JP2024135406A - Information processing device, information processing method, and computer program - Google Patents

Abstract

[Problem] To provide an information processing device, information processing method, and computer program for improving the accuracy of price estimation in the electricity market. [Solution] The information processing device disclosed herein includes a processing unit that acquires planned data on the amount of power generated by a power generation facility that generates electricity that can be traded in the electricity market, and, when a first event related to a change in the amount of power generated on a target day is detected based on the planned data, estimates the amount of fluctuation in the market price of the electricity market due to the first event on the target day. [Selected Figure] Figure 1

Description

The present disclosure relates to an information processing device, an information processing method, and a computer program.

When promoting and operating renewable energy, it may be necessary to probabilistically estimate future prices and bid volumes in the electricity market to meet the requirements of various applications. Since electricity demand and supply are mainly affected by weather, good accuracy can be achieved by estimating electricity market prices using weather forecasts from a numerical weather simulator. As a specific example, there is a method of obtaining past market prices with weather conditions similar to the weather forecast value for the target day, and estimating various quantities related to the electricity market (price and bid volume).

However, this method cannot predict sudden changes in the electricity market (such as price hikes) in advance. Market prices are determined by a variety of factors, including seasonal circumstances, problems with power generation equipment, recent world events, and natural disasters, and methods that focus on similarities in weather conditions may not be able to make accurate predictions.

JP 2019-96164 A

This embodiment provides an information processing device, an information processing method, and a computer program that improve the accuracy of estimating market prices in the electricity market.

The information processing device disclosed herein includes a processing unit that acquires planned data on the amount of electricity generated by a power generation facility that generates electricity that can be traded in the electricity market, and, when a first event related to a change in the amount of electricity generated on a target day is detected based on the planned data, estimates the amount of fluctuation in the market price of the electricity market on the target day due to the first event.

FIG. 2 is a functional block diagram of a price rise prediction device which is an information processing device according to the present embodiment. FIG. 2 is a diagram showing a specific example of an electricity market performance data area. A diagram showing electricity market performance data and specific examples from across the country. FIG. 11 is a diagram showing a specific example of weather forecast data. FIG. 4 is a diagram showing a specific example of power generation facility shutdown plan data. FIG. 13 is a diagram showing an example of output data from a price rise prediction device (prediction result calculation unit). 2 is a flowchart of an example of the overall operation of the price rise prediction device of FIG. 1 . 11 is a flowchart showing an example of the operations of a data extraction range determination unit, a normal price prediction unit, and a prediction result calculation unit. FIG. 11 is a diagram showing an example of changing the data extraction range. FIG. 11 is a diagram showing an example of creating search target data. FIG. 4 is a diagram showing a schematic diagram of a normal price prediction operation performed by a normal price prediction unit. FIG. 13 is a diagram illustrating an example of calculation by a prediction result calculation unit when a data extraction range is changed. FIG. 13 is a diagram showing another example of output data from the prediction result calculation unit. FIG. 2 is a diagram showing an example of the hardware configuration of a price rise prediction device as an information processing device according to the present embodiment.

This embodiment will be described below with reference to the drawings.

Figure 1 is a functional block diagram of a price rise prediction device 10, which is an information processing device according to this embodiment. The price rise prediction device 10 mainly comprises a processing unit 100, a data acquisition unit 200, a data storage unit 300, and an input unit 400. The processing unit 100 comprises a price rise prediction function 100A, a normal time prediction function 100B, and an electricity market prediction function 100C.

In this embodiment, it is determined whether the market price in the electricity market will soar on a target date based on the operating status of the power plant (power generation facility), specifically, the planned data on the amount of power generated by the power generation facility. For example, data on the planned value of the reduction in power generation (amount of outage) of the power plant (power generation facility) published on the Power Generation Information Website (HJKS) is acquired periodically or intermittently, and an event related to a change in the amount of power generation (a first event or an event caused by the power generation facility) is detected based on the amount of reduction or the change in the amount of reduction from the previous day. Depending on whether an event caused by the power generation facility is detected, it is estimated whether a price hike will occur.

If the first event is detected, it is determined that a price hike will occur, and if the first event is not detected, it is determined that a price hike will not occur. If it is determined that a price hike will occur, the amount of fluctuation in the market price in the electricity market due to the first event on the target day is estimated, and the market price on the target day is estimated based on this amount of fluctuation. This makes it possible to estimate the market price with high accuracy even if the market price hikes on the target day. For example, even if power generation is stopped or the amount of power generation is reduced on the target day due to a sudden cause in the power generation status of the power generation facility, it is possible to estimate the market price on the target day with high accuracy.

The reduction in power generation at a power plant is also called the amount of shutdown, since power generation is stopped by the amount of the reduction. When the power generation equipment is shut down, the amount of power generation is reduced by the rated output. When the output of the power generation equipment is reduced, the amount of reduction is the amount of reduction. In the following explanation, the term "shutdown of power generation equipment" is mainly used, but the shutdown of power generation equipment can also include a reduction in the output of the power generation equipment.

In addition, in this embodiment, in addition to the operating status of the power plant (power generation facility), weather-related events (second events or weather-related events) are detected based on the weather forecast values for the prediction target day. Whether a weather-related event is detected is used to determine whether a weather-related market price surge will occur. If a second event is detected, it is determined that a weather-related price surge will occur, and if a second event is not detected, it is determined that a weather-related price surge will not occur. If it is determined that a price surge will occur, the market price is estimated using past data (weather forecast data and electricity market performance data from previous years) with weather conditions similar to those of the prediction target day. This makes it possible to estimate the market price on the target day with high accuracy, even if the amount of power generation decreases (including the suspension of power generation) due to a sudden change in weather conditions.

In this way, according to this embodiment, the market price can be estimated with high accuracy even when the amount of power generation decreases due to a problem with the power generation equipment or when the market price rises due to a sudden change in weather conditions.

In the following explanation, a future date (prediction target date) is mainly assumed as the target date for estimating the market price, and a case will be shown in which the market price on the prediction target date is estimated. In this embodiment, estimating the market price on a future date is specifically referred to as predicting the market price, and hereinafter, estimating the market price etc. will be described as predicting the market price. This embodiment will be described in more detail below.

The data acquisition unit 200 includes a weather forecast acquisition unit 210, a power market performance acquisition unit 220, and a power generation equipment shutdown plan acquisition unit 230.

The weather forecast acquisition unit 210 acquires weather forecast data including weather forecast results (e.g., calculation results of the WRF numerical weather simulator) for AMeDAS locations nationwide from the weather forecast system 1, and sends the weather forecast data to the data storage unit 300. The weather forecast data may be acquired at any time interval, such as every hour, every day, or every week, or may be acquired by sending an acquisition request to the weather forecast system 1 at the required timing.

The electricity market performance acquisition unit 220 acquires electricity market performance data, such as market prices and transaction volume (contract volume) performance values in the electricity market, bid volume in the trading market, and virtual system price data, from the electricity market performance value management system 2, and sends the electricity market performance data to the data storage unit 300. Examples of information that can be acquired from the electricity market performance value management system 2 are disclosed on the website of the Japan Electric Power Exchange (JEPX), a general incorporated association, at http://www.jepx.org/market/index.html. The electricity market performance data may be acquired at any interval, such as once a day or once every hour, which is the unit of trading time in the electricity market, or by sending an acquisition request to the electricity market performance value management system 2 at the required timing.

The power generation equipment shutdown plan acquisition unit 230 acquires data (power generation equipment shutdown plan data) on the operation status and shutdown status of power plants (power generation equipment) by power source type nationwide from the power plant operation status management system 3, and sends the power generation equipment shutdown plan data to the data storage unit 300. Examples of information that can be acquired from the power plant operation status management system 3 include the power generation information disclosure system and that is disclosed on the HJKS website https://hjks.jepx.or.jp/hjks/top. There are no particular limitations on the timing of acquisition of the power generation equipment shutdown plan data, so long as it can be acquired intermittently. It may be acquired at any time interval, such as every hour, every day, or every week, or it may be acquired by sending an acquisition request to the power plant operation status management system 3 at the required timing.

The data storage unit 300 includes a weather forecast storage unit 310, an electricity market performance storage unit 320, and a power generation equipment shutdown plan storage unit 330.

The weather forecast storage unit 310 receives weather forecast data from the weather forecast acquisition unit 210 and stores the weather forecast data. The weather forecast storage unit 310 also stores weather forecast data acquired in the past. The stored weather forecast data is sent to the price increase prediction function 100A and also to the normal price prediction data storage unit 180 of the normal time prediction function 100B. The normal price prediction data storage unit 180 stores the weather forecast data received from the weather forecast storage unit 310.

The electricity market performance memory unit 320 receives electricity market performance data from the electricity market performance acquisition unit 220 and stores the electricity market performance data. The electricity market performance memory unit 320 also stores electricity market performance data acquired in the past. The stored electricity market performance data is sent to the price increase prediction function 100A and also to the normal price prediction data memory unit 180 of the normal time prediction function 100B. The normal price prediction data memory unit 180 stores the electricity market performance data received from the electricity market performance memory unit 320.

The power generation equipment shutdown plan storage unit 330 receives power generation equipment shutdown plan data from the power generation equipment shutdown plan acquisition unit 230 and stores the power generation equipment shutdown plan data. The power market performance storage unit 320 also stores power market performance data acquired in the past. The power generation equipment shutdown plan storage unit 330 also stores power market performance data acquired in the past. The stored power generation equipment shutdown plan data is sent to the price rise prediction function 100A.

The input unit 400 accepts input from the user and sends the input information to the price rise prediction function 100A. The input unit 400 is a device that can input information by any method, such as a keyboard, a mouse, a touch panel, a voice input unit, or a gesture input unit. The types of information accepted by the input unit 400 are shown below as [1] to [3].

[1] Information (aggregation item information) on items (e.g., cause of shutdown, type of shutdown (planned shutdown, unplanned shutdown)) to be aggregated to determine whether a price rise due to a problem with the power generation equipment has occurred. As described later, the planned shutdown value of the power generation equipment at the power plant (the planned value of the amount of power generation that is reduced from the planned amount of power generation (e.g., rated output) or the amount of shutdown) for the input items is summed up by area corresponding to the electricity market, and the summed value is used to determine whether a price rise has occurred in the electricity market. More specifically, based on the summed value, it is determined whether an event related to a change in the amount of power generation (first event or power generation equipment-related event) has been detected. If a power generation equipment-related event is detected, it is determined that a price rise due to the power generation equipment will occur, and if a power generation equipment-related event is not detected, it is determined that a price rise due to the power generation equipment will not occur. In this way, the occurrence of a price rise is determined. In addition to the cause of shutdown and the type of shutdown, the items to be aggregated may also be the type of power generation equipment.

[2] Information indicating the judgment conditions (power generation equipment cause judgment conditions) for judging whether or not there is a price hike. Examples of judgment conditions include a threshold for the amount of decrease in power generation (amount of decrease in power generation) and a threshold for the amount of change in the amount of decrease in power generation from the previous day's value.

[3] Information indicating the items to be judged for determining whether a price hike is due to weather, and the judgment conditions (weather-caused judgment conditions). Specifically, information regarding the items on which weather (weather forecast) to focus, and threshold information for comparing with the weather forecast value for that weather. As described below, it is determined whether a weather-related event (second event or weather-related event) is detected based on the weather forecast value indicated by the weather forecast item. If a weather-related event is detected, it is determined that a weather-caused price hike will occur, and if not, it is determined that a weather-caused price hike will not occur.

The price increase prediction function 100A will now be outlined.
The price hike prediction function 100A receives electricity market performance data and power generation facility shutdown plan data from the data storage unit 300, and receives tally item information and judgment conditions (power generation facility cause judgment conditions) from the input unit 400. The price hike prediction function 100A judges whether or not a hike in the electricity market price caused by the power generation facility (hereinafter, a price hike) will occur, and calculates the amount of the hike if a price hike does occur. The price hike prediction function 100A sends data indicating the calculated price hike amount to the output unit 500.

In this embodiment, a price increase is defined as, for example, a reference price being determined for each area and the price is equal to or higher than the reference price. The method of determining the reference price may be arbitrary. Based on the most recent market price, the normal price when there is no increase and the price when there is an increase (normal price + increase amount) may be respectively identified, and the price that divides these prices in half may be set as the reference price. Other methods of determination are also possible. An example of a method of determining the reference price is shown below.
Example 1: When a day for trading in the electricity market is divided into 48-hour time slots of 30 minutes each and a commodity (electricity) is traded for each time slot, the base price for each time slot is set at a uniform price of 40 yen/kWh.
Example 2: Based on the electricity market performance data for the past year, the 75th percentile value is set as the base price for each time slot.

The price hike prediction function 100A also receives weather forecast data from the weather forecast storage unit 310, and receives weather forecast items and judgment conditions (weather-caused judgment conditions) from the input unit 400. The price hike prediction function 100A judges whether or not a weather-caused price hike will occur. If it is determined that a price hike will not occur, it determines the data extraction range required to calculate the market price in the case where a weather-caused price hike does not occur (for example, 7 days from the day before the prediction target date), and sends information indicating the specified data extraction range to the normal time prediction function 100B. If it is determined that a price hike will occur, it specifies the range of the data period (data extraction range) required to calculate the market price in the case where a price hike occurs due to weather from the weather forecast data and the electricity market performance data, respectively (for example, a period in the past year when the weather conditions are similar), and sends information indicating the specified data extraction range to the normal time prediction function 100B. In this way, if it is determined that a weather-caused price hike will occur, the data extraction range is changed from normal time. The market price when a weather-related price hike occurs and the market price when a weather-related price hike does not occur are calculated by the normal forecast function 100B.

The price increase prediction function 100A will be described in further detail below.
The price rise prediction function 100A is a function for predicting price rises caused by power generation equipment, and includes a power generation equipment shutdown plan data pre-processing unit 110 (hereinafter referred to as the pre-processing unit 110), a trend analysis unit 120, a price rise determination unit (caused by power generation equipment) 130, a price sensitivity calculation unit 140, and a price rise amount prediction unit 150.

The pre-processing unit 110 pre-processes the power generation equipment shutdown plan data. More specifically, the pre-processing unit 110 receives the power generation equipment shutdown plan data from the data storage unit 300, and receives the tally item information from the input unit 400. The pre-processing unit 110 tally (sums) the planned values of the reduction amounts for the items indicated in the tally item information in the power generation equipment shutdown plan data for each area, and sends the tally data of the reduction amounts to the trend analysis unit 120.

The trend analysis unit 120 receives the decline amount data for each area from the preprocessing unit 110 and the electricity market performance data from the electricity market performance storage unit 320, and performs trend analysis of the electricity market price based on the fluctuation of the decline amount for each area (time series change) and the fluctuation of the electricity market price (time series change). More specifically, the trend analysis unit 120 receives the decline amount data for each area from the preprocessing unit 110 and the electricity market performance data from the electricity market performance storage unit 320, analyzes the fluctuation trend of the decline amount for each area and the fluctuation trend of the electricity market price, and calculates the change in the decline amount for each area (for example, the change from the previous day) and the change rate of the electricity market price relative to the change in the decline amount. The trend analysis unit 120 sends the power generation stop change amount data including the calculated information to the price increase determination unit (power generation equipment cause) 130. The change in the amount of decrease may be calculated, for example, as the difference between the amount of decrease calculated based on first planning data acquired at a first point in time (e.g., the day before the day of the prediction target) for the amount of power generated by the power generation facility on the prediction target day, and the amount of decrease calculated based on second planning data acquired at a second point in time prior to the first point in time (e.g., the day before the first point in time) for the amount of power generated by the power generation facility on the prediction target day. In other words, the amount of change in the amount of decrease can be calculated by comparing the first planning data with the second planning data.

The price hike determination unit (power generation equipment cause) 130 determines whether a price hike (price hike due to power generation equipment) caused by the shutdown of a power plant (power generation equipment) will occur. More specifically, the price hike determination unit (power generation equipment cause) 130 uses the power generation shutdown change amount data for each area received from the trend analysis unit 120 and the threshold value (at least one of the decrease amount threshold (stop amount threshold) and the decrease amount change amount (decrease amount change amount)) received from the input unit 400 to determine whether an event related to a change in power generation amount (first event or power generation equipment cause event) is detected. If a power generation equipment cause event is detected, it determines that a price hike due to power generation equipment will occur, and if a power generation equipment cause event is not detected, it determines that a price hike due to power generation equipment will not occur. Whether a power generation equipment cause event is detected is determined, for example, by whether the decrease amount of power generation amount for each area on the prediction target day exceeds a threshold value, and whether the change amount of the decrease amount (the change amount of the decrease amount from the day before the prediction target day (the day before one trading day)) exceeds a threshold value. If any of the thresholds are exceeded, a power generation equipment-related event is detected, and it is determined that a price rise due to power generation equipment will occur. The price rise determination unit (power generation equipment-related cause) 130 sends information on whether a price rise due to power generation equipment will occur to the price rise prediction unit 150, and if it determines that a price rise will occur, it sends information indicating the amount of decrease in the amount of power generation and the amount of change in the amount of decrease (decrease amount/change amount information) to the price rise prediction unit 150.

The price sensitivity calculation unit 140 calculates price sensitivity, which is an index that converts the amount of change in the amount of decrease in power generation (amount of change in the shutdown of power generation facilities) into the amount of price change in the electricity market. More specifically, the price sensitivity calculation unit 140 calculates price sensitivity, which is an index that converts the amount of change in the amount of decrease into the amount of change in the electricity market price, using the electricity market performance data received from the electricity market performance storage unit 320. The price sensitivity calculation unit 140 sends data indicating the calculated price sensitivity (price sensitivity data) to the price increase prediction unit 150.

The price surge prediction unit 150 predicts the price surge amount (price difference from the normal price on the assumption that there is no price surge due to power generation equipment) when the price surge determination unit (power generation equipment cause) 130 determines that a price surge will occur. More specifically, when the price surge determination unit (power generation equipment cause) 130 determines that a price surge due to power generation equipment will occur, the price surge prediction unit 150 receives the change amount data of the power generation decrease amount and the price sensitivity data, and predicts the price surge amount on the prediction target date. The price surge prediction unit 150 sends data indicating the predicted price surge amount (price surge amount data) to the prediction result calculation unit 182. When the price surge determination unit (power generation equipment cause) 130 determines that a price surge due to power generation equipment will not occur, the price surge prediction unit 150 may send price surge amount data indicating 0 yen as the price surge amount to the prediction result calculation unit 182.

The price hike prediction function 100A is a function for predicting weather-related price hikes and includes a price hike determination unit (weather cause) 160 and a data extraction range determination unit 170.

The price hike determination unit (weather cause) 160 receives weather forecast data and electricity market performance data from the data storage unit 300, and receives information indicating the weather forecast item and threshold from the input unit 400. The price hike determination unit (weather cause) 160 identifies the weather forecast value indicated by the weather forecast item in the weather forecast data, and detects a weather-related event (a second event or a weather-related event) based on the weather forecast value. If a weather-related event is detected, it is determined that a weather-related price hike will occur, and if not, it is determined that a weather-related price hike will not occur. Whether or not a weather-related event will occur is determined, for example, by whether or not the weather forecast value exceeds the threshold. In this case, if the weather forecast value meets the threshold, it is determined that a weather-related price hike will occur. Meeting the threshold can mean either exceeding the threshold or falling below it, depending on the type of weather. The price hike determination unit (weather cause) 160 sends information on whether or not a weather-related price hike will occur to the data extraction range determination unit 170.

The data extraction range determination unit 170 determines the data extraction range (data period) to be used by the normal price prediction unit 181 based on information from the price increase determination unit (weather cause) 160 on whether or not a weather-related price increase will occur. For example, if it is determined that a weather-related price increase will not occur, the data extraction range indicating a certain period (e.g., one week) from the day before the prediction target date (the trading day one day before) is determined. If it is determined that a weather-related price increase will occur, a similar past period including a day similar to the weather conditions indicated by the weather forecast value (or a certain period before or after that) is identified from the past weather forecast data, and a data extraction range indicating the identified period is determined. The similar past period may be any length, for example, 30 days. The data extraction range determination unit 170 sends information indicating the determined data extraction range to the normal price prediction unit 181 of the normal time prediction function 100B.

The normal time prediction function 100B includes a normal price prediction data storage unit 180, a normal price prediction unit 181, and a prediction result calculation unit 182.

The normal price prediction data storage unit 180 receives and stores actual electricity market price data and weather forecast data from the data storage unit 300. The normal price prediction data storage unit 180 may also store data previously received from the data storage unit 300.

The normal price prediction unit 181 extracts weather forecast data and electricity market price actual data from the storage unit 180 based on the data extraction range information received from the data extraction range determination unit 170. The normal price prediction unit 181 predicts the market price based on the extracted data. The prediction made by the normal price prediction unit 181 is called normal price prediction. When it is determined that a price rise due to weather will occur, the normal price prediction unit 181 can perform price prediction that takes into account the price rise due to weather, using data from a similar period in the past, by using data in the range indicated in the data extraction range information received from the data extraction range determination unit 170. When it is determined that a price rise due to weather will not occur, it can perform price prediction as usual, for example, using the most recent data (for example, data for one week prior to the prediction target date). The price predicted by the normal price prediction unit 181 is the price (normal price or first market price) assuming that there is no price rise due to power generation equipment, and the normal price reflects this price rise if there is a price rise due to weather. The normal price prediction unit 181 calculates a predicted value of the electricity market price (predicted normal price value) and sends data including the predicted value (normal price data) to the prediction result calculation unit 182.

The prediction result calculation unit 182 receives normal price data from the normal price prediction unit 181 and price hike data from the price hike prediction unit 150, and calculates the final prediction result of the electricity market price. For example, the final prediction result of the electricity market price is calculated by adding the price hike indicated by the price hike data (if there is no price hike due to the power generation equipment, the price hike can be 0 yen) to the normal price indicated by the normal price data. Calculating the final electricity market price by adding the normal price and the price hike is only one example, and other methods may be used as long as they can be calculated using a function including the normal price and the price hike. For example, the final electricity market price may be calculated by a weighted sum of the normal price and the price hike. The prediction result calculation unit 182 sends output data including the calculated prediction result, etc. to the output unit 500. The output data may include information other than the prediction result, such as the date, the normal price, and the price hike.

The output unit 500 outputs the output data received from the prediction result calculation unit 182. The output unit 500 is, for example, a display, and displays the output data on a screen. The output unit 500 may be a communication device. In this case, the output unit 500 may transmit the output data to a device of a company that uses the output data to carry out business.

For example, the output unit 500 may transmit the output data to a bidding device that generates bidding data for the electricity market and provides it to an electricity market trading system. When the bidding device determines based on the output data that the market price will be higher than a threshold on the prediction target date, it may place a buy bid for the required amount of electricity in the electricity market trading system before the prediction target date in order to buy electricity before the prediction target date. Alternatively, in order to sell electricity at a high price on the prediction target date, it may refrain from placing a sell bid for the required amount of electricity until before the prediction target date, and then place a sell bid on the prediction target date to sell electricity at a high price.

The output unit 500 may also transmit the output data to a device (DR device) that controls the demand response (DR) of a demand response (DR) business operator. If the DR device determines that the market price will be higher than a threshold on the prediction target date, it may plan to implement downward DR, which causes consumers to reduce their electricity usage, on the prediction target date. There are various other possible uses for the output data.

Before explaining the operation of this embodiment, definitions and specific examples of various parameters, data, and formulas are provided below.

FIG. 2 shows an example of an electricity market performance data area.
The spot market performance data of the electricity market includes the system price (yen/kWh) for each 30-minute time frame and the price (market price) for each area (yen/kWh). The time code is an integer value between 1 and 48 that labels each 30-minute time frame. The spot/hourly average price (yen/kWh) may also be recorded.

Figure 3 shows actual electricity market data and specific examples from across the country.
Virtual system prices are recorded. Virtual system prices have been published on the JEPX website since January 2022. Here, the virtual system price refers to the price at the intersection of the supply and demand curves when the selling bid volume and buying bid volume are increased by 0.5 GW, 1 GW, and 5 GW, respectively, using the supply and demand curve data of the system price for each time frame. For example, "virtual system price selling 500 MW" is the price at the intersection of the supply and demand curves when the selling bid volume is increased by 0.5 GW. "Virtual system price buying 1000 MW" is the price at the intersection of the supply and demand curves when the buying bid volume is increased by 1 GW. The price of the intersection corresponds to the contract price, for example.

FIG. 4 shows a specific example of weather forecast data.
Numerical weather forecast data is provided by using Weather Research and Forecasting (WRF) data at 10-minute intervals at the AMeDAS points used for observation by the Japan Meteorological Agency. The example in the figure shows average values every 30 minutes. For each area to be predicted, the temperature (T2: ℃) and solar radiation intensity (AVG4: W/m2) of all AMeDAS points contained within the area are used. The column names are "T2_code" and "AVG4_code" using the code number. For example, "T2＿00001" is the temperature at AMeDAS point 00001, and "AVG4_00001" is the solar radiation intensity at AMeDAS point 00001.

FIG. 5 shows a specific example of power generation facility shutdown plan data.
Information on power generation facility outages is published on HJKS. HJKS centrally manages operation and outage information for power generation facilities (generators) in each area. Power generation facility outage information makes it possible to understand the operation plans for power generation facilities in each area. Information includes the area, power generation company, power plant code, power plant name, power generation type, unit name (generator identifier), authorized output, outage classification (planned outage, unplanned outage, output reduction), type, reduction amount (amount of reduction in power generation), outage date and time, recovery outlook, planned recovery date, cause of outage, and last update date and time.

他の例として、下記の式(3-2)を用いてもよい。

As another example, the following formula (3-2) may be used.

式(4)の判定結果を示すデータが、価格高騰判定部（発電設備原因）１３０の出力データとなる。

Data indicating the determination result of equation (4) becomes output data of the price increase determination unit (power generation equipment cause) 130 .

を満たすか否かの判定を行う。この式が満たされる場合は、閾値を満たすと決定する。式(6)の判定結果を示すデータが、価格高騰判定部(気象原因)１６０の出力データとなる。

It is determined whether the following expression is satisfied. If this expression is satisfied, it is determined that the threshold value is satisfied. Data indicating the determination result of expression (6) becomes output data of the price increase determination unit (weather cause) 160.

For example, the similar year is selected as the year with the closest average price performance value from the day before the forecast target date t to one week before. For example, if the forecast target date t = 2022/03/17, the average price for this year is calculated from the data for the period 2022/03/10 to 2022/03/16. Similarly, for years prior to 2022, the average price is calculated from the data for the period 20XX/03/10 to 20XX/03/16. "XX" is, for example, "21", "20", "19", "18", or "17". Here, the number of years going back is 5, but it can be less than 5 or 6 or more. The year that is closest to this year's average value among these XXs is determined as the similar year. Here, the same period for each year is targeted, but the time series of the price performance value from the day before the forecast target date t to one week before may be detected and used as the similar period. In this case, the past period may be the same year as the forecast target date, or it may be a year prior to the year to which the forecast target date belongs. The number of years going back can be arbitrary.

The operation example in the flowchart of FIG. 7 is just one example, and other operations are possible. For example, if it is determined that no weather-related price hikes will occur, the data extraction range determination unit 170 may send information specifying a certain period (e.g., one week) from the day before the target prediction date to the normal price prediction data storage unit 180, and the normal price prediction data storage unit 180 may extract data based on that information. Also, that information or the above-mentioned data extraction range change information may be sent to the normal price prediction unit 181 instead of the normal price prediction data storage unit 180, and the normal price prediction unit 182 may specify a data range in the normal price prediction data storage unit 180 and read it out.

となる。
また、データ抽出範囲の変更を行った場合は、過年度データから予測した通常価格予測値と、直近データから予測した通常価格予測値との2つを利用することもできる。

It becomes.
In addition, if the data extraction range is changed, it is possible to use two types of predicted normal price values: a normal price predicted from past years' data and a normal price predicted from the most recent data.

図１０では気象データの利用地点として、N個の地点を利用した場合の例を示している。また、休日には土日及び祝日を含めている。

10 shows an example in which N locations are used as locations for using weather data. Weekends and public holidays are included in the holidays.

(Step S15)
The prediction result calculation unit 182 calculates the predicted value of the market price (area price) based on the predicted value of the normal price and the predicted value of the price hike. The following will explain the case where the data extraction range is not changed and the case where it is changed.

As described above, according to this embodiment, it is determined whether a market price surge will occur on the forecast target date based on the operating status of the power plant, specifically the planned data of the power generation amount of the power generation equipment, and if it is determined that a surge will occur, the amount of fluctuation in the market price is estimated based on the planned data. By estimating the market price based on this amount of fluctuation, the market price can be estimated with high accuracy even if the market price surges.

In addition, in this embodiment, in addition to the operating status of the power plant's power generation equipment, a determination is made as to whether a weather-related market price surge will occur based on the weather forecast value for the prediction target day, and if it is determined that a weather-related price surge will occur, the market price is estimated using past data (weather forecast data and electricity market performance data) with weather conditions similar to those of the prediction target day. This makes it possible to estimate the market price with high accuracy even if power generation decreases or stops due to a sudden change in weather conditions.

As described above, according to this embodiment, the market price can be estimated with high accuracy even when the amount of power generation decreases due to a problem with the power generation equipment, or when the market price rises due to a sudden change in weather conditions. This embodiment can improve the profits of businesses involved in the supply and demand of electricity, and can be used for fuel cell usage planning, etc.

(Modification)
In the above embodiment, a market price surge is determined based on the amount of decrease in the amount of power generation, but a similar process can be performed to determine a market price crash and predict the market price in the case of an increase in the amount of power generation. In this case, the above description should be read as a crash, and the amount of decrease should be read as an increase, etc.

[1] Hyperparameters for similarity search used below

以上が、類似検索を用いて通常価格予測値を算出する際の計算の詳細である。

The above is the details of the calculation when calculating the normal price prediction value using similarity search.

(Hardware configuration)
14 shows the hardware configuration of a price rise prediction device 10 as an information processing device according to this embodiment. The price rise prediction device 10 is configured with a computer device 600. The computer device 600 includes a CPU 601, which is a processor, an input interface 602, a display device 603, a communication device 604, a main storage device 605, and an external storage device 606, which are interconnected by a bus 607.

The CPU (Central Processing Unit) 601 executes an information processing program, which is a computer program, on the main memory device 605. The information processing program is a program that realizes each of the above-mentioned functional configurations of the device. The information processing program may be realized not by a single program, but by a combination of multiple programs or scripts. Each functional configuration is realized by the CPU 601 executing the information processing program.

The input interface 602 is a circuit for inputting operation signals from input devices such as a keyboard, mouse, and touch panel to this device.

The display device 603 displays the data output from the device. The display device 603 is, for example, but is not limited to, an LCD (liquid crystal display), an organic electroluminescence display, a CRT (cathode ray tube), or a PDP (plasma display). The data output from the computer device 600 can be displayed on the display device 603.

The communication device 604 is a circuit that enables the device to communicate wirelessly or wired with an external device. Data can be input from the external device via the communication device 604. The data input from the external device can be stored in the main memory device 605 or the external memory device 606.

The main memory device 605 stores an information processing program, data required for executing the information processing program, and data generated by executing the information processing program. The information processing program is deployed and executed on the main memory device 605. The main memory device 605 is, for example, a RAM, a DRAM, or an SRAM, but is not limited to these. Each storage unit or database in FIG. 1 may be constructed on the main memory device 605.

The external storage device 606 stores information processing programs, data required for executing the information processing programs, and data generated by executing the information processing programs. These information processing programs and data are read into the main storage device 605 when the information processing programs are executed. The external storage device 606 is, for example, a hard disk, an optical disk, a flash memory, and a magnetic tape, but is not limited to these. Each storage unit or database of the information processing device may be constructed on the external storage device 606.

The information processing program may be pre-installed in the computer device 600, or may be stored in a storage medium such as a CD-ROM. The information processing program may also be uploaded onto the Internet.

In addition, the device may be configured as a single computer device 600, or as a system consisting of multiple computer devices 600 connected to each other.

The present invention is not limited to the above-described embodiments as they are, and in the implementation stage, the components can be modified and embodied without departing from the gist of the invention. In addition, various inventions can be formed by appropriately combining multiple components disclosed in the above-described embodiments. For example, configurations in which some components are deleted from all the components shown in each embodiment are also possible. Furthermore, components described in different embodiments may be appropriately combined.

This embodiment can also have the following configuration.
[Additional Notes]
[Item 1]
An information processing device comprising: a processing unit that acquires planned data on the amount of power generation of a power generation facility that generates electricity that can be traded in the electricity market, and when a first event related to a change in the amount of power generation on a target day is detected based on the planned data, estimates an amount of fluctuation in the market price of the electricity market on the target day due to the first event.
[Item 2]
2. The information processing apparatus according to item 1, wherein the first event is an amount of change in the amount of power generation on the target day exceeding a threshold.
[Item 3]
3. The information processing device according to item 1 or 2, wherein the first event is that a difference between an amount of change in the amount of power generation on the target day and an amount of change in the amount of power generation on a trading day prior to the target day exceeds a threshold.
[Item 4]
The processing unit estimates a first market price, which is a market price in the case where the first event does not occur on the target date, based on performance data of the market price in the electricity market;
The information processing device according to any one of items 1 to 3, wherein the processing unit estimates a second market price, which is a market price in the event that the first event occurs on the target date, based on the first market price and the fluctuation amount.
[Item 5]
5. The information processing device according to item 4, wherein the processing unit calculates the second market price using a function including the first market price and the fluctuation amount.
[Item 6]
The plan data includes at least one of a cause of a fluctuation in the amount of power generated by the power generation facility, a type of the fluctuation, and a type of the power generation facility;
an input unit for allowing a user to specify at least one piece of information including a cause of the fluctuation, a type of the fluctuation, and a type of the power generation facility;
The processing unit determines whether or not the first event will occur based on a change in the total amount of power generation of the power generation facility that matches the information specified by the user.
6. The information processing device according to any one of items 1 to 5.
[Item 7]
7. The information processing device according to any one of items 1 to 6, wherein the first event is an event related to a change in a total amount of power generated by a plurality of power generation facilities present in an area covered by the electricity market.
[Item 8]
8. The information processing device according to item 7, wherein the first event is an amount of change in the total amount of power generation exceeding a threshold.
[Item 9]
Item 9. The information processing device according to item 7 or 8, wherein the first event is a difference between a sum of the amount of change in the amount of power generation on the target day and a sum of the amount of change in the amount of power generation on a trading day prior to the target day exceeding a threshold.
[Item 10]
the processing unit calculates a coefficient representing an estimated value of a fluctuation amount of the market price when an electricity trading volume increases by a unit amount based on actual data of the market price;
The processing unit calculates the fluctuation amount by multiplying the difference between the amount of change in the amount of power generation on the target date and the amount of change in the amount of power generation on a trading date prior to the target date by the coefficient.
10. The information processing device according to any one of items 1 to 9.
[Item 11]
The processing unit acquires the market price performance data and weather forecast data for a period prior to the target date,
6. The information processing device according to item 4 or 5, wherein the processing unit acquires a weather forecast value for the target day, and calculates the first market price for the target day based on the acquired weather forecast value, the performance data, and the weather forecast data.
[Item 12]
the processing unit acquires actual market price data and weather forecast data, each of the actual market price data and the weather forecast data including a first period before the target date and a second period before the first period;
6. The information processing device according to item 4 or 5, wherein the processing unit acquires a weather forecast value for the target day, and if a second weather-related event is detected based on the weather forecast value, calculates the first market price based on the performance data for the second period and the weather forecast data for the second period, and if the second event is not detected, calculates the first market price based on the performance data for the first period and the weather forecast data for the first period.
[Item 13]
The information processing device according to item 3, wherein the processing unit acquires first plan data acquired at a first point in time regarding the amount of power generation of the power generation facility on the target day, and the plan data acquired at a second point in time prior to the first point in time regarding the amount of power generation of the power generation facility on the target day, and calculates the difference based on the first plan data and the second plan data.
[Item 14]
4. The information processing device according to item 2 or 3, wherein the amount of change in the amount of power generation is an amount of decrease in the amount of power generation.
[Item 15]
4. The information processing device according to item 3, wherein the amount of change in the amount of power generation is an amount of decrease in the amount of power generation.
[Item 16]
9. The information processing device according to item 8, wherein the amount of change in the total amount of power generation is an amount of decrease in the total amount of power generation.
[Item 17]
10. The information processing device according to item 9, wherein the amount of change in the total amount of power generation is an amount of decrease in the total amount of power generation.
[Item 18]
Obtaining planned data on the amount of electricity generated by power generation facilities that generate electricity that can be traded in the electricity market;
when a first event relating to a change in power generation amount on a target day is detected based on the planning data, a fluctuation amount of the market price of the electricity market on the target day caused by the first event is estimated.
[Item 19]
Obtaining planned data on the amount of power generated by a power generation facility that generates power that can be traded in the power market;
and when a first event relating to a change in power generation amount on a target day is detected based on the planning data, estimating an amount of fluctuation in the market price of the electricity market on the target day due to the first event.

Reference Signs List 1 Weather forecast system 2 Electricity market performance value management system 3 Power plant operation status management system 10 Price rise prediction device 100 Processing unit 100A Price rise prediction function 100B Normal time prediction function 100C Electricity market prediction function 110 Power generation facility shutdown plan data pre-processing unit (pre-processing unit)
120 Trend analysis unit 130 Price increase determination unit (cause of power generation equipment)
140 Price sensitivity calculation unit 150 Price increase prediction unit 160 Price increase determination unit (weather cause)
170 Data extraction range determination unit 180 Normal price prediction data storage unit 181 Normal price prediction unit 182 Prediction result calculation unit 200 Data acquisition unit 210 Weather forecast acquisition unit 220 Electricity market performance acquisition unit 230 Power generation facility shutdown plan acquisition unit 300 Data storage unit 310 Weather forecast storage unit 320 Electricity market performance storage unit 330 Power generation facility shutdown plan storage unit 400 Input unit 500 Output unit 600 Computer device 601 CPU (processor)
602 Input interface 603 Display device 604 Communication device 605 Main memory device 606 External memory device 607 Bus

Claims (19)

An information processing device comprising: a processing unit that acquires planned data on the amount of power generation of a power generation facility that generates electricity that can be traded in the electricity market, and when a first event related to a change in the amount of power generation on a target day is detected based on the planned data, estimates an amount of fluctuation in the market price of the electricity market on the target day due to the first event. The information processing apparatus according to claim 1 , wherein the first event is an amount of change in the amount of power generation on the target day exceeding a threshold value. The information processing apparatus according to claim 1 , wherein the first event is an occurrence of a difference between an amount of change in the amount of power generation on the target day and an amount of change in the amount of power generation on a trading day prior to the target day exceeding a threshold value. The processing unit estimates a first market price, which is a market price in the case where the first event does not occur on the target date, based on performance data of the market price in the electricity market;
The information processing device according to claim 1 , wherein the processing unit estimates a second market price, which is a market price when the first event occurs on the target date, based on the first market price and the fluctuation amount. The information processing device according to claim 4 , wherein the processing unit calculates the second market price using a function including the first market price and the fluctuation amount. The plan data includes at least one of a cause of a fluctuation in the amount of power generated by the power generation facility, a type of the fluctuation, and a type of the power generation facility;
an input unit for allowing a user to specify at least one piece of information including a cause of the fluctuation, a type of the fluctuation, and a type of the power generation facility;
The processing unit determines whether or not the first event will occur based on a change in the total amount of power generation of the power generation facility that matches the information specified by the user.
The information processing device according to claim 1 . The information processing device according to claim 1 , wherein the first event is an event related to a change in a total amount of power generated by a plurality of power generation facilities present in an area covered by the electricity market. The information processing apparatus according to claim 7 , wherein the first event is an amount of change in the total amount of power generation exceeding a threshold value. The information processing device according to claim 7 , wherein the first event is a difference between a sum of the amount of change in the amount of power generation on the target day and a sum of the amount of change in the amount of power generation on a trading day prior to the target day exceeding a threshold value. the processing unit calculates a coefficient representing an estimated value of a fluctuation amount of the market price when an electricity trading volume increases by a unit amount based on actual data of the market price;
The processing unit calculates the fluctuation amount by multiplying the difference between the amount of change in the amount of power generation on the target date and the amount of change in the amount of power generation on a trading date prior to the target date by the coefficient.
The information processing device according to claim 1 . The processing unit acquires the market price performance data and weather forecast data for a period prior to the target date,
The information processing device according to claim 4 , wherein the processing unit acquires a weather forecast value for the target day, and calculates the first market price for the target day based on the acquired weather forecast value, the performance data, and the weather forecast data. the processing unit acquires actual market price data and weather forecast data, each of the actual market price data and the weather forecast data including a first period before the target date and a second period before the first period;
5. The information processing device of claim 4, wherein the processing unit acquires a weather forecast value for the target date, and if a second weather-related event is detected based on the weather forecast value, calculates the first market price based on the actual data for the second period and the weather forecast data for the second period, and if the second event is not detected, calculates the first market price based on the actual data for the first period and the weather forecast data for the first period. The information processing device of claim 3, wherein the processing unit acquires first planned data acquired at a first point in time regarding the amount of power generation of the power generation facility on the target day, and the planning data acquired at a second point in time prior to the first point in time regarding the amount of power generation of the power generation facility on the target day, and calculates the difference based on the first planned data and the second planned data. The information processing apparatus according to claim 2 , wherein the amount of change in the amount of power generation is an amount of decrease in the amount of power generation. The information processing device according to claim 3 , wherein the amount of change in the amount of power generation is an amount of decrease in the amount of power generation. The information processing device according to claim 8 , wherein the amount of change in the total amount of power generation is an amount of decrease in the total amount of power generation. The information processing device according to claim 9 , wherein the amount of change in the total amount of power generation is an amount of decrease in the total amount of power generation. Obtaining planned data on the amount of electricity generated by power generation facilities that generate electricity that can be traded in the electricity market;
when a first event relating to a change in power generation amount on a target day is detected based on the planning data, a fluctuation amount of the market price of the electricity market on the target day caused by the first event is estimated. Obtaining planned data on the amount of power generated by a power generation facility that generates power that can be traded in the power market;
and when a first event relating to a change in power generation amount on a target day is detected based on the planning data, estimating an amount of fluctuation in the market price of the electricity market on the target day due to the first event.