Electricity Market Price Forecasting System

Abstract

An electricity market price forecasting system is provided that can improve the accuracy of long-term electricity market price forecasts. [Solution] An electricity market price prediction system 100 that predicts electricity market prices includes a first acquisition unit that acquires actual price values ​​(average actual frame price values), which are past electricity market prices for each specified time period, and an all-time period actual average value, which is the average value of the actual price values ​​(average actual frame price values) for all time periods (step S11); a second acquisition unit that acquires an average value ratio, which is the ratio of the actual price values ​​(average actual frame price values) to the all-time period actual average value, for each specified time period (step S11); and a third acquisition unit that multiplies the average value ratio for each specified time period acquired by the second acquisition unit by the electricity futures price, thereby acquiring an electricity market price prediction value for each specified time period (step S12).

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, 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 a prediction system that predicts the electricity market price based on an estimated amount of electricity supplied by renewable energy, actual value information on the amount of electricity supplied, and actual value information on the electricity market price. Such predicted electricity market price values ​​are used for transactions in the wholesale electricity market.

[0004] Conventional forecasting systems provide forecasts of electricity market prices every 30 minutes (48 periods per day) from the next day up to approximately two weeks in advance, but only provide monthly average prices for long-term electricity market price forecasts, such as one month from now. Trading electricity for one month or more in advance is effective for hedging against risks such as a surge in electricity market prices. Furthermore, since it is possible to trade electricity for specified time periods even in electricity trading for one month or more in advance, it is possible to obtain forecasts of electricity market prices for each time period, in addition to monthly average prices. As such, it is desirable to improve the accuracy of long-term electricity market price forecasts. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] Japanese Patent Application Publication No. 2019-96164 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 the accuracy of long-term electricity market price predictions. [Means for solving the problem]

[0007] The problem to be solved by the present invention is as 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 for predicting electricity market prices, comprising: a first acquisition unit that acquires actual price values, which are past electricity market prices for each specified time period, and an all-time period actual average value, which is the average value of the actual price values ​​over all time periods; a second acquisition unit that acquires an average value ratio, which is the ratio of the actual price value to the all-time period actual average value, for each specified time period; and a third acquisition unit that acquires an electricity market price prediction value for each specified time period by multiplying the average value ratio for each specified time period acquired by the second acquisition unit by the electricity futures price.

[0009] In claim 2, the first acquisition unit uses an average value of the actual price values ​​in each time period as the actual price value.

[0010] In claim 3, the first acquisition unit acquires the actual price value based on calendar information.

[0011] In claim 4, the electricity futures price includes a base load price, and the third acquisition unit acquires the electricity market price forecast value by multiplying the average value ratio by the base load price.

[0012] In claim 5, the electricity futures prices include prices limited to a time period, and the third acquisition unit acquires the electricity market price forecast value for a target time period of the prices limited to a time period by multiplying the average value ratio by the price limited to the time period, and acquires the electricity market price forecast value for a time period other than the target time period by multiplying the average value ratio by the base load price.

[0013] According to claim 6, the power generation system further comprises a correction unit that corrects the predicted value of the power market price using the futures price of fuel for power generation. [Effects of the Invention]

[0014] The present invention has the following effects.

[0015] According to claim 1, the accuracy of long-term electricity market price prediction can be improved.

[0016] According to claim 2, the accuracy of the prediction of the electricity market price can be further improved.

[0017] According to claim 3, the accuracy of the prediction of the electricity market price can be further improved.

[0018] According to claim 4, the accuracy of the prediction of the electricity market price can be further improved.

[0019] According to claim 5, the accuracy of the prediction of the electricity market price can be further improved.

[0020] According to claim 6, the accuracy of the prediction of the electricity market price can be further improved. [Brief explanation of the drawings]

[0021] [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 trend in average piece price performance data. [Figure 4] 10 is a graph showing an example of the trend in the ratio (average value ratio) of the average value of actual unit prices to the average value of actual prices for all time periods. [Figure 5] 1 is a graph showing an example of a change in predicted electricity market price based on electricity futures prices. [Figure 6] 10 is a graph showing an example of the trend of predicted electricity market prices after correction based on LNG futures prices. DETAILED DESCRIPTION OF THE INVENTION

[0022] An electricity market price prediction system 100 according to one embodiment of the present invention will be described below.

[0023] 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.

[0024] 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. The information for predicting the electricity market price includes, for example, electricity futures prices and LNG futures prices, which will be described later.

[0025] The calculation unit 120 performs various calculations and can predict the 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 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.

[0027] Conventional forecasting systems provide forecasts of electricity market prices every 30 minutes (48 periods per day) from the next day up to about two weeks in advance, but only provide monthly average prices for long-term electricity market price forecasts, such as one month from now.Even for electricity trading one month from now, it is possible to trade electricity for specified time periods, so it is possible to obtain forecasts of electricity market prices for each time period, not just monthly average prices.

[0028] 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. Hereinafter, the method for predicting the electricity market price will be described with reference to the flowchart shown in Fig. 2. In this embodiment, the electricity market price is predicted for each month from one month ahead to one year ahead.

[0029] In step S11, the calculation unit 120 creates a model curve based on the actual values ​​of the past electricity market price (JPEX price). Specifically, the calculation unit 120 acquires the actual values ​​of the electricity market price every 30 minutes (48 frames) for the past several years. Hereinafter, the actual values ​​of the past electricity market price every 30 minutes (each frame) will also be referred to as the "actual frame price value." The calculation unit 120 classifies the actual frame price values ​​for the past several years by month. Then, the calculation unit 120 calculates the average value of the actual frame price values ​​for the 48 frames (each frame from 0:00 to 23:30) based on the calendar information. Specifically, the calculation unit 120 calculates the average value of the actual frame price values ​​for the 48 frames for each weekday, Saturday, and Sunday and public holiday (Sunday and public holiday) for each month from January to December. Hereinafter, the average value of the actual frame price values ​​for each frame will also be referred to as the "average actual frame price value."

[0030] Figure 3 shows an example of the trend of average frame price performance in a certain month. As shown in Figure 3, average frame price performance is calculated for weekdays, Saturdays, and Sundays and holidays.

[0031] Next, the calculation unit 120 calculates the average value of the average frame price performance over the entire time period (the average value of the average frame price performance for 48 frames) for each weekday, Saturday, and Sunday and public holiday (Sunday and public holiday) for each month. Hereinafter, the average value of the average frame price performance over the entire time period may also be referred to as the "average frame price performance over the entire time period." Then, the calculation unit 120 calculates the ratio of the average frame price performance over the entire time period average (hereinafter referred to as the "average price ratio") by dividing the average frame price performance over the entire time period average for each frame.

[0032] FIG. 4 shows an example of the trend in the average price ratio for a certain month. As shown in FIG. 4, the average price ratio is calculated for each weekday, Saturday, Sunday, and public holiday (Sunday and public holiday) for each month. In this way, the calculation unit 120 creates a model curve that shows the average price ratio for each weekday, Saturday, Sunday, and public holiday for each month. The model curve can represent the trend in price changes for the 48 frames for each month.

[0033] After performing the process of step S11, the calculation unit 120 proceeds to step S12.

[0034] In step S12, the calculation unit 120 acquires a predicted electricity market price value (predicted value of the electricity market price) based on the model curve and the electricity futures price. Specifically, the calculation unit 120 can calculate the predicted electricity market price value by multiplying the average value ratio (model curve) calculated in step S11 by the electricity futures price.

[0035] Here, "electricity futures price" refers to the price of electricity used in electricity futures trading. Electricity futures trading is a financial transaction in which electric power companies and other entities buy and sell future electricity at a predetermined electricity futures price. By determining the electricity futures price, electricity retailers can avoid the risk of unexpected losses due to price fluctuations in the electricity market. Electricity futures prices are used for electricity trading one month or more in the future. Electricity futures prices are updated monthly based on fuel costs, etc.

[0036] In this embodiment, the electricity futures price for each month from one month to one year ahead is used as the electricity futures price, and the calculation unit 120 thereby calculates the electricity market price forecast value for each month from one month to one year ahead.

[0037] Figure 5 shows an example of the trend of the predicted electricity market price for a certain month. As shown in Figure 5, the predicted electricity market price is calculated for each weekday, Saturday, and Sunday and public holiday in each month.

[0038] There are several types of electricity futures prices, and at the Tokyo Exchange (TOCOM), there are two types: base load price (base price, 24-hour average for all days) and daytime load price (middle price, average from 8:00 to 20:00 on weekdays). The "daytime load price" is an example of the "price limited to a time period" of the present invention. The calculation unit 120 calculates the electricity market price forecast value by multiplying the average value ratio (model curve) calculated in step S11 by the baseload price for all time periods on Saturdays, Sundays, and holidays, and from 20:00 to 20:00 on weekdays. On the other hand, the calculation unit 120 calculates the electricity market price forecast value by multiplying the average value ratio (model curve) calculated in step S11 by the daytime load price for the period from 8:00 to 20:00 on weekdays.

[0039] After performing the process of step S12, the calculation unit 120 proceeds to step S13.

[0040] In step S13, the calculation unit 120 corrects the electricity market price forecast value acquired in step S12 based on the LNG futures price. Here, the LNG futures price is the price of liquefied natural gas (LNG), which is a fuel for thermal power generation, in futures trading.

[0041] It is known that the LNG futures price is correlated with the electricity market price, especially late at night (between 0 and 6 a.m.). Therefore, the calculation unit 120 corrects the electricity market price forecast value acquired in step S12 using the LNG futures price.

[0042] Specifically, the calculation unit 120 converts the LNG futures price into an LNG power generation cost (the cost required to generate electricity using liquefied natural gas as fuel) taking into consideration the power generation efficiency of liquefied natural gas, exchange rates, etc. The LNG power generation cost is the average power generation cost in the power supply area. The LNG power generation cost is calculated using the following formula 1.

[0043] LNG power generation cost = LNG futures price × α + β (Equation 1)

[0044] Here, "α" is a coefficient indicating the power generation efficiency, and "β" is a predetermined value (intercept) set based on the cost of power generation, etc. For example, α=1 / 160, β=1.5.

[0045] Next, the calculation unit 120 subtracts the difference between the LNG power generation cost and the average value of the electricity market price forecast value for all time periods from the electricity market price forecast value. That is, the corrected electricity market price forecast value (corrected price forecast value) is calculated using the following formula 2. Note that the following formula 2 is an equation obtained empirically by the inventors of the present application.

[0046] Corrected price forecast value = Electricity market price forecast value - (LNG power generation cost - All-time forecast average value) (Equation 2)

[0047] Here, the "electricity market price forecast value" is calculated in step S12. The "LNG power generation cost" is calculated using the above-mentioned formula 1. The "all-time-zone forecast average value" indicates the average value of the electricity market price forecast value for all time zones (the average value of the electricity market price forecast values ​​for 48 frames calculated in step S12).

[0048] In this way, the calculation unit 120 corrects the electricity market price forecast value for each of the 48 periods based on the LNG futures price. Figure 6 shows an example of the transition of the corrected price forecast value for a certain month. As shown in Figure 6, the corrected price forecast value is calculated for each weekday, Saturday, and Sunday and public holiday (Sunday and public holiday) in each month.

[0049] After performing the process of step S13, the calculation unit 120 ends the prediction control shown in Fig. 2. In this way, the electricity market price prediction system 100 can obtain predicted electricity market price values ​​for each month, weekday, Saturday, Sunday, and holiday from one month ahead to one year ahead.

[0050] As described above, in the electricity market price prediction system 100 according to this embodiment, first, a model curve (average value ratio) is created based on past electricity market prices (actual period price values), thereby making it possible to reflect the trend of the electricity market price for each period (step S11). Next, by multiplying the model curve (average value ratio) by the electricity futures price, it is possible to obtain a predicted value of the electricity market price for each time slot (30 minutes) for one month or more into the future (step S12). Furthermore, by using the electricity futures price to predict the electricity market price, it is possible to obtain a predicted value of the electricity market price that reflects market trends for one month or more into the future. In this way, the electricity market price prediction system 100 according to this embodiment can improve the accuracy of long-term electricity market price predictions.

[0051] In step S12, the predicted electricity market price is calculated using the base load price for all time periods on weekends and holidays, and from 8 PM to 8 AM on weekdays, while the predicted electricity market price is calculated using the daytime load price for the time periods on weekdays. In this way, by using the electricity futures price (base load price or daytime load price) according to the time period to predict the electricity market price, the accuracy of the prediction of the electricity market price can be further improved.

[0052] In addition, the accuracy of electricity market price forecasts can be improved by correcting the electricity market price forecast value using LNG futures prices, which are correlated with the electricity market price.

[0053] 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 (computation unit 120) acquires (step S11) actual price values ​​(average actual price values ​​per unit time), which are past electricity market prices for each predetermined time period (e.g., 30 minutes), and an all-time period actual average value, which is the average value of the actual price values ​​(average actual price values ​​per unit time period) for all time periods; A second acquisition unit (calculation unit 120) acquires an average value ratio, which is the ratio of the actual price value (average frame price actual value) to the average actual value for the entire time period for each predetermined time period (step S11); a third acquisition unit (calculation unit 120) that multiplies the average value ratio for each of the predetermined time slots acquired by the second acquisition unit by the electricity futures price to acquire a predicted electricity market price for each of the predetermined time slots (step S12); It is equipped with the following.

[0054] This configuration can improve the accuracy of long-term electricity market price forecasts. Specifically, it is possible to obtain predicted electricity market prices for each predetermined time period (30 minutes) over a long period, such as one month or longer. It is also possible to obtain predicted electricity market prices that reflect market trends for more than one month in the future.

[0055] Moreover, the first acquisition unit As the actual price value, an average value (average actual frame price value) of the actual price values ​​(actual frame price values) in each time period is used.

[0056] This configuration can further improve the accuracy of electricity market price prediction. Specifically, by using the average value (average actual period price value) of the actual price values ​​(actual period price values) for each time period to predict the electricity market price, the accuracy of past trends in electricity market prices can be improved, and ultimately the accuracy of predictions of electricity market prices can be further improved.

[0057] Moreover, the first acquisition unit The actual price value is obtained based on calendar information.

[0058] This configuration can further improve the accuracy of electricity market price prediction. Specifically, since the electricity market price can be predicted using actual price values ​​for a nearby day or month when the electricity market price trend is similar, the accuracy of the electricity market price prediction can be further improved.

[0059] Moreover, the electricity futures price is Including base load price, The third acquisition unit The average value ratio is multiplied by the base load price to obtain the electricity market price forecast value.

[0060] This configuration can further improve the accuracy of electricity market price prediction.

[0061] Moreover, the electricity futures price is Includes time-specific pricing (e.g., intraday load pricing), The third acquisition unit For the target time period of the price limited to the time period (8:00 to 20:00 on weekdays), the average value ratio is multiplied by the price limited to the time period to obtain the electricity market price forecast value; For time periods other than the target time periods (all time periods on Saturdays, Sundays, and holidays, and from 8:00 p.m. to 8:00 a.m. on weekdays), the electricity market price forecast value is obtained by multiplying the average value ratio by the base load price.

[0062] This configuration can further improve the accuracy of electricity market price prediction. Specifically, by using the electricity futures price (base load price or daytime load price) according to the time period to predict the electricity market price, the accuracy of the prediction of the electricity market price can be further improved.

[0063] Furthermore, the electricity market price prediction system 100 according to this embodiment includes: The power generation system includes a correction unit that corrects the predicted electricity market price using a futures price of fuel for power generation.

[0064] This configuration can further improve the accuracy of electricity market price prediction. Specifically, since futures prices of fuels for power generation (for example, LNG futures prices) are correlated with electricity market prices, the accuracy of electricity market price forecasts can be improved.

[0065] 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.

[0066] For example, 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.

[0067] Furthermore, in this embodiment, the calculation unit 120 corrects the electricity market price forecast value for each of the 48 time slots based on the LNG futures price (step S13), but the correction may be performed only for time slots that have a particularly strong correlation with the LNG futures price (for example, midnight to 6:00 AM). In this case, in the above-mentioned formula 2, the average value of the electricity market price forecast value for each time slot from midnight to 6:00 AM is used instead of the average forecast value for all time slots (the average value of the electricity market price forecast values ​​for the 48 time slots).

[0068] In addition, in this embodiment, the calculation unit 120 calculates the electricity market price forecast value using the daytime load price from 8:00 to 20:00 on weekdays, but it may also calculate the electricity market price forecast value using the base load price from 8:00 to 20:00 on weekdays. [Explanation of symbols]

[0069] 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 price record value, which is a past electricity market price for each predetermined time period, and an all-time period actual average value, which is an average value of the price record value for all time periods; a second acquisition unit that acquires an average value ratio, which is a ratio of the actual price value to the average actual price value for the entire time period, for each predetermined time period; a third acquisition unit that acquires an electricity market price forecast value for each of the predetermined time slots by multiplying the average value ratio for each of the predetermined time slots acquired by the second acquisition unit by an electricity futures price; An electricity market price forecasting system comprising:

2. The first acquisition unit As the actual price value, an average value of the actual price value in each time period is used. The electricity market price forecasting system of claim 1 .

3. The first acquisition unit obtaining the actual price value based on calendar information; The electricity market price forecasting system of claim 1 .

4. The electricity futures price is Including base load price, The third acquisition unit multiplying the average value ratio by the base load price to obtain the electricity market price forecast value; The electricity market price forecasting system according to any one of claims 1 to 3.

5. The electricity futures price is Including time-limited pricing, The third acquisition unit For a target time period of the price limited to the time period, the average value ratio is multiplied by the price limited to the time period to obtain the electricity market price forecast value; For time periods other than the target time period, the average value ratio is multiplied by the base load price to obtain the electricity market price forecast value. The electricity market price forecasting system according to claim 4.

6. a correction unit that corrects the electricity market price forecast value using a futures price of fuel for power generation; The electricity market price forecasting system of claim 1 .

Images