Electric power market price prediction system

Abstract

To provide an electric power market price prediction system capable of accurately predicting an electric power market price in response to rapid price fluctuations.SOLUTION: An electric power market price prediction system 100 is provided with a first acquisition unit (computing unit 120) for acquiring a first prediction value of an electric power market price for each predetermined time slot (for example, 30 minutes) using a first prediction model created based on a first explanatory variable including a past actual value of the electric power market price (Steps 11 to 13), a difference acquisition unit (computing unit 120) which acquires a difference between a predetermined reference value set based on the actual value for each predetermined time slot and the first prediction value for each predetermined time slot (Step 14), and a second acquisition unit (computing unit 120) for acquiring a second prediction value of the electric power market price corrected from the first prediction value using a second prediction model created based on a second explanatory variable in which a variable related to the difference is added to the first explanatory variable (Steps 15 to 17).SELECTED DRAWING: Figure 2

Description

[Technical Field]

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

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

[0003] Patent Document 1 discloses an electricity market price prediction device that predicts the electricity market price based on an estimated amount of electricity supplied by renewable energy on a target prediction date, actual value information on the amount of electricity supplied, and actual value information on the electricity market price. In this way, the technology described in Patent Document 1 aims to improve the accuracy of electricity market price prediction by predicting the estimated amount of electricity supplied and the amount of electricity demanded, which have a large impact on the electricity market price. By improving the accuracy of electricity market price prediction, it is possible to improve the revenue from selling electricity.

[0004] Here, if a sudden event such as a breakdown in power generation facilities occurs, the event may cause a sudden fluctuation in the electricity market price. Patent Document 1 describes that, in order to reduce errors in the prediction of the electricity market price, data caused by a sudden price fluctuation is excluded from the actual value data of the amount of electricity supplied and the actual value data of the electricity market price.

[0005] However, the technology described in Patent Document 1 does not determine whether or not a sudden price fluctuation will occur in the electricity market price on the target prediction date, and is therefore unable to accurately predict the electricity market price in response to sudden price fluctuations. [Prior art documents] [Patent documents]

[0006] [Patent Document 1] Japanese Patent Application Publication No. 2019-96164 Summary of the Invention [Problem to be solved by the invention]

[0007] The present invention has been made in consideration of the above-mentioned circumstances, and the problem that it aims to solve is to provide an electricity market price prediction system that can accurately predict electricity market prices in response to sudden price fluctuations. [Means for solving the problem]

[0008] The problem to be solved by the present invention is as described above, and the means for solving this problem will now be described.

[0009] That is, claim 1 provides an electricity market price prediction system for predicting electricity market prices, comprising: a first acquisition unit that acquires a first predicted value of the electricity market price for each predetermined time period using a first prediction model created based on a first explanatory variable including a past actual value of the electricity market price; a difference acquisition unit that acquires a difference for each predetermined time period between a predetermined reference value set based on the actual value for each predetermined time period and the first predicted value; and a second acquisition unit that acquires a second predicted value of the electricity market price by correcting the first predicted value using a second prediction model created based on a second explanatory variable in which a variable related to the difference is added to the first explanatory variable.

[0010] In claim 2, the predetermined reference value is a value indicating an average of the performance values.

[0011] In claim 3, the predetermined reference value is a value indicating an average of the performance values ​​excluding abnormal values ​​that cause abrupt price fluctuations from the performance values.

[0012] In claim 4, the predetermined reference value is set using, of the performance values, performance values ​​acquired over a predetermined period within the most recent 14 days.

[0013] In claim 5, the variable relating to the difference is a dummy variable indicating that the magnitude of the difference is greater than a predetermined threshold value.

[0014] In claim 6, the system further comprises a prediction model creation unit that creates the second prediction model. [Effects of the Invention]

[0015] The present invention has the following effects.

[0016] According to the invention of claim 1, the electricity market price can be predicted with high accuracy in response to sudden price fluctuations.

[0017] In the invention according to claim 2, it is possible to improve the reliability of the reference value for a price range with small price fluctuations.

[0018] In the invention according to claim 3, the reliability of the reference value can be further improved.

[0019] In the invention according to claim 4, the reliability of the reference value can be further improved.

[0020] In the invention according to claim 5, the possibility of a sudden price fluctuation in the electricity market price can be easily reflected in the second predicted value.

[0021] In the invention according to claim 6, it is possible to easily obtain the second predicted value of the electricity market price. [Brief explanation of the drawings]

[0022] [Figure 1] 1 is a block diagram showing the configuration of an electricity market price prediction system according to an embodiment of the present invention. [Figure 2] 1 is a flowchart illustrating predictive control for predicting electricity market prices. [Figure 3] 10 is a graph showing an example of the transition of the first predicted value over one day. [Figure 4] 10 is a graph showing an example of the transition of the first predicted value and the actual value over one day. [Figure 5] 10 is a graph showing an example of the transition of the second predicted value (a value obtained by correcting the first predicted value) over one day. DETAILED DESCRIPTION OF THE INVENTION

[0023] An electricity market price prediction system 100 according to one embodiment of the present invention will be described below.

[0024] An electricity market price prediction system 100 according to one embodiment of the present invention predicts electricity market prices in the wholesale electricity market (hereinafter simply referred to as "electricity market prices"). As shown in Fig. 1, the electricity market price prediction system 100 includes an acquisition unit 110 and a calculation unit 120.

[0025] The acquisition unit 110 acquires various types of information. The acquisition unit 110 can acquire information for predicting the electricity market price via a network such as the Internet.

[0026] The calculation unit 120 performs various calculations and can predict the electricity market price based on the information acquired by the acquisition unit 110.

[0027] The electricity market price prediction system 100 configured in this manner obtains the predicted electricity market price results before bidding in the wholesale electricity market, allowing users to predict the bidding curve (supply and demand curve) for bidding in the wholesale electricity market.

[0028] Here, the electricity market price may suddenly fluctuate due to unexpected events such as the breakdown of power generation facilities, etc. It is very difficult to accurately predict the electricity market price in a price range with large fluctuations and the electricity market price in a price range with small fluctuations in a single process.

[0029] Therefore, in the electricity market price prediction system 100 according to this embodiment, the calculation unit 120 predicts the electricity market price by executing the prediction control shown in Fig. 2. As will be described in detail later, in the prediction control shown in Fig. 2, the calculation unit 120 performs a first-stage prediction in the processing from steps S11 to S13, and performs a second-stage prediction (correction of the first-stage predicted value) in the processing from steps S14 to S17. A method for predicting the electricity market price will be described below using the flowchart shown in Fig. 2.

[0030] In step S11, the calculation unit 120 acquires, via the acquisition unit 110, variable data of a first explanatory variable for predicting the electricity market price. Here, the "first explanatory variable" is a variable that contributes to the electricity market price. Examples of the first explanatory variable include temperature, solar radiation, wind speed, electricity demand, and fuel price. 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 variable data of the first explanatory variable from the external information. The first explanatory variable is acquired for every 48 frames obtained by dividing a day into 30-minute increments.

[0031] After performing the process of step S11, the calculation unit 120 proceeds to step S12.

[0032] In step S12, the calculation unit 120 creates a first prediction model based on the first explanatory variable.

[0033] Specifically, the calculation unit 120 creates a first prediction model of multiple regression analysis with the electricity market price as the "objective variable" and the first explanatory variable acquired in step S11 as the "explanatory variable." A regression coefficient, 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 electricity market price (objective variable). If the regression coefficient is positive, an increase in the explanatory variable will increase the electricity market price (objective variable), and if the regression coefficient is negative, an increase in the explanatory variable will decrease the electricity market price (objective variable).

[0034] After performing the process of step S12, the calculation unit 120 proceeds to step S13.

[0035] In step S13, the calculation unit 120 uses the first prediction model to obtain a predicted value (first predicted value) of the electricity market price. The first predicted value is obtained for 48 frames obtained by dividing a day into 30-minute units.

[0036] FIG. 3 is a graph showing an example of the transition of the first predicted value over one day (March 19th in FIG. 3). The horizontal axis in FIG. 3 indicates time, and the vertical axis indicates the electricity market price [yen / kWh]. The same applies to the horizontal and vertical axes in FIG. 4 and FIG. 5, which will be described later. Note that although the predicted value of the electricity market price is shown as a line in FIG. 3, FIG. 4, and FIG. 5, in reality it is obtained as a value every 30 minutes. As shown in FIG. 3, the first predicted value fluctuates depending on the time of day (time).

[0037] In this way, the processing from steps S11 to S13 provides a first-stage predicted value (first predicted value) of the electricity market price for each predetermined time period (30 minutes).

[0038] After performing the process of step S13, the calculation unit 120 proceeds to step S14.

[0039] In step S14, the calculation unit 120 obtains (calculates) the difference between the reference value of the electricity market price and the first predicted value obtained in step S13 for each predetermined time period (30 minutes). Here, the "reference value" is a value set based on the past actual value of the electricity market price (past electricity market price). The difference is calculated by subtracting the reference value from the first predicted value. Therefore, if the first predicted value is larger than the reference value, the difference will be a positive value. On the other hand, if the first predicted value is smaller than the reference value, the difference will be a negative value.

[0040] FIG. 4 is a graph showing an example of the transition of the first forecast value and the actual value on one day (March 19th). Here, the reference value at a given time t is preferably set to a value indicating the average of multiple actual values ​​at time t. Furthermore, the reference value at time t is preferably set to a value indicating the average of multiple actual values ​​at time t, excluding abnormal values ​​(the portion of the electricity market price that is subject to the sudden price fluctuation) that cause a sudden price fluctuation. Any method can be used to determine whether a sudden price fluctuation has occurred in the actual value of the electricity market price. For example, the calculation unit 120 can determine that a sudden price fluctuation has occurred in the actual value of the electricity market price if the standard deviation of the actual value for each time period is outside a predetermined range. Furthermore, the reference value is preferably set using actual values ​​acquired over a predetermined period (e.g., seven days) within the last 14 days.

[0041] After performing the process of step S14, the calculation unit 120 proceeds to step S15.

[0042] In step S15, the calculation unit 120 assigns a flag to the time period (time t) in which the magnitude (absolute value) of the difference acquired in step S14 is equal to or greater than a predetermined threshold. Here, the "time period in which the magnitude of the difference is equal to or greater than a predetermined threshold" refers to a time period in which the first predicted value significantly deviates from the reference value, indicating that a sudden price fluctuation is likely to occur. The threshold is set to, for example, 5 yen / kWh.

[0043] If there is a time period for which the difference calculated in step S14 is plus 5 yen / kWh or more, calculation unit 120 assigns a plus flag to that time period. On the other hand, if there is a time period for which the difference calculated in step S14 is minus 5 yen / kWh or less, calculation unit 120 assigns a minus flag to that time period.

[0044] In the example shown in Figure 4, the first predicted value of the electricity market price is generally lower than the actual value. Furthermore, from 7:30 to 11:30 and 12:30, the difference between the reference value of the electricity market price and the first predicted value is 5 yen / kWh or more. Therefore, a negative flag is assigned to the periods from 7:30 to 11:30 and 12:30.

[0045] After performing the process of step S15, the calculation unit 120 proceeds to step S16.

[0046] In step S16, the calculation unit 120 creates a second prediction model based on the second explanatory variable. Here, the "second explanatory variable" is the first explanatory variable to which a variable related to the difference obtained in step S14 has been added.

[0047] Here, the "variable related to the difference" is a dummy variable that indicates that the magnitude of the difference between the reference value of the electricity market price and the first predicted value is greater than a predetermined threshold (5 yen / kWh). For example, the dummy variable is set to "0" for a time period with no flag attached, "1" for a time period with a positive flag attached, and "-1" for a time period with a negative flag attached.

[0048] The processing of step S16 will be specifically described. The calculation unit 120 creates a second prediction model of multiple regression analysis with the electricity market price as the "objective variable" and the second explanatory variable as the "explanatory variable." The regression coefficients of each explanatory variable (temperature, solar radiation, wind speed, electricity demand, fuel price, etc.) other than the dummy variable are set to the same values ​​as the regression coefficients in the first prediction model. In addition, the regression coefficients of the dummy variables (variables related to differences) are determined based on the actual values ​​of abnormal values ​​(electricity market prices in the part where price fluctuates abruptly) that cause sudden price fluctuations.

[0049] After performing the process of step S16, the calculation unit 120 proceeds to step S17.

[0050] In step S17, the calculation unit 120 uses the second prediction model to obtain a predicted value (second predicted value) of the electricity market price by correcting the first predicted value obtained in step S13. The second predicted value is obtained for 48 frames obtained by dividing a day into 30-minute intervals.

[0051] Fig. 5 is a graph showing an example of the transition of the second predicted value over one day. In the example shown in Fig. 5, the first predicted value for each time slot shown in Fig. 3 is corrected to the second predicted value. Furthermore, the second predicted values ​​for the time slots from 7:30 to 11:30 and 12:30, which are flagged in step S15 (i.e., the magnitude of the difference between the reference value of the electricity market price and the first predicted value is 5 yen / kWh or more), are calculated using the second prediction model created in step S16 and are set to values ​​close to 0 yen / kWh.

[0052] After performing the process of step S17, the calculation unit 120 ends the predictive control shown in FIG.

[0053] As described above, in the electricity market price prediction system 100 according to this embodiment, a predicted value of the electricity market price (first predicted value) is first obtained using a first prediction model created based on a first explanatory variable. Even at this stage, the first predicted value of the electricity market price has a certain degree of accuracy. However, the first predicted value has a problem in terms of the prediction accuracy of the electricity market price in the portion where the price fluctuates suddenly.

[0054] Therefore, after obtaining the first predicted value, the electricity market price prediction system 100 calculates the difference between the first predicted value and the actual value of the electricity market price, and determines that there is a high possibility of a sudden price fluctuation for a time period in which the magnitude of the difference is greater than a predetermined threshold.The electricity market price prediction system 100 then re-predicts (corrects) the electricity market price for a time period in which the magnitude of the difference is greater than the predetermined threshold (a time period in which a flag is set) using a second prediction model created based on a second explanatory variable including a variable related to the difference.By predicting the electricity market price in two stages in this way, it is possible to accurately predict the electricity market price in response to sudden price fluctuations.

[0055] Furthermore, when calculating the difference, the actual value of the electricity market price, which is the object of comparison with the first predicted value, is set to the average value excluding the electricity market price in the portion with sudden price fluctuations, thereby improving the reliability of the actual value in the price range with small price fluctuations, and ultimately improving the accuracy of determining the possibility of sudden price fluctuations.

[0056] Furthermore, if a value from too long ago is used as the actual value of the electricity market price, the value may not match the current market price of the electricity market. For this reason, the electricity market price prediction system 100 according to this embodiment uses actual values ​​from a predetermined period within the most recent 14 days, preferably the most recent 7 days, as the actual value of the electricity market price. In this way, by using actual values ​​of the electricity market price that match the current market price of the electricity market, it is possible to improve the accuracy of determining the possibility of sudden price fluctuations.

[0057] As described above, the electricity market price prediction system 100 according to this embodiment: An electricity market price forecasting system 100 for forecasting an electricity market price, a first acquisition unit (calculation unit 120) that acquires a first predicted value of the electricity market price for each predetermined time period (e.g., 30 minutes) using a first prediction model created based on a first explanatory variable including a past actual value of the electricity market price (steps S11 to S13); a difference acquisition unit (calculation unit 120) that acquires a difference between a predetermined reference value set based on the actual value for each predetermined time period and the first predicted value for each predetermined time period (step S14); a second acquisition unit (calculation unit 120) that acquires a second predicted value of the electricity market price by correcting the first predicted value using a second prediction model created based on second explanatory variables in which a variable related to the difference is added to the first explanatory variables (steps S15 to S17); It is equipped with the following.

[0058] With this configuration, it is possible to accurately predict the electricity market price in response to sudden price fluctuations. Specifically, by obtaining a first forecast value for each predetermined time period and obtaining the difference between a reference value set based on actual values ​​and the first forecast value for each predetermined time period, it is possible to identify time periods during which sudden price fluctuations are likely to occur.Then, for time periods during which sudden price fluctuations are likely to occur, the first forecast value is corrected using a forecast model (second forecast model) to which a variable related to the difference is added, thereby improving the forecast accuracy of the electricity market price during time periods during which sudden price fluctuations will occur.

[0059] Moreover, the predetermined reference value is This is a value that indicates the average of the actual results.

[0060] This configuration can improve the reliability of the reference values ​​for price ranges with small price fluctuations, and can also improve the accuracy of determining the possibility of sudden price fluctuations.

[0061] Moreover, the predetermined reference value is It is a value that indicates the average of the performance values ​​excluding abnormal values ​​that cause sudden price fluctuations from the performance values.

[0062] This configuration can further improve the reliability of the reference value, and ultimately improve the accuracy of determining the possibility of sudden price fluctuations.

[0063] Moreover, the predetermined reference value is The performance value is set using performance values ​​acquired over a predetermined period within the most recent 14 days.

[0064] This configuration can further improve the reliability of the reference value, and ultimately improve the accuracy of determining the possibility of sudden price fluctuations.

[0065] The variable relating to the difference is This is a dummy variable that indicates that the magnitude of the difference is greater than a predetermined threshold value.

[0066] With this configuration, the possibility of sudden price fluctuations in the electricity market price can be easily reflected in the second predicted value.

[0067] The electricity market price prediction system 100 according to this embodiment also includes a prediction model creation unit (calculation unit 120) that creates the second prediction model.

[0068] This configuration makes it possible to easily obtain the second predicted value of the electricity market price.

[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 this 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, 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 (outsourced to) a calculation unit of another system, for example. [Explanation of symbols]

[0072] 100 Electricity Market Price Forecasting System 120 Arithmetic section

Claims

1. An electricity market price prediction system for predicting an electricity market price, a first acquisition unit that acquires a first predicted value of the electricity market price for each predetermined time period using a first prediction model created based on a first explanatory variable including a past actual value of the electricity market price; a difference acquisition unit that acquires a difference between a predetermined reference value set based on the actual value for each predetermined time period and the first predicted value for each predetermined time period; a second acquisition unit that acquires a second predicted value of the electricity market price by correcting the first predicted value using a second prediction model created based on second explanatory variables in which a variable related to the difference is added to the first explanatory variables; and Equipped with Electricity market price forecasting system.

2. The predetermined reference value is A value indicating the average of the actual results, The electricity market price forecasting system of claim 1 .

3. The predetermined reference value is A value indicating the average of the actual values ​​excluding abnormal values ​​that cause sudden price fluctuations from the actual values. The electricity market price forecasting system according to claim 2 .

4. The predetermined reference value is The performance data is set using performance data acquired over a predetermined period within the most recent 14 days. The electricity market price forecasting system according to claim 2 .

5. The variable relating to the difference is is a dummy variable indicating that the magnitude of the difference is greater than a predetermined threshold. The electricity market price forecasting system of claim 1 .

6. a prediction model creation unit that creates the second prediction model; The electricity market price forecasting system according to any one of claims 1 to 5.

Images