the following processes is executed for each specific target period (time frame: what JEPX calls a "product"). The target period is determined according to the period during which electricity bids are made, for example, as predetermined by the operator of the electricity trading market such as JEPX. For example, the target period is in 30-minute increments. In this case, one day is divided into 48 target periods. The 48 target periods can be identified by identification information from 1 to 48. For example, the identification information for 0:00 to 0:30 is "1", the identification information for 0:30 to 1:00 is "2", ..., the identification information for 23:30 to 0:00 is "48".

[0029] First, the details of the power generation scenario creation unit 102's processing will be explained. The power generation scenario creation unit 102 creates multiple power generation scenarios for a specified number of times, based on the predicted power generation value PEb for the target period on the current day and the actual power generation information for each target period, which is identified by the same identification information from the previous day or nearby identification information. The number of power generation scenarios can be any value, but for example, if it is set to several hundred, the bidding range (bid amount) can be determined with higher accuracy.

[0030] While it is possible to use past power generation figures as the power generation scenario, if there is insufficient historical data, it may not be possible to calculate the bidding range accurately. In this embodiment, by allowing the number of power generation scenarios to be specified, it becomes possible to create a more appropriate number of power generation scenarios.

[0031] The power generation scenario creation unit 102 performs the following steps. Step 1-1: The power generation scenario creation unit 102 uses t as the identification information for the target period and Δt as the number of periods for setting similar time zones. A similar time zone represents a time zone in which a power generation amount similar to that of the target period is expected to be obtained. Similar time zone data, which is power generation performance information similar to that of the target period, is extracted from the similar time zones. The power generation scenario creation unit 102 extracts similar time zone data from the range expressed by the following equation (1) in the past power generation performance information.

number

[0032] The value of Δt is set according to the power generation method, for example. For example, in the case of a power generation method where the amount of power generated changes greatly depending on the time of day, such as solar power generation, it is set to a small value such as Δt=0. Note that Δt=0 means that, from the past power generation performance information, power generation performance information for the same period as the target period of the identification information used to calculate the bidding amount for the current day is extracted as similar time period data.

[0033] Step 1-2: Let A be the number of data points extracted in Step 1-1, and let X be the neighbor ratio (%). The neighbor ratio is used to determine the range (neighbor range) that contains data (neighbor data) that is closer to the predicted power generation value PEb among similar time zone data. The neighbor ratio is specified, for example, by the user and accepted by the reception unit 101. The power generation scenario creation unit 102 calculates a neighbor range that contains m similar time zone data points represented by the following equation (2), centered on the predicted power generation value PEb, and selects m similar time zone data points included in the neighbor range as neighbor data.

number

[0034] Figure 2 shows an example of selected neighboring data. Data 201 corresponds to the predicted power generation value PEb. Multiple data points 202, represented by circles other than data 201, correspond to extracted similar time period data. Range 210 corresponds to the neighboring range. The four data points d1 to d4 within range 210 correspond to the selected neighboring data.

[0035] Step 1-3: The power generation scenario creation unit 102 calculates the error amount of neighboring data. The error amount is, for example, the difference between the predicted power generation value PEa and the actual value included in the actual power generation information (similar time period data). Hereafter, the data numbers will be 0, 1, ..., m-1 in order of increasing error amount of neighboring data. The data number is information that identifies neighboring data and is a non-negative integer.

[0036] Steps 1-4: The power generation scenario creation unit 102 plots the neighboring data on a coordinate system where the horizontal axis represents the error amount of the neighboring data and the vertical axis represents the data number. Figure 3 shows the distribution (distribution function) of the plotted neighboring data. The distribution shown in Figure 3 is a distribution function based on the empirical distribution of the error amount.

[0037] Steps 1-5: The power generation scenario creation unit 102 randomly selects a value between data numbers 0 and M-1. The selected value may be a real number. The power generation scenario creation unit 102 uses an empirical distribution as shown in Figure 3 to determine the error amount corresponding to the selected value. If a value corresponding to a neighboring data point is selected, the power generation scenario creation unit 102 can determine the error amount corresponding to the selected value by interpolation or other means.

[0038] Figure 4 shows an example of the process for calculating the error amount. In the example in Figure 4, the error amount 402 corresponding to the value 401 between data numbers 5 and 6 is calculated.

[0039] The power generation scenario creation unit 102 calculates a power generation forecast value PEc by adding the calculated error amount to the power generation forecast value PEb, and creates a new power generation scenario that includes the power generation forecast value PEc. The power generation scenario creation unit 102 repeats steps 1-5 for the specified number of power generation scenarios, creating power generation scenarios for that number of scenarios.

[0040] Next, we will explain the details of the processing of the range calculation unit 104.

[0041] The range calculation unit 104 arranges the multiple power generation scenarios created by the power generation scenario creation unit 102 in order of the power generation prediction value PEc.

[0042] Figure 5 shows an example of a list of predicted power generation values ​​PEc. In Figure 5, 10 power generation scenarios S1 to S10 (number of power generation scenarios = 10) are created, and an example is shown of the 10 predicted power generation values ​​PEc included in the power generation scenarios being listed. The value 501 corresponds to the predicted power generation value PEb.

[0043] Since the power generation scenarios are created using an empirical distribution generated based on past power generation performance data, the predicted power generation values ​​for any of the power generation scenarios can actually occur as the predicted power generation. However, since the actual power generation values ​​may contain outliers, the power generation scenarios may also contain outliers. Therefore, the range calculation unit 104 calculates the lower percentile value β L And the upper percentile value β H Using these methods, the bid range is calculated to exclude outliers as much as possible.

[0044] In the example in Figure 5, there are 10 power generation scenarios, so β=0.1 corresponds to power generation scenario S1, ..., β=0.5 corresponds to power generation scenario S5, ..., and β=1.0 corresponds to power generation scenario S10. The lower percentile value β=0.5 corresponds to the median (the most likely power generation considering prediction error). Therefore, β L , β H In configurations where input is not provided, β L =0.4, β H It is reasonable to use a range that includes β=0.5, such as =0.6, as the default value.

[0045] Specifying β<0.5 means that the predicted power generation amount will take into account the forecast error, giving more weight to the possibility of a downward revision in power generation. In other words, the amount of bids in the major trading market will generally decrease, and the risk of electricity buybacks in other electricity trading markets will decrease. Buyback risk represents the risk of buying back electricity from markets other than the major trading market.

[0046] Specifying β>0.5 results in a generated power amount that takes into account prediction errors with an emphasis on an increase in the generated power amount. That is, the bidding volume in the main trading market increases, and the additional selling risk of electricity in other electricity trading markets decreases. The additional selling risk represents the risk of selling additional electricity to markets other than the main trading market.

[0047] Therefore, considering which of the additional selling list and the buy-back risk is emphasized, the lower percentile value β L and the upper percentile value β H should desirably be made specifiable.

[0048] FIG. 6 is a diagram showing an example of correspondence information that associates the importance levels of the additional selling list and the buy-back risk (importance level of additional selling risk·buy-back risk) with combinations of set values of the lower percentile value β L and the upper percentile value β H .

[0049] For example, the user designates one of the importance levels of the additional selling risk·buy-back risk of the correspondence information as shown in FIG. 6. The reception unit 101 receives the input of the designated importance level of the additional selling risk·buy-back risk, and sets the set value corresponding to the received importance level of the additional selling risk·buy-back risk as the values of the lower percentile value β L and the upper percentile value β H .

[0050] Note that the method of designating percentile values using correspondence information as shown in FIG. 6 is an example, and other designation methods may be used. For example, the reception unit 101 may separately receive the input of the values of the lower percentile value β L and the upper percentile value β H .

[0051] Furthermore, the risks that can be anticipated are not limited to buyback risk and additional selling risk. For example, at JEPX, if the difference from the predicted power generation amount cannot be processed (traded) even in non-major trading markets, trading is forced at a penalty price called the imbalance price. The imbalance price is unfavorable to power generators and can be a risk factor. Percentile values ​​may be specified taking into account risks other than buyback risk and additional selling risk.

[0052] The range calculation unit 104 calculates the bidding range from a range that includes multiple predicted power generation values ​​PEc, with the specified lower percentile value as the lower limit and the specified upper percentile value as the upper limit. In the example in Figure 5, the lower percentile value β L = 0.5, upper percentile value β H If the value is 0.8, the range calculation unit 104 calculates the bidding range as 840kWh to 880kWh.

[0053] Next, we will explain the details of the processing in the price scenario creation unit 103.

[0054] The price scenario creation unit 103 creates a specified number of price scenarios for each target period, each scenario showing candidate pairs of major and minor trading market prices that are likely to occur.

[0055] The price scenario creation unit 103 performs the following steps: Step 2-1: The price scenario creation unit 103 uses t as the identification information for the target period and Δt as the number of periods for setting similar time zones. The price scenario creation unit 103 extracts similar time zone data from the range represented by equation (1) above in the historical price performance information.

[0056] Step 2-2: Let A be the number of data points extracted in Step 2-1, and let X be the neighbor ratio (%). The price scenario creation unit 103 calculates the radius of a circle containing m similar time zone data points represented by equation (2) above, centered on the price forecast value PCb (price forecast value for major trading markets, price forecast value for non-major trading markets) for the target period, and selects m similar time zone data points contained within that circle as neighbor data.

[0057] Figure 7 shows an example of selected neighboring data. Data 701 corresponds to the price prediction value PCb. Multiple data points 702, represented by circles other than data 701, correspond to extracted similar time zone data. Range 710 corresponds to the calculated circle. The four data points d1 to d4 within range 710 correspond to selected neighboring data.

[0058] Step 2-3: The price scenario creation unit 103 calculates the error amount for neighboring data. The error amount is, for example, the difference between the predicted price PCa and the actual price included in the price performance information (similar time period data). The error amount is calculated for both the major trading market price and the non-major trading market price.

[0059] Step 2-4: The price scenario creation unit 103 plots the neighboring data on a coordinate system where the horizontal axis represents the error amount of the major trading market price and the vertical axis represents the error amount of the non-major trading market price. Furthermore, the price scenario creation unit 103 calculates the Voronoi region (boundary) and convex hull with the neighboring data as the generating points.

[0060] Figure 8 shows an example of a Voronoi boundary and convex hull. The dashed line 802 corresponds to the Voronoi boundary. The region enclosed by the solid line 801 corresponds to the convex hull. Performing the operations in steps 2-5 and 2-6 below using a two-dimensional graph like the one shown in Figure 8 can be interpreted as corresponding to the function of creating a scenario that follows a two-dimensional empirical distribution function.

[0061] Step 2-5: The price scenario creation unit 103 randomly selects one of several neighboring data points.

[0062] Step 2-6: The price scenario creation unit 103 randomly selects points within the Voronoi region and convex hull, with the neighboring data points selected in Step 2-5 as its origin points. The price scenario creation unit 103 considers the selected points as the error amounts of the market price (error amount of the major trading market price, error amount of the non-major trading market price), and calculates a price forecast value PCc by adding this error amount to the price forecast value PCb for the target period, and creates a new price scenario that includes the price forecast value PCc.

[0063] The price scenario creation unit 103 repeats steps 2-5 and S2-6 for the specified number of price scenarios, creating price scenarios equal to the number of price scenarios.

[0064] Next, we will explain the details of the processing of the bid amount calculation unit 105.

[0065] The bid volume calculation unit 105 calculates the bid volume (w) for major trading markets that will generate stable and high revenue, using the price scenario created by the price scenario creation unit 103 within the bid range created by the range calculation unit 104. spot Calculate ).

[0066] In the following explanation, the bidding range will be W min ~W max In the power generation scenario, the amount of power generated at the percentile value β=0.5 is W mid This will be the case. Also, the amount of bids to non-major trading markets will be w intraday (=W mid -w spot ) is written as follows.

[0067] The bid volume for major trading markets that provides stable, high income is the amount that maximizes the expected return on scenarios below the lower alpha percentile when price scenarios are sorted from lowest to highest profit. spot This refers to (0≦α≦1.0). spot The problem of finding corresponds to scenario optimization based on the CVaR concept. Note that α may be a value provided by the user, or it may be given as a fixed default value (e.g., α=0.8).

[0068] The bid quantity calculation unit 105 is w spot The problem of finding is formulated as a linear programming problem, and the linear programming problem is solved. The method of formulating it as a linear programming problem is explained below. In the following explanation, the set of power generation scenarios is Z, and the set of price scenarios is Y. Price scenario y i ∈Y (where i is an integer between 1 and the number of price scenarios) includes the primary and secondary trading market prices. Power generation scenario z j ∈Z (where j is an integer between 1 and the number of power generation scenarios) includes the amount of power generated.

[0069] The definitions of the constants and variables used when formulating the problem into a linear programming problem are shown in Figures 9 and 10, respectively. The values ​​of each constant shown in Figure 9 can be obtained from the input data and the processing results of the respective units (power generation scenario creation unit 102, price scenario creation unit 103, and range calculation unit 104).

[0070] Next, we will explain the constraint equations for the linear programming problem. The bid quantity w for non-major trading markets. intraday This corresponds to an auxiliary variable, w intraday =W mid -w spot It is expressed as follows: The constraint equation that limits the amount of bids to the main trading market to within the bid range is W min ≤w spot ≤W max It is represented as follows.

[0071] The definition of the auxiliary variable representing profit is given by equation (3) below. However, R Y [·] represents the expected sales for the T scenarios with the lowest profits from the set of market price scenarios Y. The value of T can be calculated from the value of α and the number of price scenarios to be created. For example, the value of T can be calculated as the integer part of price scenario × α.

number

[0072] The method for expanding equation (3) into linear form is as follows: R YThe expansion of [·] applies the formulation of the CVaR model (e.g., Non-Patent Document 1). Specifically, by adding the constraint expressed in equation (4) below, equation (3) can be expanded into a linear constraint. yi This is an auxiliary variable with no range restrictions.

number

[0073] Add the following constraint equation (5), where γ, u yi is a newly introduced auxiliary variable. γ is a value with no range restrictions, and u yi ≥ 0.

number

[0074] Using these constraints, equation (6) below can be replaced by equation (7) below.

number

number

[0075] Next, we will explain the objective function of a linear programming problem. The objective function for maximizing the expected value of profit is expressed as max.prof.

[0076] The bid volume calculation unit 105 calculates the bid volume for major trading markets that will generate stable, high revenue (based on the CVaR approach) by solving the linear programming problem formulated in this manner. Any conventional method can be used to solve the linear programming problem.

[0077] So far, we have shown examples of scenario optimization based on CVaR, but this is not the only way. The bid amount calculation unit 105 may use a method to maximize the expected value of profit based on VaR, or a simpler method to maximize the expected value of profit. The simpler method to maximize the expected value of profit is equivalent to setting α=1 and maximizing the expected value of revenue for scenarios below the lower αth percentile.

[0078] Next, the bid amount calculation process by the information processing device 100 according to the first embodiment will be described. Figure 11 is a flowchart showing an example of the bid amount calculation process in the first embodiment.

[0079] The reception unit 101 accepts input data to be used in creating scenarios (power generation scenario, price scenario) (step S101). The power generation scenario creation unit 102 creates multiple power generation scenarios according to an empirical distribution (step S102). The range calculation unit 104 calculates the bid range for power generation from the created multiple power generation scenarios (step S103). The price scenario creation unit 103 creates multiple price scenarios according to an empirical distribution (step S104). The bid amount calculation unit 105 solves an optimization problem based on the bid range, power generation scenario, and price scenario and calculates the optimal bid amount (step S105). The output control unit 106 outputs the calculated bid amount (step S106) and terminates the bid amount calculation process.

[0080] In the JEPX market, the minimum trading unit is set at 50kWh. Therefore, the output control unit 106 calculates the bid amount w spot If the value is not a multiple of 50kWh, you may round up or down the value to make it a multiple of 50kWh and output the result.

[0081] Thus, in the first embodiment, the functions of (F1) and (F2) described above make it possible to determine the appropriate bid quantity for the major trading market with greater accuracy.

[0082] Furthermore, in technologies that determine the ratio of electricity generated to be bid on in multiple markets (such as major and minor trading markets) in order to maximize profits, there may be cases where regulations such as the requirement to trade electricity in the major trading market (spot market) as much as possible are not followed. In contrast, this embodiment calculates the optimal bid amount for the major trading market. In other words, it is possible to determine the bid amount with high accuracy while adhering to the above regulations.

[0083] (Second embodiment) In the second embodiment, the optimal bid amount is calculated without using a price scenario.

[0084] Figure 12 is a block diagram showing an example of the configuration of an information processing device 100-2 according to a second embodiment. As shown in Figure 12, the information processing device 100-2 includes a storage unit 121, a reception unit 101-2, a power generation scenario creation unit 102, a bid amount calculation unit 105-2, and an output control unit 106.

[0085] In the second embodiment, the price scenario creation unit 103 and the range calculation unit 104 are omitted, and the functions of the reception unit 101-2 and the bid quantity calculation unit 105-2 differ from those of the first embodiment. The other configurations and functions are the same as those in Figure 1, which is a block diagram of the information processing device 100 according to the first embodiment, so the same reference numerals are used, and their explanation is omitted here.

[0086] The receiving unit 101-2 differs from the receiving unit 101 of the first embodiment in that it does not need to receive data related to the creation of price scenarios. The data related to the creation of price scenarios includes, for example, historical price information, price forecast value PCb, and the number of price scenarios.

[0087] The bid amount calculation unit 105-2 calculates the bid amount based on risk information and multiple power generation scenarios. The risk information represents the importance of buyback risk and additional sell risk, respectively. For example, the risk information is a risk ratio (p:q) that specifies how much weight to give to buyback risk and additional sell risk. The risk information may be configured to be input by the reception unit 101-2.

[0088] The bid amount calculation unit 105-2 calculates the percentile value β = p / (p+q) using, for example, the risk ratio p:q. The bid amount calculation unit 105-2 then calculates the amount of power generated for the power generation scenario corresponding to the calculated percentile value β as the amount of power generated (bid amount) considering the prediction error.

[0089] Next, the bid amount calculation process by the information processing device 100-2 according to the second embodiment will be explained with reference to Figure 13. Figure 13 is a flowchart showing an example of the bid amount calculation process in the second embodiment.

[0090] The reception unit 101-2 accepts data to be used in creating a scenario (power generation scenario) (step S201). The power generation scenario creation unit 102 creates multiple power generation scenarios according to the empirical distribution (step S202). The bid amount calculation unit 105-2 calculates the optimal bid amount from the risk information (risk ratio) and power generation scenario (step S203). The output control unit 106 outputs the calculated bid amount (step S204) and terminates the bid amount calculation process.

[0091] Thus, the information processing device according to the second embodiment can calculate the optimal bid amount using risk information (risk ratio) and power generation scenarios, without using price scenarios.

[0092] As explained above, according to the first and second embodiments, the optimal amount of electricity to bid can be determined with greater accuracy.

[0093] Next, the hardware configuration of the information processing device according to the first or second embodiment will be described using Figure 14. Figure 14 is an explanatory diagram showing an example of the hardware configuration of the information processing device according to the first or second embodiment.

[0094] The information processing device according to the first or second embodiment includes a control device such as a CPU 51, a storage device such as a ROM (Read Only Memory) 52 or RAM 53, a communication I / F 54 that connects to a network for communication, and a bus 61 that connects the various parts.

[0095] The program to be executed by the information processing device according to the first or second embodiment is provided pre-installed in a ROM 52 or the like.

[0096] The program executed by the information processing device according to the first or second embodiment may be configured to be provided as a computer program product by recording it in an installable or executable file format onto a computer-readable recording medium such as a CD-ROM (Compact Disk Read Only Memory), a flexible disk (FD), a CD-R (Compact Disk Recordable), or a DVD (Digital Versatile Disk).

[0097] Furthermore, the information processing device according to the first or second embodiment may be configured to store the program on a computer connected to a network such as the Internet and provide it by allowing download via the network. Alternatively, the information processing device according to the first or second embodiment may be configured to provide or distribute the program via a network such as the Internet.

[0098] A program executed in the information processing device according to the first or second embodiment can cause a computer to function as a part of the information processing device described above. This computer can read a program from a computer-readable storage medium onto its main memory and execute it using the CPU 51.

[0099] An example of the configuration of the embodiment is described below. (Configuration Example 1) Based on multiple trading volume data, each including a first trading volume forecast and actual trading volume corresponding to the first trading volume forecast for a traded asset traded in multiple markets including the first market, and a second trading volume forecast for the said traded asset, multiple trading volume scenarios are created, each including a third trading volume forecast. Based on the multiple trading volume scenarios, a bidding range is calculated that indicates the range of trading volumes to be bid on for the first market. Multiple price scenarios are created, including multiple third price forecasts for the price, based on a plurality of price performance information, each including a first price forecast and an actual price corresponding to the first price forecast, and a second price forecast for the price. Based on the aforementioned bidding range and the multiple aforementioned price scenarios, a bid quantity representing the trading volume to be bid on for the first market is calculated. Processing section An information processing device equipped with the following features. (Configuration example 2) The processing unit calculates the bid range, with the lower percentile value representing the lower limit of the bid range being the lower percentile value, and the upper percentile value representing the upper limit of the bid range being the upper percentile value, from a range that includes a plurality of third predicted transaction volume values ​​included in a plurality of transaction volume scenarios. The information processing device described in Configuration Example 1. (Configuration Example 3) The processing unit creates the specified number of trading volume scenarios and the specified number of price scenarios. An information processing device as described in Configuration Example 1 or 2. (Configuration example 4) The processing unit obtains an empirical distribution of the error between the first predicted trading volume and the actual trading volume, and creates a plurality of trading volume scenarios including a plurality of third predicted trading volume values ​​obtained by adding a plurality of error amounts selected using the empirical distribution to the second predicted trading volume. An information processing device as described in any one of Configuration Examples 1 to 3. (Configuration example 5) The processing unit obtains an empirical distribution of the error between the first price forecast value and the actual price value, and creates a plurality of price scenarios including a plurality of third price forecast values ​​obtained by adding a plurality of error amounts selected using the empirical distribution to the second price forecast value. An information processing device as described in any one of Configuration Examples 1 to 4. (Configuration example 6) The processing unit calculates the bid amount by solving an optimization problem to find the price scenario among a plurality of price scenarios in which the bid amount is included in the bid range and which maximizes profit. An information processing device as described in any one of Configuration Examples 1 to 5. (Configuration example 7) The aforementioned optimization problem can be formulated using CVaR (Conditional Value at Risk) or VaR (Value at Risk). The information processing device described in Configuration Example 6. (Configuration example 8) The aforementioned processing unit, A trading volume scenario creation unit that creates the aforementioned trading volume scenario, A range calculation unit for calculating the aforementioned bid range, A price scenario creation unit that creates the aforementioned price scenario, A bid quantity calculation unit that calculates the aforementioned bid quantity, Equipped with, An information processing device as described in any one of Configuration Examples 1 to 7. (Configuration example 9) The subject of the transaction is electricity, The aforementioned trading volume is the amount of electricity generated. An information processing device as described in any one of Configuration Examples 1 to 8. (Configuration example 10) An information processing method performed by an information processing device, The steps include creating multiple trading volume scenarios, each including a third trading volume forecast, based on multiple trading volume performance data, each including a first trading volume forecast and actual trading volume corresponding to the first trading volume forecast for a trading target traded in multiple markets, including the first market, and a second trading volume forecast for the said trading target; A step of calculating a bid range that indicates the range of trading volume to be bid on for the first market based on multiple trading volume scenarios, The steps include creating a plurality of price scenarios, including a plurality of third price forecasts for the price, based on a plurality of price performance information, each including a first price forecast and an actual price corresponding to the first price forecast, and a second price forecast for the price; A step of calculating a bid amount representing the trading volume to be bid on for the first market based on the bid range and the multiple price scenarios, Information processing methods including (Configuration Example 11) On the computer, The steps include creating multiple trading volume scenarios, each including a third trading volume forecast, based on multiple trading volume performance data, each including a first trading volume forecast and actual trading volume corresponding to the first trading volume forecast for a trading target traded in multiple markets, including the first market, and a second trading volume forecast for the said trading target; A step of calculating a bid range that indicates the range of trading volume to be bid on for the first market based on multiple trading volume scenarios, The steps include creating a plurality of price scenarios, including a plurality of third price forecasts for the price, based on a plurality of price performance information, each including a first price forecast and an actual price corresponding to the first price forecast, and a second price forecast for the price; A step of calculating a bid amount representing the trading volume to be bid on for the first market based on the bid range and the multiple price scenarios, A program to execute. (Configuration Example 12) The system accepts input of multiple power generation performance data, each including a first power generation forecast value and an actual power generation value corresponding to the first power generation forecast value for a trading target traded in multiple markets, including the first market, and a second power generation forecast value for the said trading target. Based on the multiple data points of actual power generation and the second predicted power generation value, multiple power generation scenarios are created, including multiple third predicted power generation values. Based on risk information indicating the importance of buyback risk (buying back the trading target from a market other than the first market) and additional sell risk (selling the trading target to a market other than the first market), and multiple power generation scenarios, the bid amount representing the amount of power generation to be bid on in the first market is calculated. Processing section An information processing device equipped with the following features.

[0100] While several embodiments of the present invention have been described, these embodiments are presented as examples only and are not intended to limit the scope of the invention. These novel embodiments can be carried out in a variety of other forms, and various omissions, substitutions, and modifications can be made without departing from the spirit of the invention. These embodiments and their variations are included in the scope and spirit of the invention, as well as in the claims of the invention and its equivalents. [Explanation of Symbols]

[0101] 100, 100-2 Information Processing Device Rooms 101 and 101-2, Reception Desk 102 Power Generation Scenario Creation Department 103 Price Scenario Creation Department 104 Range Calculation Unit 105, 105-2 Bid Quantity Calculation Section 106 Output control unit 121 Storage section

Claims

1. Multiple trading volume scenarios are created, including multiple third trading volume forecasts, based on multiple trading volume historical data, each including a first trading volume forecast and actual trading volume corresponding to the first trading volume forecast for a trading target traded in multiple markets, including the first market, and a second trading volume forecast for the trading target predicted at a point in time after the multiple trading volume historical data is obtained. Based on the multiple trading volume scenarios, a bidding range is calculated that indicates the range of trading volumes to be bid on for the first market. Multiple price scenarios are created, including multiple third price forecasts for the price, based on a plurality of price performance information, each including a first price forecast value which is a predicted value of the price of the traded object in a plurality of the aforementioned markets, and the actual price corresponding to the first price forecast value, and a second price forecast value for the price. Based on the aforementioned bidding range and the multiple aforementioned price scenarios, a bid quantity representing the trading volume to be bid on for the first market is calculated. Processing section An information processing device equipped with the following features.

2. The processing unit calculates the bid range, using the lower percentile value representing the lower limit of the bid range as the lower limit value and the upper percentile value representing the upper limit of the bid range as the upper limit value, from among the range including the multiple third predicted transaction volume values ​​included in the multiple transaction volume scenarios. The information processing apparatus according to claim 1.

3. The processing unit creates the specified number of trading volume scenarios and the specified number of price scenarios. The information processing apparatus according to claim 1.

4. The processing unit obtains an empirical distribution of the error between the first predicted trading volume and the actual trading volume, and creates a plurality of trading volume scenarios including a plurality of third predicted trading volume values ​​obtained by adding a plurality of error amounts selected using the empirical distribution to the second predicted trading volume. The information processing apparatus according to claim 1.

5. The processing unit obtains an empirical distribution of the error between the first price prediction value and the actual price, and creates a plurality of price scenarios including a plurality of third price prediction values ​​obtained by adding a plurality of error amounts selected using the empirical distribution to the second price prediction value. The information processing apparatus according to claim 1.

6. The processing unit calculates the bid amount by solving an optimization problem to find the price scenario among a plurality of price scenarios in which the bid amount is included in the bid range and which maximizes profit. The information processing apparatus according to claim 1.

7. The aforementioned optimization problem is formulated using CVaR (Conditional Value at Risk) or VaR (Value at Risk). The information processing apparatus according to claim 6.

8. The aforementioned processing unit, A trading volume scenario creation unit that creates the aforementioned trading volume scenario, A range calculation unit for calculating the aforementioned bid range, A price scenario creation unit that creates the aforementioned price scenario, A bid quantity calculation unit that calculates the aforementioned bid quantity, Equipped with, The information processing apparatus according to claim 1.

9. The subject of the transaction is electricity, The aforementioned trading volume is the amount of electricity generated. The information processing apparatus according to claim 1.

10. An information processing method performed by an information processing device, The steps include creating multiple trading volume scenarios, each including a third trading volume forecast, based on a plurality of trading volume performance data, each including a first trading volume forecast and actual trading volume corresponding to the first trading volume forecast for a trading target traded in a plurality of markets, including a first market, and a second trading volume forecast for the trading target predicted at a point in time after the plurality of trading volume performance data has been obtained; A step of calculating a bid range that indicates the range of trading volume to be bid on for the first market based on multiple trading volume scenarios, The steps include creating a plurality of price scenarios, including a plurality of third price forecasts for the price, based on a plurality of price performance information, each including a first price forecast value which is a predicted value of the price of the traded object in a plurality of the aforementioned markets, and the actual price corresponding to the first price forecast value, and a second price forecast value for the price, A step of calculating a bid amount representing the trading volume to be bid on to the first market based on the bid range and the multiple price scenarios, Information processing methods including

11. On the computer, The steps include creating multiple trading volume scenarios, each including a third trading volume forecast, based on a plurality of trading volume performance data, each including a first trading volume forecast and actual trading volume corresponding to the first trading volume forecast for a trading target traded in a plurality of markets, including a first market, and a second trading volume forecast for the trading target predicted at a point in time after the plurality of trading volume performance data has been obtained; A step of calculating a bid range that indicates the range of trading volume to be bid on for the first market based on multiple trading volume scenarios, The steps include creating a plurality of price scenarios, including a plurality of third price forecasts for the price, based on a plurality of price performance information, each including a first price forecast value which is a predicted value of the price of the traded object in a plurality of the aforementioned markets, and the actual price corresponding to the first price forecast value, and a second price forecast value for the price, A step of calculating a bid amount representing the trading volume to be bid on to the first market based on the bid range and the multiple price scenarios, A program to execute.

12. Multiple trading volume scenarios are created, including multiple third trading volume forecasts, based on multiple trading volume historical data, each including a first trading volume forecast and actual trading volume corresponding to the first trading volume forecast for a trading target traded in multiple markets, including the first market, and a second trading volume forecast for the trading target predicted at a point in time after the multiple trading volume historical data is obtained. Based on risk information indicating the importance of buyback risk (buying back the trading target from a market other than the first market) and additional sell risk (selling the trading target to a market other than the first market), and multiple trading volume scenarios, the system calculates a bid amount representing the trading volume to be bid on in the first market. Processing section An information processing device equipped with the following features.

13. An information processing method performed by an information processing device, The steps include creating multiple trading volume scenarios, each including a third trading volume forecast, based on a plurality of trading volume performance data, each including a first trading volume forecast and actual trading volume corresponding to the first trading volume forecast for a trading target traded in a plurality of markets, including a first market, and a second trading volume forecast for the trading target predicted at a point in time after the plurality of trading volume performance data has been obtained; A step of calculating a bid amount representing the trading volume to be bid on in the first market, based on risk information indicating the importance of the buyback risk of buying back the trading target from a market other than the first market and the additional sell risk of selling the trading target to a market other than the first market, and a plurality of trading volume scenarios. Information processing methods including

14. A computer, The steps include creating multiple trading volume scenarios, each including a third trading volume forecast, based on a plurality of trading volume performance data, each including a first trading volume forecast and actual trading volume corresponding to the first trading volume forecast for a trading target traded in a plurality of markets, including a first market, and a second trading volume forecast for the trading target predicted at a point in time after the plurality of trading volume performance data has been obtained; A step of calculating a bid amount representing the trading volume to be bid on in the first market, based on risk information indicating the importance of the buyback risk of buying back the trading target from a market other than the first market and the additional sell risk of selling the trading target to a market other than the first market, and a plurality of trading volume scenarios. A program to execute.

Images