Output prediction device, output prediction method, and program

Abstract

This output prediction device comprises a first acquisition unit, a second acquisition unit, a power generation output prediction unit, and an output prediction unit. The first acquisition unit acquires weather prediction data including the prediction result of distribution of wind blowing in a wind power plant having a plurality of wind power generation devices. The second acquisition unit acquires operation plan data indicating an operation plan for each of the plurality of wind power generation devices. The power generation output prediction unit uses the weather prediction data and the operation plan data to generate power generation output prediction data indicating the prediction results of a power generation output of each of the plurality of wind power generation devices. The output prediction unit uses the power generation output prediction data to generate output prediction data indicating the prediction result of the output amount of electrical power from the wind power generation plant.

Description

Output prediction device, output prediction method, and program

[0001] The present invention relates to an output prediction device, an output prediction method, and a program.

[0002] Wind power generation is one type of renewable energy (hereinafter also referred to as "renewable energy"). For example, when considering the construction location of a wind power plant, predicting power generation output is important. For example, Patent Literature 1 discloses the following system. First, the system references weather actual data published by a second institution and calculates actual values ​​of weather elements for each of multiple partitions including an area based on actual values ​​of weather elements for each segment in the partition. The system also references weather actual data published by a first institution and creates a model based on the actual values ​​of weather elements calculated for each partition and the actual values ​​of renewable energy power generation output in the area, with the weather element values ​​for each partition as input and the renewable energy power generation output value for the area as output. The system references weather forecast data published by a second institution and calculates actual values ​​of weather elements for each of multiple partitions including an area based on predicted values ​​of weather elements for each segment in the partition, and then calculates a predicted value of renewable energy power generation output based on the predicted values ​​of weather elements calculated for each partition and the created model.

[0003] Japanese Patent Application Laid-Open No. 2022-121028

[0004] It is important to improve the accuracy of prediction of the amount of power output from a wind power plant. One example of a problem to be solved by the present invention is to improve the accuracy of prediction of the amount of power output from a wind power plant.

[0005] According to one aspect of the present invention, there is provided an output prediction device comprising: a first acquisition unit that acquires weather forecast data including forecast results regarding wind in an area including a wind power plant having a plurality of wind power generation devices; a second acquisition unit that acquires operation plan data indicating operation plans for each of the plurality of wind power generation devices; a power output prediction unit that uses the weather forecast data and the operation plan data to generate power output prediction data indicating forecast results of power output for each of the plurality of wind power generation devices; and a power prediction unit that uses the power output prediction data to generate output prediction data indicating forecast results of power output from the wind power plant.

[0006] According to one aspect of the present invention, there is provided an output prediction method in which a computer acquires weather forecast data including predicted results of wind distribution in a wind power plant having a plurality of wind power generation devices, acquires operation plan data indicating operation plans for each of the plurality of wind power generation devices, uses the weather forecast data and the operation plan data to generate power output prediction data indicating predicted results of power output for each of the plurality of wind power generation devices, and uses the power output prediction data to generate output prediction data indicating predicted results of the amount of power output from the wind power plant.

[0007] According to one aspect of the present invention, there is provided a program for causing a computer to have: a first acquisition unit that acquires weather forecast data including predicted results of wind distribution in a wind power plant having a plurality of wind power generation devices; a second acquisition unit that acquires operation plan data that indicates an operation plan for each of the plurality of wind power generation devices; a power output prediction unit that uses the weather forecast data and the operation plan data to generate power output prediction data that indicates predicted results of power output for each of the plurality of wind power generation devices; and an output prediction unit that uses the power output prediction data to generate output prediction data that indicates predicted results of power output from the wind power plant.

[0008] According to one aspect of the present invention, the accuracy of prediction of the amount of power output from a wind power plant is improved.

[0009] 7 is a diagram showing an example of a usage environment of an output prediction device according to an embodiment. FIG. 8 is a diagram showing an example of a functional configuration of the output prediction device. FIG. 9 is a diagram showing an example of a hardware configuration of the output prediction device. FIG. 10 is a flowchart showing an example of an operation of the output prediction device. FIG. 11 is a diagram showing a usage environment of an output prediction device according to a first modified example. FIG. 12 is a flowchart showing an example of an operation of the output prediction device shown in FIG. 5. FIG. 13 is a diagram for explaining the effect of the output prediction device shown in FIG.

[0010] Hereinafter, embodiments of the present invention will be described with reference to the drawings. In all the drawings, like components are designated by like reference numerals, and the description thereof will be omitted as appropriate.

[0011] 1 is a diagram showing an example of a usage environment of a power output prediction device 10 according to an embodiment. The power output prediction device 10 predicts the power output from a wind power plant 20. The wind power plant 20 is provided with a plurality of wind power generation devices 22. The power output of the wind power plant 20 is basically the sum of the power generated by the plurality of wind power generation devices 22. Note that, as in a modified example described below, the wind power plant 20 may have a power storage facility 24. In this case, the power output of the wind power plant 20 is the sum of the power generated by the wind power generation devices 22 adjusted by the power storage facility 24.

[0012] The wind direction, wind speed, and control conditions of the multiple wind power generation devices 22 differ from one another depending on their installation locations. As a result, the power output of the multiple wind power generation devices 22 differs from one another. The output prediction device 10 predicts the power output of each of the multiple wind power generation devices 22 and predicts the output of the wind power plant 20 using this prediction result. When making this prediction, the output prediction device 10 also uses operation plan data. The operation plan data indicates the operation plan for each of the multiple wind power generation devices 22.

[0013] 2 is a diagram illustrating an example of the functional configuration of the output prediction device 10. In this example, the output prediction device 10 includes a first acquisition unit 110, a second acquisition unit 120, a power generation output prediction unit 130, and an output prediction unit 140, and can use a storage unit 150. The storage unit 150 stores various information used by the output prediction device 10. As an example, the power generation output prediction unit 130 stores a model and operation plan data used by the power generation output prediction unit 130. An example of the model is a learning model such as a neural network or deep learning. The storage unit 150 may be a part of the output prediction device 10, or may be provided outside the output prediction device 10.

[0014] The first acquisition unit 110 acquires weather forecast data, for example, data related to weather forecasts. The weather forecast data includes forecast results regarding wind at the wind power plant 20. In other words, the weather forecast data includes forecast results regarding wind in an area including the wind power generation device 22. More specifically, the weather forecast data includes trends in the distribution of wind direction and wind speed. It is preferable that the weather forecast data also include trends in outside air temperature. The period indicated by the weather forecast data may be, for example, one week, one month, or one year. The weather forecast data is generated, for example, by an external device, but may also be generated by the first acquisition unit 110.

[0015] The second acquisition unit 120 acquires operation plan data. The operation plan data indicates an operation plan for each of the multiple wind power generation devices 22. The operation plan indicates a period during which the wind power generation device 22 will be stopped. The cause of this stoppage may be, for example, a planned stoppage due to maintenance. The operation plan data may further include power generation suppression data, i.e., a period during which power generation suppression will be performed and an upper limit on power generation output during that period. The operation plan data is created, for example, by the manager of the wind power plant 20.

[0016] The power generation output prediction unit 130 generates power generation output prediction data using weather forecast data and operation plan data.

[0017] For example, the power generation output prediction unit 130 uses a machine learning model and weather forecast data to generate individual forecast data indicating the power generation output of each of the multiple wind turbine generators 22. The individual forecast data indicates the trend in the power generation output of the wind turbine generator 22 for each unit period. As an example, the individual forecast data indicates the power generation output of the wind turbine generator 22, for example, every 30 minutes. The target period for the trend indicated by the individual forecast data is, for example, one week, but may also be one month or one year. The power generation output prediction unit 130 then compiles these multiple individual forecast data into power generation output prediction data.

[0018] Furthermore, if operation plan data is available, the power generation output prediction unit 130 uses this operation plan data to correct the individual forecast data. The operation plan data indicates the suspension periods for each wind power generation device 22. For example, the power generation output prediction unit 130 corrects the individual forecast data so that the power generation output during the suspension periods in the operation plan data is zero. Furthermore, if the operation plan data includes power generation suppression data, the power generation output prediction unit 130 corrects the individual forecast data so that the power generation output during the power generation suppression periods in the operation plan data is equal to or less than the upper limit for the power generation suppression period. In this way, the accuracy of the forecast data generated by the power generation output prediction unit 130 is improved.

[0019] When generating the individual forecast data, the power generation output prediction unit 130 uses data from the weather forecast data that indicates the trends in wind direction and wind speed in an area that includes the location where the wind power generation device 22 is installed or the location where the wind power generation device 22 is planned to be installed. The size of this area is, for example, 5 km x 5 km, but is not limited to this value. The side of the square that represents this area may be any value between 1 km and 10 km. The power generation output prediction unit 130 generates the individual forecast data by, for example, inputting this data into a machine learning model. This machine learning model is updated by re-learning at predetermined times using actual power generation output data. The actual data used here includes power generation output data and weather data. The weather data includes trends in wind direction and wind speed distribution. Preferably, the weather data also includes trends in outside temperature and humidity.

[0020] It is preferable that the power generation output prediction unit 130 uses a machine learning model generated for each wind power generation device 22. As an example, if the wind power generation device 22 has already been installed, a learning model can be generated for each wind power generation device 22 using actual power generation data of the wind power generation device 22. In this way, the accuracy of the prediction data generated by the power generation output prediction unit 130 is improved.

[0021] The power generation output prediction unit 130 may further generate the power generation output prediction data using actual power generation data indicating the power generation results of each of the multiple wind turbine generators 22. For example, the power generation output prediction unit 130 generates the power generation output prediction data using a learning model in which the actual power generation data is used as at least one of the explanatory variables.

[0022] The output prediction unit 140 generates output prediction data using the power generation output prediction data generated by the power generation output prediction unit 130. The output prediction data shows the transition of the predicted value of the output of the wind power plant 20. This transition shows the predicted value of the output every 30 minutes, for example. For example, the power generation output prediction unit 130 generates the output prediction data by adding up the individual prediction data.

[0023] The output prediction unit 140 generates output data using the output prediction data. The output data may include the power generation output prediction data itself, or may include the power generation amount obtained by integrating the predicted power generation output value over a predetermined period, or may include both of these. The former data is expressed in units of kW, for example, and the latter data is expressed in units of kWh, for example.

[0024] 3 is a diagram showing an example of the hardware configuration of the output prediction device 10. In this example, the output prediction device 10 includes a bus 1010, a processor 1020, a memory 1030, a storage device 1040, an input / output interface 1050, and a network interface 1060.

[0025] The bus 1010 is a data transmission path for transmitting and receiving data among the processor 1020, memory 1030, storage device 1040, input / output interface 1050, and network interface 1060. However, the method of connecting the processor 1020 and the like to each other is not limited to bus connection.

[0026] The processor 1020 is implemented by a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), or the like.

[0027] The memory 1030 is a main storage device realized by a RAM (Random Access Memory) or the like.

[0028] The storage device 1040 is an auxiliary storage device realized by removable media such as a hard disk drive (HDD), a solid state drive (SSD), or a memory card, or a read-only memory (ROM). The storage device 1040 stores program modules that realize each function of the output prediction device 10 (e.g., the first acquisition unit 110, the second acquisition unit 120, the power generation output prediction unit 130, and the output prediction unit 140). The processor 1020 loads each of these program modules into the memory 1030 and executes them, thereby realizing each function corresponding to the program module. The storage device 1040 may also function as the memory unit 150.

[0029] The input / output interface 1050 is an interface for connecting the output prediction device 10 to various input / output devices.

[0030] The network interface 1060 is an interface for connecting the output prediction device 10 to a network. This network is, for example, a local area network (LAN) or a wide area network (WAN). The network interface 1060 may be connected to the network wirelessly or by wire. The output prediction device 10 may communicate with an external device via the network interface 1060.

[0031] 4 is a flowchart showing an example of the operation of the power output prediction device 10. The power output prediction device 10 performs this processing to predict the power output of the wind power plant 20, for example, before the wind power plant 20 is constructed. As another example, the power output prediction device 10 performs this processing when predicting the power output of a wind power plant 20 that is already in operation.

[0032] First, the first acquisition unit 110 acquires weather forecast data, and the second acquisition unit 120 acquires operation plan data (step S10).

[0033] Next, the power generation output prediction unit 130 generates power generation output prediction data using the weather forecast data. The power generation output prediction data includes individual forecast data for each wind turbine generator 22 (step S20). Next, the power generation output prediction unit 130 corrects the individual forecast data using the operation plan data, thereby correcting the power generation output prediction data (step S30).

[0034] The output prediction unit 140 then generates output data using the power generation output prediction data (step S40).

[0035] As described above, by using the output prediction device 10 of this embodiment, the prediction accuracy of the amount of power output from the wind power plant 20 is improved.

[0036] (Modification 1) Fig. 5 is a diagram showing a usage environment of the output prediction device 10 according to Modification 1. In the example shown in this figure, the wind power plant 20 has a power storage facility 24. The power storage facility 24 stores the power generated by the wind power generation device 22 as energy and outputs this stored energy as power. An example of the power storage facility 24 is a storage battery and its control device. However, the power storage facility 24 is not limited to this example.

[0037] The charge / discharge plan for the power storage facility 24 may be determined in advance. In this case, the output prediction device 10 generates output prediction data by further using charge / discharge plan data indicating the charge / discharge plan for the power storage facility 24. For example, the output prediction unit 140 corrects the output prediction data described in the embodiment by using the charge / discharge plan data. For example, during a period in which the power storage facility 24 is discharging, the output prediction unit 140 adds the amount of discharge to the output prediction data, and during a period in which the power storage facility 24 is charging, the output prediction unit 140 subtracts the amount of charge from the output prediction data.

[0038] Fig. 6 is a flowchart showing an example of the operation of the output prediction device 10 shown in Fig. 5. First, the first acquisition unit 110 acquires weather forecast data, and the second acquisition unit 120 acquires operation plan data and charge / discharge plan data (step S10).

[0039] Next, the power generation output prediction unit 130 generates power generation output prediction data using the weather forecast data (step S20), and corrects the power generation output prediction data using the operation plan data (step S30).

[0040] Then, the output prediction unit 140 generates output prediction data using the power generation output prediction data and the charge / discharge plan data (step S40).

[0041] This also improves the accuracy of predicting the amount of power output from the wind power plant 20. Furthermore, the amount of power output from the wind power plant 20 is leveled.

[0042] (Modification 2) Fig. 7 is a diagram showing a usage environment of the output prediction device 10 according to Modification 2. In the example shown in this figure, a power storage facility 24 is provided in parallel with each of the plurality of wind power generation devices 22. A charge / discharge plan for each of the plurality of power storage facilities 24 is set individually.

[0043] For example, consider a case where the first to fifth wind power generation devices 22 are lined up in this order from windward. In this case, as shown in FIG. 8 , the peaks of the power generation outputs of the first to fifth wind power generation devices 22 will occur at different times. It is preferable to create a charge / discharge plan for the power storage facility 24 so that charging occurs at the timing of these peaks. In this case, the charge / discharge plan data may be created by the output prediction unit 140. In this case, the charge / discharge plan data may be used as is, or may be modified by the manager of the wind power plant 20 before use.

[0044] In this modification, the charge / discharge plan data is set for each of the plurality of power storage facilities 24. In this way, the prediction accuracy of the amount of power output from the wind power plant 20 is improved, and the amount of power output from the wind power plant 20 is further leveled.

[0045] Although the embodiments of the present invention have been described above with reference to the drawings, these are merely examples of the present invention, and various other configurations can also be adopted.

[0046] In addition, although the flowcharts used in the above description show multiple steps (processes) in a sequential order, the order of steps executed in each embodiment is not limited to the order shown. In each embodiment, the order of the steps shown in the drawings can be changed as long as it does not cause any problems in terms of content. Furthermore, the above-described embodiments can be combined as long as the content is not contradictory.

[0047] REFERENCE SIGNS LIST 10 Output prediction device 20 Wind power plant 22 Wind power generation device 24 Power storage facility 110 First acquisition unit 120 Second acquisition unit 130 Power generation output prediction unit 140 Output prediction unit 150 Storage unit

Claims

1. An output prediction device comprising: a first acquisition unit that acquires weather forecast data including predicted results of wind distribution in a wind power plant having a plurality of wind power generation devices; a second acquisition unit that acquires operation plan data that indicates an operation plan for each of the plurality of wind power generation devices; a power output prediction unit that uses the weather forecast data and the operation plan data to generate power output prediction data that indicates predicted results of power output for each of the plurality of wind power generation devices; and an output prediction unit that uses the power output prediction data to generate output prediction data that indicates predicted results of power output from the wind power plant.

2. An output prediction device according to claim 1, wherein the power generation output prediction unit generates the power generation output prediction data using a machine learning model provided for each of the plurality of wind power generation devices.

3. An output prediction device according to claim 1 or 2, wherein the weather forecast data includes distribution of wind direction and wind speed.

4. An output prediction device according to claim 3, wherein the weather forecast data further includes an outside temperature.

5. An output prediction device according to any one of claims 1 to 4, wherein the wind power plant has a power storage facility, and the output prediction unit generates the output prediction data using charge / discharge plan data indicating a charge / discharge plan for the power storage facility.

6. An output prediction device according to claim 5, wherein the storage facility is provided in parallel with each of the plurality of wind power generation devices, and the charge / discharge plan data is set for each of the plurality of storage facilities.

7. An output prediction device according to any one of claims 1 to 6, wherein the power generation output prediction unit further generates the power generation output prediction data using power generation performance data indicating the power generation performance of each of the plurality of wind power generation devices.

8. An output prediction method in which a computer acquires weather forecast data including predicted results of wind distribution at a wind power plant having a plurality of wind power generation devices, acquires operation plan data indicating operation plans for each of the plurality of wind power generation devices, uses the weather forecast data and the operation plan data to generate power output prediction data indicating predicted results of power output for each of the plurality of wind power generation devices, and uses the power output prediction data to generate output prediction data indicating predicted results of the amount of power output from the wind power plant.

9. A program that causes a computer to have: a first acquisition unit that acquires weather forecast data including predicted results of wind distribution at a wind power plant having multiple wind power generation devices; a second acquisition unit that acquires operation plan data that indicates an operation plan for each of the multiple wind power generation devices; a power output prediction unit that uses the weather forecast data and the operation plan data to generate power output prediction data that indicates predicted results of power output for each of the multiple wind power generation devices; and an output prediction unit that uses the power output prediction data to generate output prediction data that indicates predicted results of the amount of power output from the wind power plant.

Images