Power generation amount prediction device, future power generation amount calculation process program, future power generation amount calculation processing method, and display method

Abstract

To provide a power generation amount prediction device capable of predicting a future power generation amount such that a ratio calculated using data on a power generation amount predicted in the past and data on a power generation amount measured in the past is within a predetermined threshold range.SOLUTION: A power generation amount prediction device comprises a processor, which calculates a future power generation amount such that a ratio related to the difference between a future power generation amount indicative of a predicted value of a future power generation amount calculated based upon an error between past power generation amount measurement data indicative of a measured value of a power generation amount in a past predetermined period and past power generation amount prediction data indicative of a predicted value of a power generation amount in the past predetermined period, and a measured value for the future power generation amount is within a predetermined threshold range.SELECTED DRAWING: Figure 2

Description

[Technical Field]

[0001] The present invention relates to a power generation amount prediction device, a future power generation amount calculation processing program, a future power generation amount calculation processing method, and a display method. [Background technology]

[0002] Patent Document 1 discloses a solar radiation correction method that acquires solar radiation data, acquires meteorological data, and corrects the acquired solar radiation data based on the acquired meteorological data. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] International Publication No. 2022 / 024960 Summary of the Invention [Problem to be solved by the invention]

[0004] An imbalance system has been introduced to ensure a stable supply of electricity. Under the imbalance system, electric power companies that supply electricity are required to prepare power generation plans that outline how much power they plan to generate on a future day, such as the next day.

[0005] The present disclosure aims to provide a power generation prediction device that can predict future power generation with reduced variation in the ratio calculated using data on previously predicted power generation and data on previously measured power generation. [Means for solving the problem]

[0006] In order to achieve the above object, the power generation prediction device according to the first aspect includes a processor, and the processor calculates a future power generation amount, which indicates a predicted value of future power generation amount calculated based on an error between past power generation actual data indicating the actual measured value of power generation amount in a predetermined period in the past and past power generation prediction data indicating the predicted value of power generation amount in the predetermined period in the past, and the ratio of the difference between the actual measured value and the future power generation amount is within a predetermined threshold range. [Effects of the Invention]

[0007] The first aspect of the power generation prediction device has the advantage of being able to predict future power generation such that the ratio calculated using data on previously predicted power generation and data on previously measured power generation falls within a predetermined threshold range. [Brief explanation of the drawings]

[0008] [Figure 1] 1 is a diagram illustrating an outline of a power generation amount prediction device according to an embodiment of the present invention. [Figure 2] FIG. 2 is a block diagram showing an example of a hardware configuration of the power generation amount prediction device. [Figure 3] 10 is a flowchart illustrating an example of processing executed by a future power generation amount calculation processing program. [Figure 4] 10 is a sub-flowchart showing an example of a process for searching for a correction value 60. [Figure 5] 10 is a sub-flowchart showing an example of a process for searching for a correction value 70. [Figure 6] 10 is a flowchart illustrating an example of processing executed by an imbalance power generation ratio calculation processing program. [Figure 7] FIG. 7 is a diagram illustrating the details of the processing shown in FIG. 6. [Figure 8] FIG. 10 is a diagram showing a line graph of the monthly integrated imbalance power generation ratio for fiscal year N, displayed on the display unit. [Figure 9] This shows a graph of the monthly cumulative imbalance power generation ratio 90 for solar power plant A for fiscal year 2022. [Figure 10] This shows a graph of the monthly cumulative imbalance power generation ratio of 95 for solar power plant B for fiscal year 2022. [Figure 11] This shows a graph of the monthly cumulative imbalance power generation ratio of 95 for solar power plant B for fiscal year 2022. DETAILED DESCRIPTION OF THE INVENTION

[0009] Hereinafter, this embodiment will be described in detail with reference to the drawings.

[0010] FIG. 1 is a diagram illustrating an outline of a power generation amount prediction device 10 according to this embodiment.

[0011] The power generation prediction device 10 according to this embodiment calculates a predicted value of future power generation per fixed time period, for example, of a solar power plant A, using a future power generation calculation processing program 100 (described later) (see FIG. 3). The fixed time period may be, for example, 30 minutes, 60 minutes, 24 hours, etc. Hereinafter, the predicted value of future power generation per fixed time period calculated by the power generation prediction device 10 will be referred to as "future power generation 30."

[0012] Furthermore, the power generation prediction device 10 can calculate the imbalance power generation ratio of, for example, the solar power plant A using an imbalance power generation ratio calculation processing program 200, which will be described later, and display the imbalance power generation ratio in the form of a graph to the user (see FIGS. 6 and 8). Hereinafter, the monthly imbalance power generation ratio calculated by the power generation prediction device 10 will be referred to as a "monthly accumulated imbalance power generation ratio 88."

[0013] Here, the imbalance is a value indicating the difference between the value of the power generation amount according to the power generation plan and the value of the actual power generation amount. As will be described in detail later, the monthly cumulative imbalance power generation ratio 88 is a value indicating the ratio related to the monthly imbalance of the solar power plant.

[0014] The configuration of the power generation amount prediction device 10 according to this embodiment will be described with reference to Fig. 2. Fig. 2 is a block diagram showing an example of the hardware configuration of the power generation amount prediction device 10. Note that, for example, a PC (Personal Computer), a tablet terminal, etc. may also be used as the power generation amount prediction device 10.

[0015] The power generation amount prediction device 10 according to this embodiment includes a processor 11, a display unit 12, an operation unit 13, a storage unit 14, and a communication unit 15. The processor 11 includes a CPU 11A, a ROM 11B, a RAM 11C, a non-volatile memory 11D, and an input / output interface (I / O) 11E.

[0016] The processor 11 is a central processing unit, and reads out, for example, a future power generation amount calculation processing program 100, an imbalance power generation ratio calculation processing program 200, and various other programs from, for example, the storage unit .

[0017] The processor 11 uses the RAM 11C as a work area and executes the read program to control the power generation amount prediction device 10.

[0018] The CPU 11A, the ROM 11B, the RAM 11C, and the nonvolatile memory 11D are connected to the I / O unit 11E via a bus so as to be able to communicate with each other.

[0019] The display unit 12 is configured by, for example, a liquid crystal display or an organic EL display, etc. The display unit 12 displays information and the like according to user operations and the processing of the power generation amount prediction device 10, etc.

[0020] The operation unit 13 includes operation keys, operation buttons, a power button, and the like provided in the power generation amount prediction device 10. The operation unit 13 and the display unit 12 may be integrated into one unit using, for example, a touch panel.

[0021] The storage unit 14 is configured by, for example, an SSD (Solid State Drive) or an HDD (Hard Disk Drive). The storage unit 14 stores a future power generation amount calculation processing program 100. The storage unit 14 also stores an imbalance power generation ratio calculation processing program 200.

[0022] Furthermore, the storage unit 14 may store future power generation actual data 40 and power generation actual measurement data 50. Here, the future power generation actual data 40 indicates the future power generation 30 calculated in the past by the power generation prediction device 10. The future power generation 30 calculated by the power generation prediction device 10 is accumulated as the future power generation actual data 40. Furthermore, the power generation actual measurement data 50 indicates the actual measurement value of the actual power generation in, for example, solar power plant A. The actual measurement value of the power generation in solar power plant A is accumulated as the power generation actual measurement data 50. The future power generation actual data 40 is an example of "past power generation prediction data". Furthermore, the power generation actual measurement data 50 is an example of "past power generation actual measurement data".

[0023] The communication unit 15 is an interface for communicatively connecting to, for example, an external device, a network, etc. The communication unit 15 uses a communication standard such as Wi-Fi (registered trademark), Bluetooth (registered trademark), or LAN (Local Area Network).

[0024] The processor 11, the display unit 12, the operation unit 13, the storage unit 14, and the communication unit 15 are electrically connected by a system bus.

[0025] Next, with reference to FIG. 3, the processing of the future power generation amount calculation processing program 100 executed by the CPU 11A will be described.

[0026] 3 is a flowchart showing an example of processing executed by the CPU 11A in accordance with the future power generation amount calculation processing program 100. The processing shown in FIG. 3 is started by an instruction from a user who uses the power generation amount prediction device 10, for example, a manager of a power plant.

[0027] In step S100, the CPU 11A accepts a target date for calculating the future power generation amount 30. Hereinafter, the target date for calculating the future power generation amount 30 will be referred to as a "target date." For example, when calculating the future power generation amount 30 for August 2, 2023 as of August 1, 2023, August 2, 2023 is accepted as the target date. The target date is set by the user.

[0028] Although details will be described later, the CPU 11A may also accept a "target time" indicating a target time for calculating the future power generation amount 30 in addition to the target date.

[0029] In step S102, the CPU 11A determines whether it is possible to acquire the most recent actual future power generation amount data 40 and the most recent measured power generation amount data 50. The most recent actual future power generation amount data 40 may be, for example, actual future power generation amount data 40 from 2 to 12 days or 7 to 14 days before the target date, but it is more preferable if it is actual future power generation amount data 40 from several hours before. Specifically, it is assumed that, as of August 1, 2023, the future power generation amount 30 for August 2, 2023, which is the target date, is calculated. In this case, in step S102, the CPU 11A determines whether or not the future power generation amount record data 40 and the power generation amount actual measurement data 50 for at least one of July 31, 30, 29, 28, 27, 26, 25, 24, 23, 22, 21, and 20, 2023, can be acquired from, for example, the storage unit 14. Note that periods such as 2 to 12 days and 7 to 14 days before the target date are examples of "predetermined past periods."

[0030] If the most recent actual future power generation amount data 40 and the most recent measured power generation amount data 50 can be acquired, the CPU 11A makes a positive determination in step S102 and executes the process of step S104. On the other hand, if at least one of the most recent actual future power generation amount data 40 and the most recent measured power generation amount data 50 cannot be acquired, the CPU 11A makes a negative determination in step S102 and executes the process of step S108.

[0031] In step S104, the CPU 11A acquires the most recent future power generation amount actual data 40 and the most recent power generation amount actual measurement data 50.

[0032] In step S106, the CPU 11A executes a process of searching for the correction value 60 using the acquired most recent future power generation amount result data 40 and the most recent power generation amount actual measurement data 50. Details of the process in step S106 will be described later (see FIG. 4).

[0033] Meanwhile, in step S108, the CPU 11A acquires the monthly future power generation amount performance data 40 of the previous fiscal year and the monthly measured power generation amount data 50 of the previous fiscal year. For example, if the target date is August 2, 2023, the CPU 11A acquires the monthly future power generation amount performance data 40 and the measured power generation amount data 50 of April, May, ..., August, ..., March of fiscal year 2022.

[0034] In step S110, the CPU 11A executes a process of searching for the correction value 70 using the acquired monthly future power generation result data 40 of the previous year and monthly measured power generation data 50 of the previous year. Details of the process of step S110 will be described later (see FIG. 5).

[0035] In step S112, the CPU 11A acquires weather data for the target day. Here, the weather data for the target day refers to data related to weather, including forecasts, for the future target day. Examples of weather data for the target day include forecasted wind direction values, forecasted atmospheric pressure values, forecasted weather values, and forecasted solar radiation values. The weather data for the target day may be acquired, for example, from an external server communicably connected to the power generation amount prediction device 10. The weather data for the target day includes hourly weather data for the target day.

[0036] In step S114, the CPU 11A uses the acquired weather data for the target day to calculate the predicted power generation amount for the target day 20. The predicted power generation amount 20 is a value calculated using existing technology.

[0037] As an example of existing technology, a model constructed by a supervised method using, for example, regression analysis is used. For example, past weather data is used as the training data. Examples of past weather data include values ​​indicating the actual wind direction during a certain period in the past, values ​​indicating the actual air pressure during a certain period in the past, values ​​indicating the actual weather during a certain period in the past, and values ​​indicating the actual amount of solar radiation during a certain period in the past. When the above model is used, the weather data for the target day is input into the model constructed using the above training data, and the predicted power generation amount value 20 is calculated.

[0038] In addition, the predicted value of the amount of solar radiation for the target day may be calculated using existing technology, and the CPU 11A may calculate the predicted power generation amount value 20 by performing a process of converting the predicted value of the amount of solar radiation calculated using existing technology into the amount of power generation.

[0039] In step S116, the CPU 11A applies the correction value 60 or the correction value 70 to the predicted power generation amount 20 to calculate the future power generation amount 30 for the target day.

[0040] In detail, when the search is performed in the process of step S106, the CPU 11A calculates the future power generation amount 30 by applying the correction value 60 to the power generation prediction value 20. On the other hand, when the search is performed in the process of step S110, the CPU 11A calculates the future power generation amount 30 by applying the correction value 70 to the power generation prediction value 20.

[0041] Through the above-described process, the power generation amount prediction device 10 can calculate the future power generation amount 30 for the target day in units of, for example, 30 minutes.

[0042] Although the process of calculating the future power generation amount 30 for a target day has been described above, the power generation amount prediction device 10 may be configured to calculate the future power generation amount 30 for a target time. In this case, for example, in the process of step S100, the CPU 11A accepts a target time for calculating the future power generation amount 30. In detail, for example, when calculating the future power generation amount 30 from 2:00 PM to 2:30 PM as of 1:00 PM on August 1, 2023, the CPU 11A can accept the 30 minutes from 2:00 PM to 2:30 PM as the target time.

[0043] In the process of step S102 when calculating the future power generation amount 30 for the target time, the CPU 11A may further determine whether or not it is possible to acquire the latest future power generation amount record data 40 and the latest power generation amount actual measurement data 50. For example, at 1:00 PM on August 1, 2023, it may determine whether or not it is possible to acquire the future power generation amount record data 40 and the power generation amount actual measurement data 50 from, for example, 1 to 6 hours before 1:00 PM on August 1, 2023 as the latest future power generation amount record data 40 and the latest power generation amount actual measurement data 50. If the latest future power generation amount record data 40 and the latest power generation amount actual measurement data 50 are available, the search process for the correction value 60 may be performed using the acquired data.

[0044] Next, the process of step S106 will be described in detail with reference to Fig. 4. Fig. 4 is a sub-flowchart showing an example of the process of searching for the correction value 60.

[0045] In step S200, the CPU 11A applies the adjustment value X to the most recent actual future power generation amount data 40 acquired in step S104 to calculate adjusted actual future power generation amount data 40A. Note that, when executing the process of searching for the correction value 60, step S200 is executed repeatedly, but in the first process, an initial value is applied as the adjustment value X. Note that the initial value is, for example, "1.0".

[0046] In step S202, the CPU 11A calculates an error 62 between the adjusted actual future power generation amount data 40A and the most recent measured power generation amount data 50 acquired in step S106. The error between the adjusted actual future power generation amount data 40A and the most recent measured power generation amount data 50 is calculated using an index such as RMSE (Root Mean Squared Error) or MAE (Mean Absolute Error). The error 62 is calculated using, for example, the adjusted actual future power generation amount data 40A and the measured power generation amount data 50 at the same time on the same day.

[0047] In step S204, the CPU 11A determines whether or not the correction value 60 can be acquired. Here, the correction value 60 is defined as the minimum value of the error 62 between the adjusted future power generation amount actual data 40A and the most recent power generation amount actual measurement data 50. The determination in step S204 is affirmative when the correction value 60 can be acquired, for example, because a predetermined condition is satisfied. The details of step S204 will be described later.

[0048] If the determination in step S204 is positive, the CPU 11A executes the process of step S208. If the determination in step S204 is negative, the CPU 11A executes the process of step S206.

[0049] In step S206, CPU 11A updates adjustment value X to a new value. Note that adjustment value X is updated using, for example, gradient descent or Newton's method. After updating adjustment value X, CPU 11A executes the process of step S200 again.

[0050] Here, the processing of step S204 will be described in detail. The "predetermined condition" mentioned above in step S204 is, for example, a condition for determining whether the processing of steps S200, S202, and S206 has been repeatedly executed until the threshold value is reached. In this case, if the processing of steps S200, S202, and S206 has been executed until the threshold value is reached, the CPU 11A makes a positive determination. Alternatively, the predetermined condition may be specified as, for example, repeatedly executing the processing of steps S200, S202, and S206 until the adjustment value X reaches the threshold value. In this way, if the processing has been repeatedly executed until the predetermined condition is met, the CPU 11A acquires the smallest value of the calculated errors 62 as the correction value 60 in step S204.

[0051] In step S208, CPU 11A stores the acquired correction value 60 in storage unit 14, for example.

[0052] In this manner, the CPU 11A searches for the minimum value of the error 62 between the adjusted future power generation actual data 40A and the most recent power generation actual measurement data 50.

[0053] As described above, according to the processes of steps S200 to S208, the correction value 60 can be calculated using the most recent future power generation amount actual data 40 and the most recent power generation amount actual measurement data 50. The correction value 60 may be calculated for each 30-minute period of the target day. That is, the CPU 11A may calculate the correction value 60 from 8:00 AM to 8:30:00 AM on August 2, 2023, then calculate the correction value 60 until 9:00 AM, and so on, calculating the correction value 60 for each 30-minute period of one day.

[0054] Next, the process of step S110 will be described in detail with reference to Fig. 5. Fig. 5 is a sub-flowchart showing an example of the process of searching for the correction value 70.

[0055] In step S300, the CPU 11A acquires the future power generation result data 40 and the power generation actual measurement data 50 for month M, which indicates a certain month, from the future power generation result data 40 and the power generation actual measurement data 50 for each month of the previous year acquired in step S108.

[0056] In step S302, the CPU 11A applies the adjustment value X to the future power generation actual data 40 for month M to calculate the adjusted future power generation actual data 40B for month M. In the first process of step S302, an initial value is applied as the adjustment value X. In addition, the initial value is, for example, "1.0."

[0057] In step S304, an error 72 is calculated between the adjusted future power generation amount actual data 40B for month M and the power generation amount actual measurement data 50 for month M. The error 72 between the adjusted future power generation amount actual data 40B for month M and the power generation amount actual measurement data 50 for month M is calculated using an index such as RMSE (Root Mean Squared Error) or MAE (Mean Absolute Error).

[0058] In step S306, the CPU 11A determines whether or not the correction value 70 can be acquired. Here, the correction value 70 is defined as the minimum value of the error 72 between the adjusted future power generation amount actual data 40B for month M and the power generation amount actual measurement data 50 for month M. The determination in step S306 is affirmative, for example, when a predetermined condition is satisfied. The details of step S306 will be described later.

[0059] If the determination in step S306 is positive, CPU 11A executes the process of step S310. If the determination in step S306 is negative, CPU 11A executes the process of step S308.

[0060] In step S308, the CPU 11A updates the adjustment value X to a new value. The adjustment value X is updated using, for example, the gradient descent method or the Newton method. After updating the adjustment value X, the CPU 11A executes the process of step S302 again.

[0061] Here, the processing of step S306 will be described in detail. The predetermined condition in the processing of step S306 is, for example, a condition for determining whether the processing of steps S302, S304, and S308 has been repeatedly executed until the threshold value is reached. In this case, if the processing of steps S302, S304, and S308 has been executed until the threshold value is reached, the CPU 11A makes a positive determination. Alternatively, the predetermined condition may be, for example, that the processing of steps S302, S304, and S308 be repeatedly executed until the adjustment value X reaches the threshold value. In this way, if the processing has been repeatedly executed until the predetermined condition is met, the CPU 11A acquires the smallest value of the calculated errors 72 as the correction value 70 in step S306.

[0062] In step S310, CPU 11A stores the calculated correction value 70 in storage unit 14, for example, as correction value 70 for month M.

[0063] In step S312, the CPU 11A determines whether or not the correction values ​​70 for all months have been calculated using the monthly future power generation amount record data 40 and the actual power generation amount data 50 for the previous year acquired in step S108. Specifically, it is assumed that the future power generation amount record data 40 and the actual power generation amount data 50 for, for example, 12 months have been acquired in step S108. In this case, the CPU 11A executes the processes of steps S302 to S310 using the monthly future power generation amount record data 40 and the actual power generation amount data 50 to calculate the 12 correction values ​​70 for 12 months, i.e., 12 correction values ​​70. If the correction values ​​70 for all the acquired months have been calculated, the CPU 11A makes a positive determination in step S312 and ends the process of searching for the correction values ​​70.

[0064] On the other hand, if the processes of steps S302 to S310 have not been executed for all the acquired monthly data (the future power generation amount result data 40 and the power generation amount actual measurement data 50 for the previous year), the CPU 11A executes the process of step S314.

[0065] In step S314, the CPU 11A updates month M to the next month, for example. Specifically, for example, it is assumed that in step S300, the future power generation amount record data 40 and the power generation amount actual measurement data 50 for April 2022 have been acquired as month M. In this case, the processes of steps S302 to S310 are executed using the future power generation amount record data 40 and the power generation amount actual measurement data 50 for April 2022. In the process of step S314, the CPU 11A updates month M to May, which is the month following April.

[0066] In step S316, CPU 11A sets the adjustment value X updated in step S308 to the initial value, for example. After executing the process of step S316, CPU 11A executes the process of step S300.

[0067] As described above, the CPU 11A searches for the minimum value of the error 72 between the adjusted future power generation result data 40B for the Mth month and the power generation actual measurement data 50 for the Mth month.

[0068] Although the above description has been made regarding the process when data for the previous year is acquired in step S108, for example, future power generation performance data 40 and power generation actual measurement data 50 for each month for the past several years may be acquired in step S108. In this case, the process in Fig. 5 may be configured to calculate the correction value 70 for each month for all acquired years.

[0069] As described above, according to the processing of steps S300 to S316, the correction value 70 for each month can be calculated using the actual future power generation amount data 40 for each month of the previous year and the actual power generation amount data 50 for each month of the previous year. Also, for example, the correction value 70 for each day of each month may be calculated using the actual future power generation amount data 40 for each day of the previous year and the actual power generation amount data 50 for each day of the previous year.

[0070] 3 to 5, it is possible to calculate a correction value (correction value 60 or correction value 70) according to the obtainable future power generation amount record data 40 and the obtainable power generation amount actual measurement data 50. Furthermore, according to this embodiment, it is possible to calculate the future power generation amount 30 using a correction value according to the obtainable future power generation amount record data 40 and the obtainable power generation amount actual measurement data 50.

[0071] Here, the details of the process in step S116 of FIG. 3 , in which the CPU 11A applies the correction value 70 to the power generation forecast value 20 to calculate the future power generation amount 30, will be described. In step S116, the correction value 70 applied by the CPU 11A to the power generation forecast value 20 is the correction value 70 for the month corresponding to the month including the target day, or the correction value 70 for the corresponding day in the month including the target day. Specifically, for example, when the power generation forecast value 20 for August 2, 2023 is calculated as the target day, the CPU 11A applies the correction value 70 for August, or the correction value 70 for the day in August corresponding to the target day, among the monthly correction values ​​70 for past years, to the power generation forecast value 20 to calculate the future power generation amount 30. In this way, the future power generation amount 30 can be calculated using the correction value 70 for the month including the target day or the correction value 70 for a past day in a past month.

[0072] Next, the processing of the imbalance power generation ratio calculation processing program 200 executed by the CPU 11A will be described with reference to FIGS.

[0073] Fig. 6 is a flowchart showing an example of processing executed by the CPU 11A in accordance with the imbalance power generation ratio calculation processing program 200. Fig. 7 is a diagram for explaining the details of the processing shown in Fig. 6.

[0074] The imbalance power generation ratio calculation processing program 200 is executed in response to a user instruction. The imbalance power generation ratio calculation processing program 200 allows the user to view the monthly integrated imbalance power generation ratio for a target year, which may represent, for example, a certain past year.

[0075] In step S400 of Fig. 6, the CPU 11A accepts a target year. In the following description, it is assumed that the CPU 11A accepts fiscal year N as the target year. The target year is specified by the user. For example, if the user uses the imbalance power generation ratio calculation processing program 200 as of August 1, 2023, the user can specify a year before fiscal year 2022 as the target year.

[0076] In step S402, CPU 11A acquires future power generation amount actual data 40 for month L of year N. Column 510 of table 500 shown in Fig. 7 indicates hourly future power generation amount actual data 40 for a certain day in month L of year N acquired in step S402.

[0077] In step S404, CPU 11A acquires power generation amount actual measurement data 50 for month L of year N. Column 520 of table 500 shown in Fig. 7 indicates the acquired power generation amount actual measurement data 50 for each hour of a certain day in month L of year N.

[0078] In step S406, CPU 11A calculates a difference between forecast and actual value 80 using actual future power generation amount data 40 for fiscal year N and actual power generation amount data 50 for fiscal year N. The difference between forecast and actual value 80 is a value indicating the difference between actual future power generation amount data 40 and actual power generation amount data 50. The difference between forecast and actual value 80 is calculated for each day of month L of fiscal year N, for example, in 30-minute increments. Column 530 of table 500 shown in FIG. 7 indicates the difference between forecast and actual value 80 for each hour of a certain day in month L of fiscal year N. Each value of the difference between forecast and actual value 80 shown in column 530 corresponds to the difference between the deficit imbalance value and the surplus imbalance value shown in table 500, in accordance with a general imbalance charge calculation for a power generation forecast.

[0079] In step S408, CPU 11A calculates daily accumulated difference between budget and actual results 82. Here, daily accumulated difference between budget and actual results 82 is a value obtained by accumulating difference between budget and actual results 80 on a daily basis. In other words, daily accumulated difference between budget and actual results 82 is the sum of the calculated difference between budget and actual results 80 per day. Column 610 of table 600 in FIG. 7 shows daily accumulated difference between budget and actual results 82, which is the sum of difference between budget and actual results 80 shown in column 530 of table 500.

[0080] In step S410, CPU 11A calculates monthly accumulated difference between forecast and actual results 84 for month L. Here, monthly accumulated difference between forecast and actual results 84 for month L is a value calculated by integrating daily accumulated difference between forecast and actual results 82 by the number of days in month L. In other words, monthly accumulated difference between forecast and actual results 84 for month L is a value calculated by summing up daily accumulated difference between forecast and actual results 82 for the number of days in month L. Specifically, for example, if month L has 31 days, 31 daily accumulated difference between forecast and actual results 82 are added together to calculate monthly accumulated difference between forecast and actual results 84 for month L. Table 700 in FIG. 7 shows monthly accumulated difference between forecast and actual results 84 for month L of year N.

[0081] In step S412, the CPU 11A uses the power generation amount actual measurement data 50 for the Nth year to calculate a monthly integrated actual measurement value 86 indicating the integrated value of the power generation amount actual measurement data 50 for the Lth month of the Nth year. Table 700 in Fig. 7 shows the monthly integrated actual measurement value 86 for the Lth month of the Nth year.

[0082] In step S414, the CPU 11A calculates a monthly integrated imbalance power generation ratio 88 for month L, which indicates a value obtained by dividing the monthly integrated difference between forecast and actual value 84 by the monthly integrated actual measured value 86. The monthly integrated imbalance power generation ratio 88 is an example of "a ratio related to the difference between the future power generation amount and the actual measured value of the future power generation amount."

[0083] In step S416, the CPU 11A determines whether the monthly integrated imbalance power generation ratios 88 for all months of the Nth year have been calculated. If the monthly integrated imbalance power generation ratios 88 for all months of the Nth year have been calculated, the CPU 11A executes the process of step S420. On the other hand, if the monthly integrated imbalance power generation ratios 88 for all months of the Nth year have not been calculated, in other words, if there is a monthly integrated imbalance power generation ratio 88 for a month that has not yet been calculated, the CPU 11A executes the process of step S418.

[0084] In the process of step S418, CPU 11A updates month L. Specifically, month L is updated to the next month. Specifically, if month L is set to April, May is set as the new month L. Thereafter, CPU 11A executes the process of step S402.

[0085] In step S420, CPU 11A displays, for example, a graph of the monthly integrated imbalance power generation ratio for fiscal year N. An example of the graph displayed in step S420 is shown in FIG.

[0086] FIG. 8 is a diagram showing a line graph of the monthly integrated imbalance power generation ratio for the Nth fiscal year, which is displayed on the display unit 12 by the processing in step S420.

[0087] 8 shows the monthly accumulated imbalance power generation ratio for fiscal year 2022 calculated using the future power generation amount 30 calculated using the correction value 70. The correction value 70 is a value found using the RMSE error 72. In this way, the CPU 11A can graph the monthly accumulated imbalance power generation ratios 88 for 12 months of fiscal year N calculated in the processing of steps S400 to S418 and display the graph on the display unit 12.

[0088] Graph GB in Figure 8 shows the monthly cumulative imbalance power generation ratio of solar power plant A in fiscal year 2022. The vertical axis represents the cumulative imbalance power generation ratio. The horizontal axis of graph GB represents the month.

[0089] According to the imbalance power generation ratio calculation processing program 200 described above, it is possible to calculate the monthly integrated imbalance power generation ratio for, for example, the past year using the actual data of the future power generation amount 30 (future power generation actual data 40) calculated by the power generation amount prediction device 10 according to this embodiment. Furthermore, according to the imbalance power generation ratio calculation processing program 200, it is possible to graph the changes in the monthly integrated imbalance power generation ratio for, for example, the past year and display it on the display unit 12.

[0090] 9 to 11, the effects of the power generation amount prediction device 10 according to this embodiment will be described. Below, a comparison result is shown between a monthly integrated imbalance power generation ratio 90 calculated using the future power generation amount 30 calculated by the power generation amount prediction device 10 according to this embodiment and a monthly integrated imbalance power generation ratio 95 calculated using a predicted value of the power generation amount calculated without using the power generation amount prediction device 10 according to this embodiment.

[0091] The graphs shown in Figures 9 to 11 below all show the monthly cumulative imbalance power generation ratio for fiscal year 2022. In each graph, the vertical axis shows the cumulative imbalance power generation ratio, and the horizontal axis shows each month.

[0092] Figure 9 shows a graph of the monthly cumulative imbalance power generation ratio 90 for solar power plant A in fiscal year 2022.

[0093] Graph GB shown in Fig. 9 is the same as that shown in Fig. 8. The monthly integrated imbalance power generation ratio 90 shown in graph GB is calculated using the future power generation amount 30 calculated by the power generation amount prediction device 10 according to this embodiment. The future power generation amount 30 is calculated by applying a correction value 70. The correction value 70 is searched for using an error 72 due to RMSE.

[0094] The monthly cumulative imbalance power generation ratio 90 in graph GB is less than or equal to 0 in all months, and the difference between the maximum and minimum values ​​is 0.23.

[0095] Figure 10 shows a graph of the monthly cumulative imbalance power generation ratio 95 for fiscal year 2022, calculated at solar power plant B, which is different from solar power plant A.

[0096] The difference between the maximum and minimum values ​​of the monthly integrated imbalance power generation ratio 95 shown in the graph GC of FIG. 10 is 0.6.

[0097] Graph GD in Figure 11 shows the monthly cumulative imbalance power generation ratio of 95 for fiscal year 2022, calculated from forecast values ​​that are lowered by multiplying the forecast values ​​by a constant to avoid generating a shortage imbalance at solar power plant B.

[0098] The monthly cumulative imbalance power generation ratio 95 of graph GD is less than or equal to 0 in all months, and the difference between the maximum and minimum values ​​is 0.6.

[0099] 9 to 11, the monthly integrated imbalance power generation ratio 90 calculated using the future power generation amount 30 shown in FIG. 9 is less than or equal to 0 in every month, and exhibits a smoother result with less monthly variation, compared to the monthly integrated imbalance power generation ratio 95 shown in FIGS. 10 and 11 without using the future power generation amount 30. This means that by using the future power generation amount 30 calculated by applying the correction value 60 or 70, it is possible to prevent the monthly integrated imbalance power generation ratio from becoming less than 0 and from suddenly increasing or decreasing. In other words, the power generation prediction device 10 according to this embodiment can calculate the future power generation amount 30 that does not cause a shortage imbalance in the monthly integration and reduces the difference in the monthly integrated imbalance power generation ratio 88 between seasons.

[0100] 9-11, it can be seen that the future power generation amount 30 calculated by the power generation amount prediction device 10 according to this embodiment can suppress the value of the monthly integrated imbalance power generation ratio 88 to be greater than or equal to -0.5 and less than or equal to 0. In other words, the power generation amount prediction device 10 according to this embodiment can calculate the future power generation amount 30 such that the value of the monthly integrated imbalance power generation ratio 88 is greater than or equal to -0.5 and less than or equal to 0. This means that the relationship of surplus imbalance > shortage imbalance holds true in every month of the year (a situation in which the shortage imbalance is offset by the surplus imbalance in the monthly integration calculation, and no shortage imbalance occurs), and the variation in the monthly integrated imbalance power generation ratio 88 throughout the year is reduced compared to the monthly imbalance power generation ratio calculated without using this embodiment. The value of the monthly integrated imbalance power generation ratio 88 is an example of a "threshold value," and the value "greater than or equal to -0.5 and less than or equal to 0" is an example of a "predetermined threshold range."

[0101] This threshold range is set to be greater than or equal to -0.5 and less than or equal to 0, preferably greater than or equal to -0.3 and less than or equal to 0. It has been confirmed that the monthly imbalance power generation ratio calculated without using the method according to this embodiment is greater than or equal to 0 in some months, and that the difference between the annual maximum and minimum values ​​is greater than or equal to 0.5 due to variations in weather forecasts used for power generation prediction. Furthermore, it has been confirmed at multiple locations that the monthly integrated imbalance power generation ratio 88 calculated using the method according to this embodiment is in the range of greater than or equal to -0.3 and less than or equal to 0. In light of the above, the threshold range is set to be greater than or equal to -0.5 and less than or equal to 0, preferably greater than or equal to -0.3 and less than or equal to 0.

[0102] The imbalance power generation ratio calculation processing program 200 may be configured to, for example, allow a user to set a threshold range and display the monthly integrated imbalance power generation ratio 88 based on the set threshold range.

[0103] In detail, for example, the monthly integrated imbalance power generation ratio 90 for a certain year calculated by a calculation method different from the monthly integrated imbalance power generation ratio 88 may be acquired first, and when the threshold range is set, if there is a monthly integrated imbalance power generation ratio 90 outside the threshold range for even one month, a notification to that effect may be given, and the monthly integrated imbalance power generation ratio 88 may be acquired and displayed.

[0104] For example, when the threshold range is set to "-0.5 or more and 0 or less," if there is even one month in which the monthly integrated imbalance power generation ratio 90 is outside the range of -0.5 or more and 0 or less, the monthly integrated imbalance power generation ratio 88 may be displayed after being notified of this or the relevant month being highlighted. The threshold range set by the user is an example of a "predetermined threshold range."

[0105] The monthly integrated imbalance power generation ratio 90 is calculated by the imbalance power generation ratio calculation processing program 200, for example.

[0106] The monthly integrated imbalance power generation ratio 90 is a monthly imbalance power generation ratio calculated using a predicted value of future power generation calculated from the error between the measured power generation data 50 for a predetermined period in the past and the actual future power generation data 40 for the predetermined period in the past. More specifically, the monthly integrated imbalance power generation ratio 90 is a monthly imbalance power generation ratio obtained using a predicted value of future power generation calculated by a process that simply determines the error between the measured power generation data 50 and the actual future power generation data 40 for the predetermined period in the past, rather than by the process shown in FIGS. 5 and 6 above. The predicted value of future power generation calculated from the error between the measured power generation data 50 and the actual future power generation data 40 for the predetermined period in the past is an example of a "first future power generation." The monthly integrated imbalance power generation ratio 90 obtained using the first future power generation can be displayed on the display unit 12, for example, in response to a user instruction.

[0107] According to the above configuration, when the user sets the threshold range, it is possible to detect the month in which the monthly integrated imbalance power generation ratio 90 falls outside the threshold range. Furthermore, when the monthly integrated imbalance power generation ratio 90 falls outside the range, the CPU 11A performs the processes shown in Figures 3 to 6 to calculate the monthly integrated imbalance power generation ratio 88 within the threshold range using the future power generation amount 30.

[0108] Furthermore, with the above configuration, the monthly integrated imbalance power generation ratio 90 and the monthly integrated imbalance power generation ratio 90 can be graphed and displayed on the display unit 12, for example, as shown in FIG.

[0109] In the present embodiment, the future power generation amount calculation processing program 100 and the imbalance power generation ratio calculation processing program 200 are described as being stored in the storage unit 14. However, the future power generation amount calculation processing program 100 and the imbalance power generation ratio calculation processing program 200 may be provided by a storage medium such as a CD-ROM, or may be downloaded via a network.

[0110] (Addendum) (Appendix 1) a processor; The processor: calculating a future power generation amount, which indicates a predicted value of a future power generation amount calculated based on an error between past power generation amount actual measurement data, which indicates a measured value of a power generation amount in a predetermined period in the past, and past power generation amount prediction data, which indicates a predicted value of a power generation amount in the predetermined period in the past, and the future power generation amount, such that a ratio of a difference between the future power generation amount and the actual measured value falls within a predetermined threshold range; Power generation forecasting device. (Appendix 2) The predetermined threshold range is a range of -0.5 or more and 0 or less. 2. The power generation prediction device according to claim 1. (Appendix 3) The ratio of the difference between the future power generation amount and the actual measurement value of the future power generation amount is calculated by dividing the value obtained by integrating the difference between the future power generation amount and the actual measurement value over a predetermined period by the value obtained by integrating the actual measurement value over the predetermined period. 3. The power generation prediction device according to claim 1 or 2. (Appendix 4) The processor: calculating a minimum value of the error between the past measured power generation amount data and the past predicted power generation amount data; calculating the future power generation amount using the minimum value; 4. The power generation amount prediction device according to any one of appendices 1 to 3. (Appendix 5) On the computer, calculating a future power generation amount, which indicates a predicted value of a future power generation amount calculated based on an error between past power generation amount actual measurement data, which indicates a measured value of a power generation amount in a predetermined period in the past, and past power generation amount prediction data, which indicates a predicted value of a power generation amount in the predetermined period in the past, and the future power generation amount, such that a ratio of a difference between the future power generation amount and the actual measured value falls within a predetermined threshold range; A future power generation calculation processing program that executes the processing. (Appendix 6) The processor: calculating a future power generation amount, which indicates a predicted value of a future power generation amount calculated based on an error between past power generation amount actual measurement data, which indicates a measured value of a power generation amount in a predetermined period in the past, and past power generation amount prediction data, which indicates a predicted value of a power generation amount in the predetermined period in the past, and the future power generation amount, such that a ratio of a difference between the future power generation amount and the actual measured value falls within a predetermined threshold range; A method for calculating future power generation. (Appendix 7) a ratio obtained using a first future power generation amount, which indicates a predicted value of a future power generation amount calculated from an error between past power generation amount actual measurement data, which indicates a measured value of a power generation amount in a predetermined period in the past, and past power generation amount prediction data, which indicates a predicted value of a power generation amount in the predetermined period in the past, is displayed on a display means so as to be within a predetermined threshold range; Display method. (Appendix 8) The predetermined threshold range is a range of -0.5 or more and 0 or less. Display method as described in Appendix 7. (Appendix 9) If the calculated ratio exceeds the predetermined threshold, a future power generation amount different from the first future power generation amount is calculated; Using the calculated future power generation amount, a ratio that falls within the predetermined threshold range is displayed. Display method as described in Appendix 7 or Appendix 8. [Explanation of symbols]

[0111] 10...power generation prediction device, 11...processor, 12...display unit, 14...memory unit, 20...power generation prediction value, 30...future power generation, 40...future power generation actual data, 40A, 40B...adjusted future power generation actual data, 50...actual power generation data, 60...correction value, 62...error, 70...correction value, 72...error, 80...forecast / actual difference, 82...daily accumulated forecast / actual difference, 84...monthly accumulated forecast / actual difference, 86...monthly accumulated actual value, 88...monthly accumulated imbalance power generation ratio, 90...monthly accumulated imbalance power generation ratio, 100...future power generation calculation processing program, 200...imbalance power generation ratio calculation processing program.

Claims

1. a processor; The processor: calculating a future power generation amount, which indicates a predicted value of a future power generation amount calculated based on an error between past power generation amount actual measurement data, which indicates a measured value of a power generation amount in a predetermined period in the past, and past power generation amount prediction data, which indicates a predicted value of a power generation amount in the predetermined period in the past, and the future power generation amount, such that a ratio of a difference between the future power generation amount and the actual measured value falls within a predetermined threshold range; Power generation forecasting device.

2. The predetermined threshold range is a range of −0.5 or more and 0 or less. The power generation prediction device according to claim 1.

3. The ratio of the difference between the future power generation amount and the actual measurement value of the future power generation amount is calculated by dividing the value obtained by integrating the difference between the future power generation amount and the actual measurement value over a predetermined period by the value obtained by integrating the actual measurement value over the predetermined period. The power generation prediction device according to claim 1.

4. The processor: calculating a minimum value of the error between the past measured power generation amount data and the past predicted power generation amount data; calculating the future power generation amount using the minimum value; The power generation prediction device according to claim 1.

5. On the computer, calculating a future power generation amount, which indicates a predicted value of a future power generation amount calculated based on an error between past power generation amount actual measurement data, which indicates a measured value of a power generation amount in a predetermined period in the past, and past power generation amount prediction data, which indicates a predicted value of a power generation amount in the predetermined period in the past, and the future power generation amount, such that a ratio of a difference between the future power generation amount and the actual measured value falls within a predetermined threshold range; A future power generation calculation processing program that executes the processing.

6. The processor: calculating a future power generation amount, which indicates a predicted value of a future power generation amount calculated based on an error between past power generation amount actual measurement data, which indicates a measured value of a power generation amount in a predetermined period in the past, and past power generation amount prediction data, which indicates a predicted value of a power generation amount in the predetermined period in the past, and the future power generation amount, such that a ratio of a difference between the future power generation amount and the actual measured value falls within a predetermined threshold range; A method for calculating future power generation.

7. a ratio obtained using a first future power generation amount, which indicates a predicted value of a future power generation amount calculated from an error between past power generation amount actual measurement data, which indicates a measured value of a power generation amount in a predetermined period in the past, and past power generation amount prediction data, which indicates a predicted value of a power generation amount in the predetermined period in the past, is displayed on a display means so as to be within a predetermined threshold range; Display method.

8. The predetermined threshold range is a range of −0.5 or more and 0 or less. The display method according to claim 7.

9. If the calculated ratio exceeds the predetermined threshold range, a future power generation amount different from the first future power generation amount is calculated; Using the calculated future power generation amount, a ratio that falls within the predetermined threshold range is displayed. The display method according to claim 7.

Images