Electricity market price prediction system

Abstract

To provide an electricity market price prediction system capable of improving prediction accuracy while reducing the cost of predicting wholesale electricity market prices.SOLUTION: An electricity market price forecasting system 100, which forecasts wholesale electricity market prices for bidding in a wholesale electricity market, is provided with: a price forecasting unit for forecasting wholesale electricity market prices based on multiple explanatory variables (step S11); and a re-forecasting determination unit for determining whether or not to re-forecast the wholesale electricity market prices (steps S13 to S15) based on predetermined determination materials when there is a change in at least one of the explanatory variables during a period from the time of forecasting the wholesale electricity market prices (step S12) to the time of bidding in the wholesale electricity market (YES in step S16).SELECTED DRAWING: Figure 3

Description

[Technical field]

[0001] The present invention relates to a technology for an electricity market price forecasting system for forecasting wholesale electricity market prices. [Background technology]

[0002] Conventionally, an electricity market price prediction system for predicting wholesale electricity market prices has been publicly known, as described in Patent Document 1, for example.

[0003] Patent Document 1 describes an electricity market price prediction device that includes an acquisition unit that acquires an estimated supply amount of electricity generated by renewable energy (e.g., solar power generation) on a future prediction target date, and a prediction unit that predicts the electricity market price on the prediction target date based on the estimated supply amount of electricity. The electricity market price prediction device described in Patent Document 1 aims to improve the prediction accuracy of the electricity market price by the above-mentioned configuration.

[0004] On the other hand, if a weather forecast or the like changes between the time of predicting the electricity market price and the time of bidding in the electricity market, the predicted value of the estimated supply of electricity will also change according to the change in the weather forecast. Therefore, in order to maintain the prediction accuracy, it becomes necessary to re-predict the electricity market price. However, if the re-predicted result does not change much, the cost required for the re-prediction will be wasted. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] JP 2019-96164 A Summary of the Invention [Problem to be solved by the invention]

[0006] The present invention has been made in consideration of the above-mentioned circumstances, and the problem that the present invention aims to solve is to provide an electricity market price prediction system that can improve prediction accuracy while reducing the costs associated with predicting wholesale electricity market prices. [Means for solving the problem]

[0007] The problem to be solved by the present invention has been described above, and the means for solving this problem will now be described.

[0008] That is, claim 1 provides an electricity market price prediction system that predicts a wholesale electricity market price for bidding in the wholesale electricity market, and includes a price prediction unit that predicts the wholesale electricity market price based on a plurality of explanatory variables, and a re-prediction determination unit that, if there is a change in at least one of the plurality of explanatory variables between the time of predicting the wholesale electricity market price and the time of bidding in the wholesale electricity market, determines whether or not to re-predict the wholesale electricity market price based on predetermined judgment criteria.

[0009] In claim 2, the predetermined criteria include a contribution rate of the changed explanatory variable to the wholesale electricity market price.

[0010] In claim 3, the re-prediction determination unit calculates the contribution rate of the changed explanatory variable by multiple regression analysis.

[0011] In claim 4, the predetermined criteria include the amount of change in the explanatory variable that has changed.

[0012] In claim 5, the predetermined criteria include the electricity selling revenue and the cost of re-predicting the wholesale electricity market price.

[0013] In claim 6, the explanatory variables include at least one of temperature, solar radiation, wind speed, power demand, and fuel price. Effect of the Invention

[0014] The present invention has the following advantages.

[0015] According to claim 1, it is possible to reduce the cost of predicting wholesale electricity market prices and improve the prediction accuracy.

[0016] According to claim 2, the accuracy of determining whether or not the wholesale electricity market price should be re-predicted can be improved.

[0017] According to claim 3, the accuracy of determining whether or not the wholesale electricity market price should be re-predicted can be improved.

[0018] According to claim 4, the accuracy of determining whether or not the wholesale electricity market price should be re-predicted can be improved.

[0019] According to claim 5, the accuracy of determining whether or not the wholesale electricity market price should be re-predicted can be improved.

[0020] In claim 6, the prediction accuracy can be improved. [Brief description of the drawings]

[0021] [Figure 1] 1 is a block diagram showing a configuration of an electricity market price prediction system according to an embodiment of the present invention. [Diagram 2] A diagram showing the process leading up to bidding for wholesale electricity market prices. [Diagram 3] 4 is a flowchart showing prediction control for predicting wholesale electricity market prices. [Figure 4] A table showing an example of explanatory variables and regression coefficients. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0022] The following describes an electricity market price prediction system 100 according to one embodiment of the present invention.

[0023] An electricity market price prediction system 100 according to an embodiment of the present invention predicts an electricity market price in a wholesale electricity market (hereinafter referred to as a "wholesale electricity market price"). As shown in FIG. 1, the electricity market price prediction system 100 includes an acquisition unit 110 and a calculation unit 120.

[0024] The acquisition unit 110 acquires various types of information. The acquisition unit 110 can acquire information for predicting the wholesale electricity market price and the like via a network such as the Internet.

[0025] The calculation unit 120 performs various calculations. The calculation unit 120 can predict the wholesale electricity market price based on the information acquired by the acquisition unit 110.

[0026] The electricity market price prediction system 100 configured in this manner allows the user to predict the bidding curve (supply and demand curve) for bidding in the wholesale electricity market by obtaining the predicted results of the wholesale electricity market price up until the time of bidding in the wholesale electricity market.

[0027] The main wholesale electricity markets are the day-ahead market (spot market) and the intraday market (hour-ahead market). The day-ahead market is a market in which electricity is traded the day before actual supply and demand. In the day-ahead market, trading takes place for 48 products (48 frames) dividing a day into 30-minute increments. On the other hand, the intraday market is a market in which trading takes place up to one hour before actual supply and demand, and is mainly responsible for adjusting surpluses and shortages in the day-ahead market. In the following, the electricity market price prediction system 100 according to this embodiment will be described as predicting wholesale electricity market prices in the day-ahead market (spot market).

[0028] An example of the flow leading up to bidding in the wholesale electricity market (day-ahead market) will be explained below in chronological order with reference to FIG. 2. Details of each process will be described later. In the following, the day on which bidding is conducted in the day-ahead market will be referred to as the "same day", and the day of actual supply and demand (i.e., the day after the day on which bidding is conducted in the day-ahead market) will be referred to as the "next day". In the wholesale electricity market (day-ahead market), bidding must be conducted by 10:00 a.m. on the same day.

[0029] First, taking into account the time until bidding, for example at 6:00 a.m. on the day, the wholesale electricity market price [yen / kWh] for the next day is predicted. The predicted wholesale electricity market price is obtained for 48 frames, each dividing the day into 30-minute intervals. Then, at 9:00 a.m. on the day, the predicted value of the bidding curve is determined based on the predicted result of the wholesale electricity market price. Then, by 10:00 a.m. on the day, bidding in the wholesale electricity market (day-ahead market) is conducted based on the predicted value of the determined bidding curve.

[0030] Here, in some cases, after the forecast results of the wholesale electricity market price are output, the data for forecasting the wholesale electricity market price may change before the bidding curve is determined (9:00 a.m. on the day) due to a sudden shutdown of a power plant, the occurrence of a typhoon, a sudden deterioration in weather, etc. If the data changes, the forecast results of the wholesale electricity market price will also change, so essentially, if there is a change in the data, it is necessary to re-predict the wholesale electricity market price.

[0031] However, since re-predicting the wholesale electricity market price also requires costs, such as, for example, fees for outsourcing to external companies and system usage fees, if the wholesale electricity market price is re-predicted every time the data for the prediction changes, the costs for the re-prediction may accumulate and increase. Furthermore, if the results of the re-prediction do not change much even if the data for the prediction changes, the costs for the re-prediction may be wasted.

[0032] Therefore, when there is a change in the data for prediction, the electricity market price prediction system 100 of this embodiment does not immediately re-predict the wholesale electricity market price, but determines whether or not to re-predict the wholesale electricity market price, and then performs the re-prediction as necessary.

[0033] Specifically, the calculation unit 120 acquires a prediction result of the wholesale electricity market price before bidding in the wholesale electricity market (day-ahead market) by executing the prediction control shown in Fig. 3. A method of acquiring a prediction result of the wholesale electricity market price will be described below with reference to the flowchart shown in Fig. 3. In this embodiment, the prediction control shown in Fig. 3 is assumed to start at 6 a.m. on the day.

[0034] In step S11, the calculation unit 120 acquires variable data for predicting (used in) the wholesale electricity market price. Here, the "variable data" is a variable that contributes to the wholesale electricity market price. The variable data includes, for example, temperature, solar radiation, wind speed, electricity demand, fuel price, etc. The calculation unit 120 acquires external information such as electricity demand, weather information, power plant information, market information, fuel information, and exchange rate information via the acquisition unit 110, and acquires the variable data from the external information. The variable data is acquired every 48 frames, which is obtained by dividing a day into 30-minute units. The calculation unit 120 acquires the latest variable data as of 6:00 a.m. on the day via the acquisition unit 110.

[0035] After acquiring the variable data at 6 am on the current day, the calculation unit 120 periodically acquires the latest variable data. After performing the process of step S11, the calculation unit 120 proceeds to step S12.

[0036] In step S12, the calculation unit 120 predicts the wholesale electricity market price. The wholesale electricity market price here indicates the electricity market price (contract price) in the wholesale electricity market (day-ahead market). The calculation unit 120 predicts the wholesale electricity market price based on the variable data acquired in step S11.

[0037] Specifically, the calculation unit 120 obtains a prediction result of the wholesale electricity market price by multiple regression analysis using the wholesale electricity market price as the "objective variable" and the variable data obtained in step S11 as the "explanatory variables." The prediction result of the wholesale electricity market price is obtained for 48 frames obtained by dividing a day into 30-minute units.

[0038] Fig. 4 shows explanatory variables and regression coefficients of multiple regression analysis in one of the 48 frames. As shown in Fig. 4, a regression coefficient (regression coefficient of the changed explanatory variable in the equation of multiple regression analysis used in step S12) which is the coefficient of the explanatory variable is set for each explanatory variable (temperature, solar radiation, wind speed, electricity demand, fuel price, etc.). The regression coefficient indicates the contribution rate (weight) of each explanatory variable to the wholesale electricity market price (objective variable). Note that if the regression coefficient is positive, the wholesale electricity market price (objective variable) increases when the explanatory variable increases, and if the regression coefficient is negative, the wholesale electricity market price (objective variable) decreases when the explanatory variable increases.

[0039] In addition, the regression coefficients are set for each time period (48 frames). For example, the regression coefficient for temperature (explanatory variable) is set to be high in the early morning and during the day, and low at night and late at night.

[0040] After performing the process of step S12, the calculation unit 120 proceeds to step S13.

[0041] In step S13, the calculation unit 120 determines whether or not there has been a change in the variable data (explanatory variables) acquired in step S11 (at 6 a.m. on the current day). Specifically, the calculation unit 120 acquires the latest variable data and determines whether or not there has been a change in the variable data by comparing it with the variable data acquired at 6 a.m. on the current day. The calculation unit 120 determines that there has been a change in the variable data when there has been a change in one or more explanatory variables among the multiple explanatory variables.

[0042] It is preferable that the calculation unit 120 determines whether or not there has been a change in the variable data, taking into consideration the amount of change in the variable data (the amount of change in the latest variable data relative to the variable data acquired at 6 a.m. on the day). For example, the calculation unit 120 may determine that there has been a change in the variable data when the amount of change in the variable data is greater than a predetermined threshold value X, and may determine that there has been no change in the variable data when the amount of change in the variable data is equal to or less than the threshold value X. The threshold value X may be set to a different value for each explanatory variable.

[0043] If the calculation unit 120 determines that there is a change in the variable data (explanatory variables) ("YES" in step S13), the process proceeds to step S14. On the other hand, if the calculation unit 120 determines that there is no change in the variable data (explanatory variables) ("NO" in step S13), the process proceeds to step S16.

[0044] In step S14, the calculation unit 120 determines whether the contribution rate of the changed explanatory variable is high. Specifically, when the absolute value of the regression coefficient of the explanatory variable determined to have changed in step S13 is equal to or greater than a predetermined threshold Y, the calculation unit 120 determines that the contribution rate of the changed explanatory variable is high. On the other hand, when the absolute value of the regression coefficient of the changed explanatory variable is not equal to or greater than the threshold Y (is less than the threshold Y), the calculation unit 120 determines that the contribution rate of the changed explanatory variable is not high.

[0045] If the calculation unit 120 determines that the contribution rate of the changed explanatory variable is high ("YES" in step S14), the process proceeds to step S15. On the other hand, if the calculation unit 120 determines that the contribution rate of the changed explanatory variable is not high ("NO" in step S14), the process proceeds to step S16.

[0046] When the contribution rate of the changed explanatory variable is high (YES in step S14), it means that the fluctuation in the wholesale electricity market price (objective variable) accompanying the change in the explanatory variable is relatively large. On the other hand, when the contribution rate of the changed explanatory variable is not high (YES in step S14), it means that the fluctuation in the wholesale electricity market price (objective variable) accompanying the change in the explanatory variable is relatively small (i.e., the prediction result does not change much even if the wholesale electricity market price is predicted again).

[0047] In addition, when multiple explanatory variables have changed, the calculation unit 120 determines whether the contribution rate of the changed explanatory variables is high or not, taking into consideration the contribution rates of the changed explanatory variables. For example, the calculation unit 120 may make the determination in step S14 by comparing the sum of the contribution rates of the changed explanatory variables with a predetermined threshold value Z.

[0048] In step S15, the calculation unit 120 determines whether the predicted increase in the electricity selling revenue is less than the reforecast cost (predicted increase in the electricity selling revenue<reforecast cost). Here, the "predicted increase in the electricity selling revenue" refers to the value (increase) of the electricity selling revenue predicted to increase by reforecast. In addition, the "reforecast cost" refers to the cost required to reforecast the wholesale electricity market price. For example, if the time period of the changed explanatory variable is a time period (e.g., late night) when the bid amount (amount of electricity sold, amount of electricity purchased) is low, the predicted increase in the electricity selling revenue will be small.

[0049] If the calculation unit 120 determines that the predicted increase in electricity selling revenue is less than the re-predicted cost ("YES" in step S15), the calculation unit 120 proceeds to step S16. On the other hand, if the calculation unit 120 determines that the predicted increase in electricity selling revenue is not less than the re-predicted cost ("NO" in step S15), the calculation unit 120 returns the process to step S12. When the calculation unit 120 returns the process to step S12, the calculation unit 120 re-predicts the wholesale electricity market price using the latest variable data.

[0050] If the predicted increase in electricity selling revenue is less than the re-prediction cost (YES in step S15), it means that if the wholesale electricity market price is re-predicted, it is predicted that the electricity selling revenue will be offset (no profit will be obtained). On the other hand, if the predicted increase in electricity selling revenue is not less than the re-prediction cost (NO in step S15), it means that even if the wholesale electricity market price is re-predicted, it is predicted that electricity selling revenue will be obtained.

[0051] In step S16, the calculation unit 120 determines whether it is time (9:00 am) to determine the bid curve (supply and demand curve).

[0052] If the calculation unit 120 determines that it is time to determine the bid curve (9:00 a.m.) ("YES" in step S16), the process proceeds to step S17. On the other hand, if the calculation unit 120 determines that it is not yet time to determine the bid curve (9:00 a.m.) ("NO" in step S16), the process returns to step S13.

[0053] In step S17, the calculation unit 120 determines the predicted value of the bidding curve based on the prediction result of the wholesale electricity market price. Based on the determined bidding curve, the user can make a bid in the wholesale electricity market (day-ahead market) by 10:00 a.m. on the day. The actual value of the wholesale electricity market price in the wholesale electricity market is used to determine whether or not to re-predict from the next time onwards.

[0054] An example of whether or not the wholesale electricity market price can be re-predicted will be described below. For example, the temperature forecast value for the next day at 11pm has changed significantly, but since 11pm is a time period in which the contribution rate of temperature (explanatory variable) is low (NO in step S14) and the amount of bids in the wholesale electricity market is low (the predicted increase in electricity sales revenue is small, YES in step S15), the wholesale electricity market price is not re-predicted.

[0055] For example, the temperature forecast for 12 o'clock the following day has changed significantly, but since 12 o'clock is a time period in which the contribution rate of temperature (explanatory variable) is high (YES in step S14) and also a time period in which there is a large amount of bidding in the wholesale electricity market (large predicted increase in electricity sales revenue, NO in step S15), a re-prediction of the wholesale electricity market price is performed.

[0056] As described above, in the electricity market price prediction system 100 according to one embodiment of the present invention, a decision is made as to whether or not to re-predict the wholesale electricity market price, taking into consideration the amount of change in the explanatory variables (variable data for prediction) and the contribution rate of the explanatory variables between the time the wholesale electricity market price is predicted (6:00 a.m. on the day) and the time the bidding curve is determined (9:00 a.m. on the day). Then, only when it is determined that a cost benefit can be obtained, the wholesale electricity market price is re-predicted. This makes it possible to suppress the occurrence of unnecessary re-prediction costs, and ultimately to reduce the costs involved in prediction.

[0057] As described above, the electricity market price prediction system 100 according to this embodiment: 1. An electricity market price forecasting system 100 for forecasting wholesale electricity market prices for bidding in a wholesale electricity market, comprising: A price prediction unit (calculation unit 120) predicts the wholesale electricity market price based on a plurality of explanatory variables (step S11); a re-prediction determination unit (calculation unit 120) that, when at least one of the multiple explanatory variables has changed between the time of predicting the wholesale electricity market price (step S12) and the time of bidding in the wholesale electricity market (YES in step S16), determines whether or not to re-predict the wholesale electricity market price based on predetermined judgment criteria (steps S13 to S15); It is equipped with the following.

[0058] With this configuration, it is possible to reduce the cost of predicting wholesale electricity market prices while improving the prediction accuracy.

[0059] In addition, the predetermined criteria include: This includes the contribution rate of the changed explanatory variable to the wholesale electricity market price (step S14).

[0060] With this configuration, it is possible to improve the accuracy of determining whether or not to re-predict the wholesale electricity market price.

[0061] Further, the re-prediction determination unit The contribution rate of the changed explanatory variable is calculated by multiple regression analysis.

[0062] With this configuration, it is possible to improve the accuracy of determining whether or not to re-predict the wholesale electricity market price.

[0063] In addition, the predetermined criteria include: The amount of change in the explanatory variables that have changed (step S13) is included.

[0064] With this configuration, it is possible to improve the accuracy of determining whether or not to re-predict the wholesale electricity market price.

[0065] In addition, the predetermined criteria include: This includes the electricity sales revenue and the cost of re-predicting the wholesale electricity market price (step S15).

[0066] With this configuration, it is possible to improve the accuracy of determining whether or not to re-predict the wholesale electricity market price.

[0067] In addition, the explanatory variables include: At least one of temperature, solar radiation, wind speed, electricity demand, or fuel price is included.

[0068] With this configuration, it is possible to improve prediction accuracy.

[0069] Although the embodiment of the present invention has been described above, the present invention is not limited to the above configuration, and various modifications are possible within the scope of the invention described in the claims.

[0070] For example, in the present embodiment, the contribution rate of the explanatory variables is calculated by multiple regression analysis, but it may be calculated by other methods such as random forest.

[0071] Furthermore, in this embodiment, the bidding curve is determined by the calculation unit 120, but the bidding curve may be determined by a user based on the results of a prediction of the wholesale electricity market price.

[0072] In addition, in this embodiment, a re-prediction of the wholesale electricity market price is performed when the contribution rate of the changed explanatory variable is high (YES in step S14) and the predicted increase in electricity sales revenue is equal to or greater than the re-prediction cost (NO in step S15). However, a re-prediction of the wholesale electricity market price may also be performed when either one of the following conditions is met: when the contribution rate of the changed explanatory variable is high (YES in step S14) or when the predicted increase in electricity sales revenue is equal to or greater than the re-prediction cost (NO in step S15).

[0073] For example, if the temperature forecast for 6 p.m. the following day changes significantly, 6 p.m. is a time period in which the contribution rate of temperature (explanatory variable) is low (NO in step S14), and since this is a time period in which there is a large amount of bidding in the wholesale electricity market (large predicted increase in electricity sales revenue, NO in step S15), a re-prediction of the wholesale electricity market price may be performed.

[0074] For example, if the temperature forecast for 6:00 the next day changes significantly, 6:00 is a time period with low bidding volume in the wholesale electricity market (low predicted increase in electricity sales revenue, YES in step S15), but it is also a time period with a high contribution rate of temperature (explanatory variable) (YES in step S14), so a re-prediction of the wholesale electricity market price may be performed.

[0075] Furthermore, the processing performed by the calculation unit 120 does not necessarily have to be performed by the calculation unit 120 alone, but may be performed in cooperation with (contract with) a calculation unit of another system, for example. [Explanation of symbols]

[0076] 100 Electricity Market Price Prediction System 120 Arithmetic section

Claims

1. 1. An electricity market price prediction system for predicting a wholesale electricity market price for bidding in a wholesale electricity market, comprising: A price prediction unit that predicts the wholesale electricity market price based on a plurality of explanatory variables; a re-prediction determination unit that, when there is a change in at least one of the plurality of explanatory variables between the time of predicting the wholesale electricity market price and the time of bidding in the wholesale electricity market, determines whether or not to re-predict the wholesale electricity market price based on a predetermined judgment material; Equipped with Electricity market price forecasting system.

2. The predetermined criteria for judgment include: The contribution rate of the changed explanatory variable to the wholesale electricity market price is included. The electricity market price forecasting system of claim 1 .

3. The re-prediction determination unit Calculating the contribution rate of the changed explanatory variable by multiple regression analysis; The electricity market price forecasting system according to claim 2 .

4. The predetermined criteria for judgment include: The amount of change in the explanatory variable that has been changed is included. The electricity market price forecasting system of claim 1 .

5. The predetermined criteria for judgment include: The cost of re-forecasting the electricity sales revenue and the wholesale electricity market price is included. The electricity market price forecasting system of claim 1 .

6. The explanatory variables include: At least one of temperature, solar radiation, wind speed, power demand, or fuel price is included; The electricity market price forecasting system of claim 1 .

Images