Power generation estimation device, power generation estimation method, and program

Abstract

To provide a power generation amount estimating device capable of precisely estimating a power generation amount of a solar panel.SOLUTION: A power generation amount estimating device (2) includes an acquisition unit (21) that acquires weather prediction data, and an estimation unit (22) that estimates a power generation amount of a solar panel from predicted insolation amount data and predicted temperature data included in the weather prediction data by using a learned model that has learned, as teacher data, a set of a measured insolation amount, a measured temperature in the surrounding area of the solar panel, and a measured power generation amount of the solar panel.SELECTED DRAWING: Figure 3

Description

[Technical field]

[0001] The present invention relates to a technique for estimating the amount of power generated by photovoltaic power generation. [Background technology]

[0002] An invention disclosed in the following Patent Document 1 is a technology for predicting the amount of power generated by photovoltaic power generation. Patent Document 1 calculates the degree of clear weather corresponding to weather forecast data based on the relationship between the degree of clear weather and weather stored in a storage means, and calculates the amount of solar radiation at each time by multiplying the degree of clear weather obtained as a result of the calculation by the theoretical maximum amount of solar radiation at each time during the predicted time period. Then, the amount of solar radiation is multiplied by information related to the conversion efficiency, which indicates the relationship between the amount of solar radiation and power generation output, to calculate the power generation output. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2013-164286 A (Published on August 22, 2013) Summary of the Invention [Problem to be solved by the invention]

[0004] In Patent Document 1, the power generation output is calculated by using the degree of clear sky and the conversion efficiency. However, there is a problem in that when the accuracy of each of the degree of clear sky and the conversion efficiency is reduced, the accuracy of the calculated power generation output may be significantly reduced.

[0005] An object of one aspect of the present invention is to provide a power generation amount estimation device, a power generation amount estimation method, and a program that are capable of accurately estimating the amount of power generated by a solar panel. [Means for solving the problem]

[0006] In order to solve the above problems, a power generation estimation device according to one embodiment of the present invention includes an acquisition unit that acquires weather forecast data, and an estimation unit that estimates the power generation amount of the solar panel from the predicted data of the solar radiation and the predicted data of the temperature included in the weather forecast data using a trained model that has been trained using a set of actually measured solar radiation, the actually measured air temperature around the solar panel, and the actually measured power generation amount of the solar panel as teacher data.

[0007] In order to solve the above problems, a power generation estimation device according to one embodiment of the present invention includes an acquisition unit that acquires weather forecast data, and an estimation unit that estimates the power generation amount of the solar panel from the predicted data of the solar radiation contained in the weather forecast data using a trained model trained on a pair of actually measured solar radiation and the actually measured power generation amount of the solar panel as teacher data.

[0008] In addition, in order to solve the above problem, a power generation estimation method according to one embodiment of the present invention includes a step of acquiring weather forecast data, and a step of estimating the power generation amount of the solar panel from the predicted data of the solar radiation and the predicted data of the temperature included in the weather forecast data using a trained model trained on a set of actually measured solar radiation, actually measured air temperature around the solar panel, and actually measured power generation amount of the solar panel as teacher data.

[0009] In addition, in order to solve the above problem, a program according to one embodiment of the present invention causes a computer to execute a process of acquiring weather forecast data and a process of estimating the power generation amount of the solar panel from the predicted data of the solar radiation and the predicted data of the temperature contained in the weather forecast data, using a trained model trained on a set of actually measured solar radiation, actually measured air temperature around the solar panel, and actually measured power generation amount of the solar panel as teacher data. Effect of the Invention

[0010] According to one aspect of the present invention, it is possible to accurately estimate the amount of power generated by a solar panel. [Brief description of the drawings]

[0011] [Figure 1] 1 is a block diagram showing a functional configuration of a system including a learning device according to an embodiment of the present invention. [Diagram 2] 1 is a flowchart for explaining a processing procedure of a system including a learning device according to an embodiment of the present invention. [Diagram 3] 1 is a block diagram showing a functional configuration of a power generation amount estimation device according to an embodiment of the present invention. [Figure 4] 4 is a flowchart for explaining a processing procedure of the power generation amount estimation device according to one embodiment of the present invention. [Diagram 5] 11 is a flowchart for explaining details of a process for converting a horizontal plane global solar radiation into another type of solar radiation. [Figure 6] 10 is a flowchart for explaining details of a process for converting an air temperature into an air temperature on the rear surface of a solar panel. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0012] <Example of system configuration including learning device 1> 1 is a functional block diagram showing a schematic configuration of a system including a learning device according to an embodiment of the present invention. The system includes a learning device 1, a solar radiation meter 12, a temperature meter 13, an RTC (Real Time Clock) 14, and a power generation meter 15.

[0013] The solar radiation measuring device 12 is installed around the solar panels included in the PV (Photovoltaic Power Generation) system, measures the amount of solar radiation, and outputs the measurement result to the learning device 1. The solar radiation measuring device 12 is one of four types: a measuring device for measuring the amount of global solar radiation on a horizontal surface, a measuring device for measuring the amount of global solar radiation on an inclined surface, a measuring device for measuring the amount of direct solar radiation on a horizontal surface, and a measuring device for measuring the amount of direct solar radiation on an inclined surface, and is installed around the solar panels according to the characteristics of each measuring device.

[0014] The temperature measuring device 13 is installed on the solar panel, measures the temperature on the back side of the solar panel, and outputs the measurement result to the learning device 1. The temperature measuring device 13 may also be installed near the solar panel and measure the temperature around the solar panel.

[0015] The RTC 14 keeps track of time and outputs at least the current date and time as time information to the learning device 1. The time information includes the month, date, hour, minute, and second, but it is up to the user to decide which of these pieces of information to use.

[0016] The power generation amount measuring device 15 is installed on the solar panel, measures the current power generation amount of the solar panel, and outputs the measurement result to the learning device 1.

[0017] The learning device 1 trains a learning model using sets of the amount of solar radiation actually measured by a solar radiation meter 12, the temperature on the back side of the solar panel actually measured by a temperature meter 13, time information at the time of actual measurement output from an RTC 14, and the amount of power generated by the solar panel actually measured by a power generation meter 15 as teacher data, and outputs the trained model 11. This learning model is trained to estimate the amount of power generated by the solar panel from the amount of solar radiation, the temperature on the back side of the solar panel, and the time information.

[0018] The learning device 1 may train the learning model using as teacher data a set of the amount of solar radiation actually measured by the solar radiation measuring device 12, the air temperature on the back side of the solar panel actually measured by the air temperature measuring device 13, and the amount of power generated by the solar panel actually measured by the power generation measuring device 15. The learning device 1 may train the learning model using as teacher data a set of the amount of solar radiation actually measured by the solar radiation measuring device 12, time information at the time of actual measurement output from the RTC 14, and the amount of power generated by the solar panel actually measured by the power generation measuring device 15.

[0019] The learning device 1 trains the learning model with, for example, one year's worth of data for six hours per day (time periods when solar power generation is possible). The amount of power generated by a solar panel has the strongest correlation with the amount of solar radiation, but it also has correlation with temperature, season, time of day, etc., so by training the learning model with data including these, it is possible to improve the prediction accuracy of the trained model.

[0020] <Processing Procedure of the System Including the Learning Device 1> 2 is a flowchart for explaining the processing procedure of a system including the learning device 1. First, the solar radiation measuring device 12 measures the amount of solar radiation and outputs it to the learning device 1 (S11). As described above, depending on the type of solar radiation measuring device 12, any one of the following is output: global solar radiation on a horizontal surface, global solar radiation on an inclined surface, direct solar radiation on a horizontal surface, and direct solar radiation on an inclined surface.

[0021] Next, the temperature measuring device 13 measures the temperature on the back side of the solar panel and outputs it to the learning device 1 (S12). In addition, the RTC 14 acquires current time information and outputs it to the learning device 1 (S13). In addition, the power generation measuring device 15 measures the amount of power generated by the solar panel and outputs it to the learning device 1 (S14).

[0022] Next, the learning device 1 trains the learning model using as teacher data a set of the amount of solar radiation actually measured by the solar radiation measuring instrument 12, the temperature on the back side of the solar panel actually measured by the temperature measuring instrument 13, the time information at the time of actual measurement output from the RTC 14, and the amount of power generated by the solar panel actually measured by the power generation measuring instrument 15 (S15).

[0023] Then, the learning device 1 determines whether or not to end the process (S16). If the process is not to be ended (S16, No), the learning device 1 returns to step S11 and repeats the subsequent processes. If the process is to be ended (S16, Yes), the learning device 1 outputs the trained model 11 and ends the process.

[0024] In addition, one year's worth of actual data (measured solar radiation, measured temperature on the back side of the solar panel, time information at the time of actual measurement, measured amount of power generated by the solar panel) may be acquired and stored, and the learning device 1 may learn from the one year's worth of actual data to create a trained model.

[0025] <Configuration example of power generation amount estimation device 2> 3 is a functional block diagram showing a schematic configuration of a power generation amount estimation device 2 according to one embodiment of the present invention. The power generation amount estimation device 2 includes an acquisition unit 21 and an estimation unit 22. The estimation unit 22 includes a solar radiation conversion unit 221, an air temperature conversion unit 222, a control unit 223, and a trained model 11 created by the learning device 1.

[0026] The acquisition unit 21 is configured with a man-machine interface such as a keyboard, a mouse, a touch panel, or a network adapter, and acquires weather forecast data from the outside and outputs it to the estimation unit 22. The weather forecast data includes information such as forecast data of horizontal plane global solar radiation, forecast data of temperature, date and time (date, time zone), latitude, longitude, etc.

[0027] For example, when the acquisition unit 21 is configured with a network adapter or the like, it receives weather forecast data from a server or the like owned by a weather forecast data provider such as the Japan Meteorological Agency or a private weather forecasting company via a wide area communication network such as the Internet. In addition, when the acquisition unit 21 is configured with a man-machine interface or the like, the user may input the weather forecast data.

[0028] The solar radiation conversion unit 221 converts the predicted data of the horizontal plane global solar radiation included in the weather forecast data into the solar radiation corresponding to the type of solar radiation when the learning model was trained. For example, if the solar radiation meter 12 shown in FIG. 1 is a measuring instrument that measures the inclined plane global solar radiation, the solar radiation conversion unit 221 converts the predicted data of the horizontal plane global solar radiation included in the weather forecast data into the predicted data of the inclined plane global solar radiation. This inclined plane global solar radiation is the solar radiation that takes into account the installation angle and orientation of the solar panel, and when creating a learning model that estimates the power generation amount of the solar panel, the prediction accuracy can be improved by using this inclined plane global solar radiation, but the above-mentioned horizontal plane global solar radiation, horizontal plane direct solar radiation, or inclined plane direct solar radiation may also be used.

[0029] The solar radiation conversion unit 221 converts the predicted data of the horizontal surface global solar radiation into predicted data of the inclined surface global solar radiation, for example, by using a direct / diffuse separation model such as Erbs or an inclined surface solar radiation model. In this conversion process, predicted data of the horizontal surface direct solar radiation and predicted data of the inclined surface direct solar radiation are also calculated.

[0030] Furthermore, the solar radiation conversion unit 221 calculates the theoretical maximum value of the global solar radiation on a horizontal plane from information such as latitude, longitude, date and time, etc. Then, the solar radiation conversion unit 221 can convert the calculated theoretical maximum value of the global solar radiation on a horizontal plane into the theoretical maximum value of the global solar radiation on an inclined plane, the theoretical maximum value of the direct solar radiation on a horizontal plane, and the theoretical maximum value of the direct solar radiation on an inclined plane, for example, using a direct / diffuse separation model such as Erbs or an inclined solar radiation model.

[0031] The temperature conversion unit 222 converts the predicted temperature data T bf Convert to the temperature on the back side of the solar panel. Temperature on the back side of the solar panel T af can be calculated using the following equations (Equation 1) and (Equation 2).

[0032] T af =T bf +γΔT (Eq. 1) T af :Temperature after correction of weather forecast data (℃) T bf :Temperature before correction of weather forecast data (℃) γ: correction coefficient ΔT: Temperature correction (℃)

[0033]

number

[0034] The control unit 223 holds information such as the type of solar radiation when the learning model is trained, and outputs the type of solar radiation to the solar radiation conversion unit 221 when the solar radiation conversion unit 221 converts the predicted data of the horizontal plane global solar radiation into another type of solar radiation. The control unit 223 also outputs the predicted data of the inclined plane global solar radiation calculated by the solar radiation conversion unit 221 and the predicted data of the horizontal plane global solar radiation to the air temperature conversion unit 222. The control unit 223 also obtains time information of the prediction target from the weather forecast data when estimating the power generation amount of the solar panel, and outputs it to the trained model 11.

[0035] The trained model 11 receives as input the predicted data of the amount of solar radiation output from the solar radiation conversion unit 221, the predicted data of the air temperature on the back side of the solar panel output from the air temperature conversion unit 222, and the time information output from the control unit 223, and outputs an estimate of the amount of power generated by the solar panel. This estimated value of the amount of power generated is displayed, for example, on a display device.

[0036] <Processing Procedure of Power Generation Amount Estimation Device 2> 4 is a flowchart for explaining the processing procedure of the power generation amount estimation device 2. First, the solar radiation conversion unit 221 extracts the horizontal plane global solar radiation at the time of prediction from the weather forecast data acquired by the acquisition unit 21, converts it to another type of solar radiation, and outputs it to the trained model 11 (S21). Details of this process will be described later.

[0037] Next, the temperature conversion unit 222 extracts forecast data of the temperature at the time of prediction from the weather forecast data acquired by the acquisition unit 21, converts it into forecast data of the temperature on the back surface of the solar panel, and outputs it to the trained model 11 (S22). Details of this process will be described later.

[0038] Next, the control unit 223 acquires time information of the prediction target from the weather forecast data and outputs it to the trained model 11 (S23). Then, the trained model 11 receives the predicted data of the amount of solar radiation converted by the solar radiation conversion unit 221, the predicted data of the temperature on the back side of the solar panel converted by the temperature conversion unit 222, and the time information of the prediction target output from the control unit 223 (S24). The trained model 11 outputs the estimated result of the amount of power generated by the solar panel to the outside (S25) and ends the process.

[0039] 5 is a flowchart for explaining the details of the process of converting the horizontal global solar radiation into other types of solar radiation (the process of step S21 in FIG. 4). First, the solar radiation conversion unit 221 acquires weather forecast data for the prediction target time period (S31). Next, the solar radiation conversion unit 221 calculates the theoretical maximum value of the horizontal global solar radiation at each time in the prediction target time period (S32). This theoretical maximum value of the horizontal global solar radiation is the horizontal global solar radiation on a clear day, and can be calculated from information such as latitude, longitude, date and time.

[0040] Next, the solar radiation conversion unit 221 determines whether the type of the actually measured solar radiation (the type of solar radiation when the learning model was trained) is the horizontal surface global solar radiation (S33). If the actually measured solar radiation is not the horizontal surface global solar radiation (S33, No), the process proceeds to step S37.

[0041] If the measured solar radiation is the horizontal global solar radiation (S33, Yes), the solar radiation conversion unit 221 judges whether the theoretical maximum value of the horizontal global solar radiation is equal to or greater than the predicted data of the horizontal global solar radiation (S34). If the theoretical maximum value of the horizontal global solar radiation is smaller than the predicted data of the horizontal global solar radiation (S34, No), the predicted data of the horizontal global solar radiation is replaced with the theoretical maximum value of the horizontal global solar radiation (S35), and the process proceeds to step S36. This process is for preventing the predicted data of the horizontal global solar radiation from being larger than the theoretical maximum value of the horizontal global solar radiation. If the theoretical maximum value of the horizontal global solar radiation is equal to or greater than the predicted data of the horizontal global solar radiation (S34, Yes), the process proceeds to step S36.

[0042] In step S36, the solar radiation conversion unit 221 outputs the predicted data of the horizontal plane global solar radiation to the trained model 11, and ends the process. In addition, in step S37, the solar radiation conversion unit 221 converts the theoretical maximum value of the horizontal plane global solar radiation into the theoretical maximum value of the inclined plane global solar radiation. At this time, the theoretical maximum value of the horizontal plane direct solar radiation and the theoretical maximum value of the inclined plane direct solar radiation are also calculated.

[0043] Then, the solar radiation amount conversion unit 221 converts the predicted data of the horizontal surface global solar radiation into predicted data of the inclined surface global solar radiation (S38). At this time, the predicted data of the horizontal surface direct solar radiation and the predicted data of the inclined surface direct solar radiation are also calculated.

[0044] Next, the solar radiation conversion unit 221 judges whether the type of the actually measured solar radiation is the inclined surface global solar radiation (S39). If the actually measured solar radiation is not the inclined surface global solar radiation (S39, No), the process proceeds to step S43.

[0045] If the measured solar radiation is the inclined surface global solar radiation (S39, Yes), the solar radiation conversion unit 221 judges whether the theoretical maximum value of the inclined surface global solar radiation is equal to or greater than the predicted data of the inclined surface global solar radiation (S40). If the theoretical maximum value of the inclined surface global solar radiation is smaller than the predicted data of the inclined surface global solar radiation (S40, No), the predicted data of the inclined surface global solar radiation is replaced with the theoretical maximum value of the inclined surface global solar radiation (S41), and the process proceeds to step S42. This process is for preventing the predicted data of the inclined surface global solar radiation from being larger than the theoretical maximum value of the inclined surface global solar radiation. If the theoretical maximum value of the inclined surface global solar radiation is equal to or greater than the predicted data of the inclined surface global solar radiation (S40, Yes), the process proceeds to step S42.

[0046] In step S42, the solar radiation conversion unit 221 outputs the predicted data of the inclined surface global solar radiation to the trained model 11, and ends the process. In addition, in step S43, the solar radiation conversion unit 221 determines whether the type of the actually measured solar radiation is the horizontal surface direct solar radiation. If the actually measured solar radiation is not the horizontal surface direct solar radiation (S43, No), the process proceeds to step S47.

[0047] If the measured solar radiation is the horizontal surface direct solar radiation (S43, Yes), the solar radiation conversion unit 221 judges whether the theoretical maximum value of the horizontal surface direct solar radiation is equal to or greater than the predicted data of the horizontal surface direct solar radiation (S44). If the theoretical maximum value of the horizontal surface direct solar radiation is smaller than the predicted data of the horizontal surface direct solar radiation (S44, No), the predicted data of the horizontal surface direct solar radiation is replaced with the theoretical maximum value of the horizontal surface direct solar radiation (S45), and the process proceeds to step S46. This process is for preventing the predicted data of the horizontal surface direct solar radiation from being larger than the theoretical maximum value of the horizontal surface direct solar radiation. If the theoretical maximum value of the horizontal surface direct solar radiation is equal to or greater than the predicted data of the horizontal surface direct solar radiation (S44, Yes), the process proceeds to step S46.

[0048] In step S46, the solar radiation conversion unit 221 outputs the predicted data of the amount of direct solar radiation on a horizontal surface to the trained model 11, and ends the process. In addition, in step S47, since the type of the actually measured amount of solar radiation is the amount of direct solar radiation on an inclined surface, the solar radiation conversion unit 221 determines whether the theoretical maximum value of the amount of direct solar radiation on an inclined surface is equal to or greater than the predicted data of the amount of direct solar radiation on an inclined surface.

[0049] If the theoretical maximum value of the inclined surface direct solar radiation is smaller than the predicted data of the inclined surface direct solar radiation (S47, No), the predicted data of the inclined surface direct solar radiation is replaced with the theoretical maximum value of the inclined surface direct solar radiation (S48), and the process proceeds to step S49. This process is for preventing the predicted data of the inclined surface direct solar radiation from becoming larger than the theoretical maximum value of the inclined surface direct solar radiation. Also, if the theoretical maximum value of the inclined surface direct solar radiation is equal to or larger than the predicted data of the inclined surface direct solar radiation (S47, Yes), the process proceeds to step S49.

[0050] In step S49, the solar radiation conversion unit 221 outputs the predicted data of the amount of direct solar radiation on the inclined surface to the trained model 11, and ends the processing.

[0051] 6 is a flowchart for explaining the details of the process of converting the air temperature into the air temperature on the back side of the solar panel (the process of step S22 in FIG. 4). First, the air temperature conversion unit 222 acquires weather forecast data for the prediction time period from the acquisition unit 21 (S51).

[0052] Next, the temperature conversion unit 222 acquires the inclined surface global solar radiation at each time of the prediction target time zone calculated by the solar radiation conversion unit 221 from the control unit 223 (S52). Then, the temperature conversion unit 222 converts the temperature of the weather forecast data at each time of the prediction target time zone into the temperature of the back surface of the solar panel using the above (Equation 1) and (Equation 2) (S53), and ends the process.

[0053] As machine learning, NN (Neural Network), RNN (Recurrent NN), etc. can be applied, but it is also possible to apply technologies that can predict the amount of power generated by solar panels from the amount of solar radiation and temperature in weather forecast data, such as support vector machines, decision tree analysis, and multiple regression analysis.

[0054] As described above, according to the power generation amount estimation device 2 of this embodiment, the estimation unit 22 estimates the power generation amount of the solar panel from the predicted data of the amount of solar radiation and the predicted data of the temperature included in the weather forecast data, using a trained model trained with a set of the actually measured amount of solar radiation, the actually measured temperature on the back side of the solar panel, and the actually measured amount of power generation of the solar panel as teacher data. Therefore, it is not necessary to calculate the degree of clear weather, conversion efficiency, etc., and it is possible to accurately estimate the power generation amount of the solar panel.

[0055] In addition, the trained model can be a learning model trained using a set of the measured amount of solar radiation, the measured temperature on the back side of the solar panel, the time information at the time of the actual measurement, and the measured amount of power generated by the solar panel as teacher data. Therefore, by training the learning model including the time information at the time of the actual measurement, which has a high correlation with the amount of power generated by the solar panel, it is possible to estimate the amount of power generated by the solar panel with even greater accuracy.

[0056] Furthermore, even if the type of solar radiation in the actual measurement data (teacher data) is different from the type of solar radiation in the weather forecast data, the solar radiation conversion unit 221 can match the type of solar radiation in the weather forecast data to the type of solar radiation in the actual measurement data (teacher data), making it possible to estimate the power generation amount of the solar panel with even greater accuracy.

[0057] For example, if the type of solar radiation in the actual measurement data (teacher data) is inclined surface global solar radiation and the type of solar radiation in the weather forecast data is horizontal surface global solar radiation, the solar radiation conversion unit 221 can convert the horizontal surface global solar radiation in the weather forecast data into the inclined surface global solar radiation in the actual measurement data (teacher data), making it possible to estimate the power generation amount of the solar panel with even greater accuracy.

[0058] In addition, when the type of solar radiation in the actual measurement data (teacher data) is horizontal plane direct solar radiation and the type of solar radiation in the weather forecast data is horizontal plane global solar radiation, the solar radiation conversion unit 221 converts the horizontal plane global solar radiation in the weather forecast data into the horizontal plane direct solar radiation of the actual measurement data (teacher data), making it possible to estimate the power generation amount of the solar panel with even greater accuracy.

[0059] In addition, when the type of solar radiation in the actual measurement data (teacher data) is direct solar radiation on an inclined surface and the type of solar radiation in the weather forecast data is global solar radiation on a horizontal surface, the solar radiation conversion unit 221 converts the global solar radiation on a horizontal surface in the weather forecast data into the direct solar radiation on an inclined surface in the actual measurement data (teacher data), thereby making it possible to estimate the power generation amount of the solar panel with even greater accuracy.

[0060] Of course, it is also possible to handle cases where the type of solar radiation in the actual measurement data (teacher data) and the type of solar radiation in the weather forecast data match, such as when the type of solar radiation in the actual measurement data (teacher data) is global solar radiation on a horizontal surface and the type of solar radiation in the weather forecast data is global solar radiation on a horizontal surface.

[0061] [Software implementation example] The control blocks of the power generation amount estimation device 2 (particularly the solar radiation conversion unit 221, the air temperature conversion unit 222, and the control unit 223) may be realized by a logic circuit (hardware) formed in an integrated circuit (IC chip) or the like, or may be realized by software.

[0062] In the latter case, the power generation amount estimation device 2 includes a computer that executes instructions of a program, which is software that realizes each function. The computer includes, for example, one or more processors, and a computer-readable recording medium that stores the program. The object of the present invention is achieved by the processor reading the program from the recording medium and executing it in the computer. The processor may be, for example, a CPU (Central Processing Unit). The recording medium may be a "non-transient tangible medium", such as a ROM (Read Only Memory), a tape, a disk, a card, a semiconductor memory, a programmable logic circuit, or the like. The device may further include a RAM (Random Access Memory) that expands the program. The program may be supplied to the computer via any transmission medium (such as a communication network or a broadcast wave) that can transmit the program. Note that one aspect of the present invention may also be realized in the form of a data signal embedded in a carrier wave, in which the program is embodied by electronic transmission.

[0063] 〔summary〕 The power generation estimation device according to aspect 1 of the present invention includes an acquisition unit that acquires weather forecast data, and an estimation unit that estimates the power generation capacity of the solar panel from the predicted data on the amount of solar radiation and the predicted data on air temperature contained in the weather forecast data, using a trained model trained on a set of actually measured solar radiation, actually measured air temperature around the solar panel, and actually measured power generation capacity of the solar panel as teacher data.

[0064] In the power generation estimation device according to aspect 2 of the present invention, in the above-mentioned aspect 1, the trained model is a learning model trained using a set of actually measured solar radiation, actually measured air temperature around the solar panel, time information at the time of the actual measurement, and actually measured power generation of the solar panel as teacher data.

[0065] A power generation estimation device according to a third aspect of the present invention is configured as follows: in the first or second aspect described above, the actually measured solar radiation is actual measured data of horizontal plane global solar radiation; the estimation unit includes a temperature conversion unit that converts air temperature prediction data included in the weather forecast data into air temperature prediction data around the solar panel; and the predicted data of horizontal plane global solar radiation included in the weather forecast data and the predicted data of the air temperature around the solar panel are input into the trained model.

[0066] The power generation estimation device according to aspect 4 of the present invention is, in the above-mentioned aspect 1 or 2, characterized in that the actually measured solar radiation is actual measurement data of inclined surface global solar radiation corresponding to the solar panel, and the estimation unit includes a solar radiation conversion unit that converts predicted data of horizontal surface global solar radiation included in the weather forecast data into predicted data of inclined surface global solar radiation corresponding to the solar panel, and a temperature conversion unit that converts predicted data of air temperature included in the weather forecast data into predicted data of air temperature around the solar panel, and the predicted data of inclined surface global solar radiation and the predicted data of air temperature around the solar panel are input into the trained model.

[0067] The power generation estimation device of aspect 5 of the present invention is, in the above-mentioned aspect 1 or 2, characterized in that the actually measured solar radiation is actual measured data of horizontal plane direct solar radiation, and the estimation unit includes a solar radiation conversion unit that converts predicted data of horizontal plane global solar radiation included in the weather forecast data into predicted data of the horizontal plane direct solar radiation, and a temperature conversion unit that converts predicted data of air temperature included in the weather forecast data into predicted data of air temperature around the solar panel, and the predicted data of horizontal plane direct solar radiation and the predicted data of the air temperature around the solar panel are input into the trained model.

[0068] The power generation estimation device according to aspect 6 of the present invention is, in the above-mentioned aspect 1 or 2, characterized in that the actually measured solar radiation is actual measurement data of inclined surface direct solar radiation corresponding to the solar panel, and the estimation unit includes a solar radiation conversion unit that converts predicted data of horizontal surface global solar radiation included in the weather forecast data into predicted data of inclined surface direct solar radiation corresponding to the solar panel, and a temperature conversion unit that converts predicted data of air temperature included in the weather forecast data into predicted data of air temperature around the solar panel, and the predicted data of inclined surface direct solar radiation and the predicted data of air temperature around the solar panel are input into the trained model.

[0069] In the power generation amount estimation device according to a seventh aspect of the present invention, in any one of the third to sixth aspects, the temperature conversion unit converts the predicted data T bf The predicted temperature data around the solar panel T af Convert to.

[0070] T af =T bf +γΔT (Eq. 1) T af :Temperature after correction of weather forecast data (℃) T bf :Temperature before correction of weather forecast data (℃) γ: correction coefficient ΔT: Temperature correction (℃)

[0071]

number

[0072] The power generation estimation device according to aspect 8 of the present invention includes an acquisition unit that acquires weather forecast data, and an estimation unit that estimates the power generation amount of the solar panel from the predicted data of the solar radiation included in the weather forecast data using a trained model trained on a pair of actually measured solar radiation and the actually measured power generation amount of the solar panel as teacher data.

[0073] A power generation estimation device according to a ninth aspect of the present invention is, in the above-mentioned eighth aspect, a learning model which is trained using a set of actually measured solar radiation, time information at the time of the actual measurement, and the actually measured power generation amount of the solar panel as training data.

[0074] A power generation estimation method according to aspect 10 of the present invention includes the steps of acquiring weather forecast data, and estimating the power generation of the solar panel from the predicted data on the amount of solar radiation and the predicted data on temperature contained in the weather forecast data using a trained model trained on a set of actually measured solar radiation, actually measured air temperature around the solar panel, and actually measured power generation of the solar panel as teacher data.

[0075] A program according to aspect 11 of the present invention causes a computer to execute a process of acquiring weather forecast data, and a process of estimating the amount of power generation of a solar panel from the predicted data of solar radiation and predicted data of temperature contained in the weather forecast data, using a trained model trained on a set of actually measured solar radiation, actually measured air temperature around the solar panel, and actually measured amount of power generation of the solar panel as teacher data.

[0076] [Additional Notes] The present invention is not limited to the above-described embodiments, and various modifications are possible within the scope of the claims. Embodiments obtained by appropriately combining the technical means disclosed in different embodiments are also included in the technical scope of the present invention. [Explanation of symbols]

[0077] 1 Learning device 2. Power generation estimation device 11 Pre-trained models 12 Solar radiation meter 13 Temperature Measuring Device 14 RTC 15 Power generation measuring device 21 Acquisition Department 22 Estimation part 221 Solar radiation converter 222 Temperature converter 223 Control Unit

Claims

1. An acquisition unit that acquires weather forecast data; an estimation unit that estimates the amount of power generated by the solar panel from predicted data of the amount of solar radiation and predicted data of air temperature included in the weather forecast data, using a trained model that has been trained using a set of actually measured solar radiation, actually measured air temperature around the solar panel, and actually measured amount of power generated by the solar panel as teacher data; A power generation estimation device comprising:

2. The trained model is a learning model trained using a set of the measured amount of solar radiation, the measured temperature around the solar panel, the time information at the time of the actual measurement, and the measured amount of power generation of the solar panel as teacher data. The power generation estimation device according to claim 1 .

3. The measured solar radiation amount is actual measurement data of horizontal plane global solar radiation, The estimation unit is a temperature conversion unit that converts the temperature prediction data included in the weather prediction data into prediction data of the temperature around the solar panel; The forecast data of horizontal plane global solar radiation included in the weather forecast data and the forecast data of the air temperature around the solar panel are input into the trained model. The power generation estimation device according to claim 1 .

4. The actually measured amount of solar radiation is actual measurement data of the inclined surface global solar radiation corresponding to the solar panel, The estimation unit is a solar radiation conversion unit that converts prediction data of horizontal surface global solar radiation included in the weather forecast data into prediction data of inclined surface global solar radiation corresponding to the solar panel; a temperature conversion unit that converts the temperature prediction data included in the weather prediction data into prediction data of the temperature around the solar panel, The predicted data of the inclined surface global solar radiation and the predicted data of the air temperature around the solar panel are input into the trained model. The power generation estimation device according to claim 1 .

5. The measured solar radiation amount is actual measurement data of direct solar radiation on a horizontal surface, The estimation unit is a solar radiation conversion unit that converts prediction data of horizontal plane global solar radiation included in the weather forecast data into prediction data of the horizontal plane direct solar radiation; a temperature conversion unit that converts the temperature prediction data included in the weather prediction data into prediction data of the temperature around the solar panel, The predicted data of the horizontal surface direct solar radiation and the predicted data of the air temperature around the solar panel are input into the trained model. The power generation estimation device according to claim 1 .

6. The measured amount of solar radiation is measured data of an amount of solar radiation directly incident on an inclined surface corresponding to the solar panel, The estimation unit is a solar radiation conversion unit that converts forecast data of the horizontal surface global solar radiation amount included in the weather forecast data into forecast data of the inclined surface direct solar radiation amount corresponding to the solar panel; a temperature conversion unit that converts the temperature prediction data included in the weather prediction data into prediction data of the temperature around the solar panel, The predicted data of the amount of direct solar radiation on the inclined surface and the predicted data of the air temperature around the solar panel are input into the trained model. The power generation estimation device according to claim 1 .

7. The temperature conversion unit converts the predicted data T of the temperature included in the weather forecast data into the predicted data T by using the following formulas (Formula 1) and (Formula 2). ｂｆ The predicted data T of the temperature around the solar panel ａｆ Convert to The power generation amount estimation device according to any one of claims 3 to 6. T ａｆ = T ｂｆ +γΔT...(Formula 1) T ａｆ :Temperature after correcting weather forecast data (℃) T ｂｆ:Temperature before correction of weather forecast data (℃) γ: correction coefficient ΔT: Temperature correction (℃) [0010] I Ｔ，Ｇ :Slope total solar radiation (W / m 2 ) I ｄ : Horizontal solar radiation (W / m 2 ) ρ: Ground reflectance δ: Stefan-Boltzmann constant (=5.670374419×10 -8 [W / (mK 2 ) 2 ]) W ｓ :Wind speed of weather forecast data (m / s) h: Solar altitude (°)

8. An acquisition unit that acquires weather forecast data; an estimation unit that estimates the amount of power generated by the solar panel from predicted data of the amount of solar radiation included in the weather forecast data using a trained model that has been trained using a pair of actually measured amounts of solar radiation and actually measured amounts of power generated by the solar panel as teacher data; A power generation estimation device comprising:

9. The trained model is a learning model trained using a set of the measured amount of solar radiation, the time information at the time of the actual measurement, and the measured amount of power generation of the solar panel as teacher data. The power generation estimation device according to claim 8.

10. obtaining weather forecast data; a step of estimating the amount of power generated by the solar panel from predicted data of the amount of solar radiation and predicted data of temperature included in the weather forecast data, using a trained model trained on a set of actually measured solar radiation, actually measured air temperature around the solar panel, and actually measured amount of power generated by the solar panel as teacher data; A method for estimating power generation, comprising:

11. On the computer, Obtaining weather forecast data; A process of estimating the amount of power generated by the solar panel from predicted data of the amount of solar radiation and predicted data of temperature included in the weather forecast data, using a trained model trained on a set of actually measured solar radiation, actually measured air temperature around the solar panel, and actually measured amount of power generated by the solar panel as teacher data; A program that executes the following.

Images