Photovoltaic power generation performance evaluation device, photovoltaic power generation performance evaluation method, and program

Abstract

To provide a photovoltaic power generation performance evaluation device capable of evaluating the power generation performance of a photovoltaic power plant even if no past power generation data exists.SOLUTION: A photovoltaic power generation performance evaluation device includes: a storage unit that stores a learning model created by machine learning in which weather conditions are used as explanatory variables and a result obtained by simulating the power generation performance of a photovoltaic power plant on the basis of the weather conditions is used as a target variable; a data acquisition unit that acquires weather data measured at the photovoltaic power plant and power generation data measured by a plurality of types of electric power equipment installed at the photovoltaic power plant; and a data processing unit that evaluates the power generation performance of the photovoltaic power plant on the basis of a comparison result between a first power generation performance calculated by inputting the weather data into the learning model and a second power generation performance derived from the power generation data.SELECTED DRAWING: Figure 2

Description

[Technical field]

[0001] An embodiment of the present invention relates to a photovoltaic power generation performance evaluation device, a photovoltaic power generation performance evaluation method, and a program. [Background technology]

[0002] Photovoltaic power generation is easily affected by the environment. Therefore, it is important to check whether the expected power generation performance is being achieved. In general, power generation performance is evaluated using meteorological data such as current solar radiation and temperature, and actual power generation data. In addition, in order to evaluate whether the power generation performance is deteriorating, an evaluation method may be adopted in which the power generation amount calculated from past meteorological data is compared with the current power generation amount.

[0003] However, immediately after the start of operation of a solar power plant, there is no past power generation data. Therefore, it is difficult to detect a decrease in power generation performance using the above evaluation method. In addition, the above evaluation method is based on the premise that the past power generation data is from a period when optimal power generation was performed, so it is difficult to determine whether the amount of power generation that was initially expected is being obtained. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Patent No. 6093465 Summary of the Invention [Problem to be solved by the invention]

[0005] Therefore, the embodiments of the present invention propose a photovoltaic power generation performance evaluation device, a photovoltaic power generation performance evaluation method, and a program that are capable of evaluating the power generation performance of a photovoltaic power plant even if no past power generation data exists. [Means for solving the problem]

[0006] A photovoltaic power generation performance evaluation device according to one embodiment includes a memory unit that stores a learning model created by machine learning using weather conditions as explanatory variables and the result of simulating the power generation performance of the photovoltaic power plant based on the weather conditions as a target variable, a data acquisition unit that acquires weather data measured at the photovoltaic power plant and power generation data measured by multiple types of power equipment installed at the photovoltaic power plant, and a data processing unit that evaluates the power generation performance of the photovoltaic power plant based on a comparison result between a first power generation performance calculated by inputting the weather data into the learning model and a second power generation performance derived from the power generation data. Effect of the Invention

[0007] According to this embodiment, even if there is no past power generation data, it is possible to evaluate the power generation performance of photovoltaic power generation. [Brief description of the drawings]

[0008] [Figure 1] 1 is a block diagram showing a configuration of a solar power plant according to a first embodiment. [Diagram 2] FIG. 1 is a block diagram showing an example of the configuration of a photovoltaic power generation performance evaluation device. [Diagram 3] 3 is a flowchart showing the steps of a method for evaluating the power generation performance of a photovoltaic power plant according to the first embodiment. [Figure 4] FIG. 11 is a diagram for explaining another example of a method for determining a malfunction of a solar power plant. [Diagram 5] FIG. 11 is a diagram showing an example of a learning model database stored in a storage unit according to a first modified example. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0009] DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, embodiments of the present invention will be described with reference to the drawings. The present invention is not limited to the following embodiments.

[0010] (First embodiment) Fig. 1 is a block diagram showing the configuration of a photovoltaic power plant according to the first embodiment. The photovoltaic power plant 100 shown in Fig. 1 includes a PV panel 110, a junction box 120, an intermediate substation 130, a high-voltage substation 140, a power measuring instrument 150, an environmental measuring instrument 160, and a photovoltaic power generation performance evaluation device 200.

[0011] The PV panel 110 converts incident sunlight into electricity.

[0012] The junction box 120 aggregates the output power of the multiple PV panels 110. Although three PV panels 110 are connected to one junction box 120 in Fig. 1, the number of connected PV panels 110 is not particularly limited.

[0013] A plurality of PCS (Power Conditioning System) packages 131 are installed in the intermediate substation 130. Each PCS package 131 has a PCS 132 and an intermediate transformer 133. The PCS 132 is, for example, an inverter that converts a DC voltage input from the connection box 120 into an AC voltage. The intermediate transformer 133 converts an output voltage of the PCS 132 into a predetermined voltage.

[0014] The extra-high voltage substation 140 includes a high-voltage switch gear 141, an extra-high voltage transformer 142, and an extra-high voltage switch gear 143. The high-voltage switch gear 141 switches whether or not to supply the high-voltage current input from each intermediate transformer 133 to the extra-high voltage transformer 142. When the high-voltage switch gear 141 is in the on state, the high-voltage current is supplied to the extra-high voltage transformer 142. The extra-high voltage transformer 142 converts the intermediate voltage input from each intermediate transformer 133 via the high-voltage switch gear 141 into an extra-high voltage. The extra-high voltage switch gear 143 switches whether or not to output the output power of the extra-high voltage transformer 142 to the commercial grid 300. When the extra-high voltage switch gear 143 is in the on state, the output power of the extra-high voltage transformer 142 is supplied to the commercial grid 300 from an interconnection point P, which is the output end of the extra-high voltage switch gear 143, as the power generated by the photovoltaic power plant 100.

[0015] The power measuring device 150 measures the amount of power of each of the power devices, such as the junction box 120, the PCS 132, the intermediate transformer 133, the high-voltage switch gear 141, the extra-high-voltage transformer 142, and the extra-high-voltage switch gear 143. In this embodiment, the measurement data of the power measuring device 150 corresponds to power generation data actually measured in the photovoltaic power plant 100.

[0016] The environmental measuring device 160 includes a pyranometer 161, a thermometer 162, and an anemometer 163. The pyranometer 161 measures the solar radiation intensity of sunlight incident on the PV panel 110. The thermometer 162 measures the temperature of the photovoltaic power plant 100. The anemometer 163 measures the wind power and wind direction of the photovoltaic power plant 100. Note that the environmental measuring device 160 is not limited to the pyranometer 161, the thermometer 162, and the anemometer 163. The environmental measuring device 160 may include, for example, a hygrometer that measures the humidity of the photovoltaic power plant 100, a rain gauge or a snow gauge that measures the amount of rainfall or snowfall, and a thermometer that measures the temperature of the PV panel 110. In this embodiment, the measurement data of the various devices constituting the environmental measuring device 160 corresponds to the weather data actually measured in the photovoltaic power plant 100.

[0017] In this embodiment, the power measuring instrument 150 and the environmental measuring instrument 160 transmit measurement data via a network to the photovoltaic power generation performance evaluation device 200. The photovoltaic power generation performance evaluation device 200 evaluates the power generation performance of the photovoltaic power plant 100 using the measurement data.

[0018] Here, the power generation performance of a solar power plant is often calculated separately. If the configuration of the solar power plant is simple, it can be calculated using the method specified in JISC8907 (method of estimating the amount of power generated by a solar power generation system), for example.

[0019] However, when the site is large or the panel layout is complex, it is difficult to calculate with simple calculations. Therefore, specialized software that can calculate the amount of power that can be obtained under certain weather conditions is often used.

[0020] Here, examples of commonly used software include PVSYST, Helios3D, and PVcase, but other software may be used. In these software, various conditions are set, such as weather and solar radiation conditions, the arrangement of the PV panels 110, the performance of the PV panels 110, the PCS 132, and the intermediate transformer 133, cable performance and loss, and loss due to albedo and dirt. This makes it possible to calculate the power generation performance under certain weather conditions without requiring past weather data.

[0021] However, most of the above software is on-premise software and requires a lot of time for calculations, making it difficult to automatically calculate the weather conditions obtained from data collection.

[0022] Therefore, in this embodiment, certain weather conditions and the simulation results under those conditions are set as explanatory variables and objective variables, respectively, and the responses are learned in advance by AI (Artificial Intelligence), machine learning, or the like, so that the photovoltaic power generation performance evaluation device 200 does not require the above-mentioned software and can calculate the power generation performance under certain weather conditions in real time. The photovoltaic power generation performance evaluation device 200 according to this embodiment will be described below.

[0023] Fig. 2 is a block diagram showing a configuration example of the photovoltaic power generation performance evaluation device 200. As shown in Fig. 2, the photovoltaic power generation performance evaluation device 200 includes a data acquisition unit 220, a storage unit 210, a data acquisition unit 220, a data processing unit 230, and an output unit 240. The photovoltaic power generation performance evaluation device 200 is configured by a computer having hardware resources such as a CPU, a ROM, a RAM, a HDD, and a GPU, and in which information processing by software is realized using the hardware resources by the CPU executing various programs. Furthermore, a photovoltaic power generation evaluation method using the photovoltaic power generation performance evaluation device 200 can be realized by causing the computer to execute various programs.

[0024] The storage unit 210 stores a learning model for calculating the power generation performance of the photovoltaic power plant 100. This learning model is created in advance by machine learning the results of simulating the amount of power generation expected from various weather data. For the weather data used in the machine learning, for example, actual past weather data such as weather databases provided and sold by the Japan Meteorological Agency or weather companies can be used.

[0025] In this embodiment, since the power generation performance is obtained by the simulation result, the power generation data actually measured in the past in the photovoltaic power plant 100 is not necessary. In addition, in this embodiment, it is not necessary to use the weather data actually measured in the past as the weather data used for machine learning. That is, the data itself input as the explanatory variables may be dummy weather data instead of actual weather data. The dummy weather data is weather data in which the range of the variables of the weather conditions such as the wind speed, wind direction, and temperature is preset within the range that can be taken according to the simulation conditions. Thus, according to this embodiment, even if there is no actual power generation data before the operation of the photovoltaic power plant 100 starts or for a while after the operation starts, it is possible to compare the power generation performance calculated using the learning model with the actual power generation performance. Of course, after the power generation data is accumulated, the data may be used.

[0026] The data acquiring unit 220 acquires measurement data from the power measuring device 150 and the environmental measuring device 160 directly or via a network.

[0027] The data processing unit 230 processes the measurement data input from the data acquisition unit 220 and evaluates the power generation performance of the photovoltaic power plant 100.

[0028] The output unit 240 outputs predetermined information. For example, the output unit 240 outputs the evaluation result of the data processing unit 230. The photovoltaic power generation performance evaluation device 200 according to this embodiment includes a device for displaying images, such as a display that outputs the evaluation result. In other words, the output unit 240 controls the image displayed on the display. Note that this display may be separate from the computer main body, or may be integrated with the computer main body.

[0029] The photovoltaic power generation performance evaluation device 200 according to this embodiment may control images displayed on a display of another computer connected via a network. In this case, an output unit 240 of the other computer may control the output of the evaluation results derived by the data processing unit 230.

[0030] In addition, in this embodiment, a display is exemplified as a device that displays an image, but other modes may be used. For example, information may be displayed using a projector. Furthermore, a printer that prints information on a paper medium may be used instead of a display. In other words, the objects controlled by the output unit 240 may include a projector or a printer.

[0031] Next, a method for evaluating the power generation performance of the photovoltaic power plant 100 will be described.

[0032] Fig. 3 is a flowchart showing the procedure of the method for evaluating the power generation performance of the photovoltaic power plant 100 according to the first embodiment. The flowchart shown in Fig. 3 includes a process for creating a learning model (steps S11 to S14) and a process for evaluating the power generation performance of the photovoltaic power plant 100 by the photovoltaic power generation performance evaluation device 200 using the learning model (steps S21 to S27).

[0033] First, the process of creating a learning model will be described. In this embodiment, the process of creating a learning model is executed by a machine learning device (not shown) separate from the photovoltaic power generation performance evaluation device 200. In the flowchart of Fig. 3, first, meteorological data is input to the machine learning device (step S11). As described above, this meteorological data may be actual meteorological data or dummy meteorological data.

[0034] Next, the machine learning device performs a power generation simulation based on the input weather data (step S12). In step S12, for example, the machine learning device calculates the amount of power generation using a simulation model in which each device of the photovoltaic power plant 100 shown in FIG.

[0035] The machine learning device creates a learning model by machine learning the weather data, which is an input condition, and the power generation amount, which is an output result (step S13). The machine learning used in step S13 includes, for example, a neural network (NN), a support vector machine (SVM), a decision tree, a random forest, a Bayesian method, or a logistic regression. However, the machine learning used in step S13 is not limited to the above-mentioned method, and may be another method.

[0036] The created learning model is stored from the machine learning device in the storage unit 210 of the photovoltaic power generation performance evaluation device 200 (step S14). This ends the process of creating the learning model.

[0037] Next, a description will be given of the power generation performance evaluation process performed by the photovoltaic power generation performance evaluation device 200. In the flowchart shown in Fig. 3, first, the data acquisition unit 220 acquires measurement data from the power measuring device 150 and the environmental measuring device 160 (step S21).

[0038] Next, the data processing unit 230 performs data pre-processing on the meteorological data, which is the measurement data of the environmental measuring device 160 (step S22). In step S22, the data processing unit 230 performs processing to convert the meteorological data into learning data so that the data can be applied to the learning model stored in the storage unit 210. Specifically, the data processing unit 230 performs, for example, input value range matching, input dimension matching, time resampling processing, outlier removal, and missing value processing on the meteorological data.

[0039] Next, the data processing unit 230 inputs the pre-processed weather data to the learning model read from the storage unit 210 (step S23). As a result, a first power generation performance is calculated (step S24). The first power generation performance indicates the amount of power generation of the photovoltaic power plant 100 that is estimated by machine learning under the weather conditions measured in the photovoltaic power plant 100.

[0040] Next, the data processing unit 230 derives a second power generation performance based on the measurement data of the power measuring device 150 (step S25). The second power generation performance indicates the actual amount of power generated by the entire photovoltaic power plant 100 measured by the power measuring device 150. The data processing unit 230, for example, extracts the amount of power measured at the interconnection point P from the measurement data of the power measuring device 150, and derives the extracted amount of power as the second power generation performance.

[0041] Next, the data processing unit 230 evaluates the power generation performance of the photovoltaic power plant 100 by comparing the second power generation performance with the first power generation performance (step S26). For example, it is assumed that the estimated power generation amount calculated as the first power generation performance is a guaranteed value of the power generation amount of the photovoltaic power plant 100 determined in advance for certain weather conditions. In this case, when the difference between the actual power generation amount derived as the second power generation performance and the estimated power generation amount is larger than a preset threshold value, the data processing unit 230 determines that some malfunction such as an abnormality or deterioration has occurred in the photovoltaic power plant 100. However, the method of determining the malfunction of the photovoltaic power plant 100 is not limited to the difference between the estimated power generation amount and the actual power generation amount.

[0042] Fig. 4 is a diagram for explaining another example of a method for determining whether or not there is a malfunction in the photovoltaic power plant 100. Fig. 4 shows a change over time in the ratio (actual power generation amount / estimated power generation amount) obtained by dividing the actual power generation amount by the estimated power generation amount. When this ratio is smaller than the numerical value 1.0, for example, the data processing unit 230 determines that a malfunction has occurred in the photovoltaic power plant 100. In Fig. 4, an abnormal value (one plot point significantly below 1.0) has occurred in the period from June 11 to June 16.

[0043] Finally, the output unit 240 outputs the evaluation result of the data processing unit 230 (step S27). In step S27, the output unit 240 displays, for example, an image of the judgment result of the malfunction of the photovoltaic power plant 100 as shown in FIG. 4. At this time, if the measured power generation amount is less than the estimated power generation amount, or if the tendency of the power generation amount expressed by the relational expression between the measured power generation amount and the estimated power generation amount is different, the output unit 240 displays the image in a color different from that in the normal state. In addition, the output unit 240 may notify the abnormality of the photovoltaic power plant 100 to a communication terminal used by the manager of the photovoltaic power plant 100 via email or SNS (Social Networking Service). In this case, since the abnormality is notified early, the manager can grasp the power generation performance of the photovoltaic power plant 100 in real time.

[0044] According to the present embodiment described above, a learning model used for evaluating the power generation performance of the photovoltaic power plant 100 is pre-stored in the storage unit 210. Furthermore, the data processing unit 230 evaluates the power generation performance of the photovoltaic power plant 100 based on a comparison result between an estimated power generation amount calculated using this learning model and an actual power generation amount derived from measurement data of the power measuring device 150. Therefore, even if there is no past power generation data, it is possible to evaluate the power generation performance of the photovoltaic power plant 100.

[0045] In addition, in this embodiment, on-premise software is not required, so the calculation time for the amount of power generation can be shortened.

[0046] Furthermore, in this embodiment, the power generation performance of the photovoltaic power plant 100 can be evaluated in real time by installing the photovoltaic power generation performance evaluation device 200. Therefore, it is possible to provide a service of installing the photovoltaic power generation performance evaluation device 200 to customers who are unable to evaluate the power generation performance of a photovoltaic power plant.

[0047] (First Modification) The following describes Modification 1. The configuration of the solar power plant according to this modification is similar to that of the solar power plant 100 according to the above-described first embodiment, and therefore a detailed description thereof will be omitted.

[0048] In the first embodiment, the data processing unit 230 of the photovoltaic power generation performance evaluation device 200 evaluates the power generation performance based on the power generation amount of the entire photovoltaic power plant 100, i.e., the power generation amount at the interconnection point P. On the other hand, in this modification, the data processing unit 230 evaluates the power generation performance for each of the intermediate substation 130, the PCS 132, or the junction box 120 in addition to the interconnection point P.

[0049] Fig. 5 is a diagram showing an example of a learning model database stored in the storage unit 210 according to the first modified example. In the learning model database 400 shown in Fig. 5, learning models M1 to M3 are stored for each machine learning target device, for example, the intermediate substation 130, the PCS 132, or the junction box 120. The learning models M1 to M3 can be created by the learning model creation process (steps S11 to S14) of the first embodiment described above.

[0050] When the learning models M1-M3 are created and stored in the memory unit 210, the data processing unit 230 calculates the estimated power generation amount for each of the intermediate substations 130, the PCS 132, or the junction box 120 from the learning models M1-M3. The data processing unit 230 also derives the actual power generation amount for each of the intermediate substations 130, the PCS 132, or the junction box 120 from the measurement data of the power measuring device 150. Then, the data processing unit 230 evaluates the power generation performance for each of the intermediate substations 130, the PCS 132, or the junction box 120 based on the result of comparing the estimated power generation amount with the actual power generation amount.

[0051] In this modification, explanatory variables may be constructed from multiple combinations of power devices installed in the solar power plant 100 to create a learning model. In this case, the more the measurement points are divided into multiple points, the easier it becomes to identify the cause and location of an abnormality when it occurs.

[0052] Furthermore, for example, in a solar power plant 100 in which multiple PCSs 132 are installed, when the data processing unit 230 compares the actual power generation amount with the expected power generation amount calculated from the learning model as described above, even if it is difficult to determine whether an abnormality exists from a single piece of power generation data, having multiple pieces of data makes statistical analysis possible, making it easier to determine whether an abnormality exists.

[0053] For example, when N (N is an integer of 2 or more) PCSs 132 are installed in the photovoltaic power plant 100, it becomes possible to calculate N system performance coefficients, so-called PR values. Here, when all the PCSs 132 are normal, the variation (deviation) of the N PR values ​​becomes small. On the other hand, when a performance degradation occurs in a specific part in the photovoltaic power plant 100, a specific PR value tends to decrease compared to the other (N-1) PR values. The data processing unit 230 can determine a malfunction of the photovoltaic power plant 100 based on the presence or absence of detection of such a trend. Note that the objects to be compared as multiple data are not limited to measurement points. Moreover, the target data is not limited to PR values, and may be data measured in the photovoltaic power plant 100, such as weather data and power generation amount. Furthermore, the statistical analysis method is not particularly limited as long as it is a method that can compare and analyze based on multiple data.

[0054] (Second Modification) The following describes Modification 2. The configuration of the solar power plant according to this modification is also similar to that of the solar power plant 100 according to the above-described first embodiment, and therefore a detailed description thereof will be omitted.

[0055] In this modification, the data processing unit 230 updates the learning model stored in the storage unit 210 from the initially created version as needed. For example, if the difference between the first power generation performance based on the actual weather data measured by the environmental measuring device 160 and the second power generation performance based on the measurement data of the power measuring device 150 is outside the allowable range, the data processing unit 230 re-learns using the second power generation performance as the objective variable, or performs transfer learning or fine tuning on the learning model. This makes it possible to evaluate the power generation performance of the photovoltaic power plant 100 with higher accuracy.

[0056] (Third Modification) The following describes Modification 3. The configuration of the solar power plant according to this modification is also similar to that of the solar power plant 100 according to the above-described first embodiment, and therefore a detailed description thereof will be omitted.

[0057] In the photovoltaic power plant 100, while there is generated power, there is also power consumption that consumes power. For example, the power consumption of an air conditioner that cools the PCS 132 and the photovoltaic power generation performance evaluation device 200 having a communication function is large. The power consumption of an air conditioner is easily affected by the environmental temperature, and is therefore correlated with weather condition data.

[0058] Therefore, in this modification, in the learning model creation process (step S13), data on power consumption is also input to the explanatory variables in addition to the weather data. This improves the accuracy of the learning model. As a result, the possibility of determining whether the cause of the deterioration of power generation performance is due to an abnormality in the power equipment or due to power consumption in the photovoltaic power plant 100 is improved.

[0059] In addition, in this modification, the data processing unit 230 may use or exclude weather data and power generation data under specific conditions when determining the first power generation performance and the second power generation performance. For example, during low solar radiation, the calculation of the power generation amount is likely to be unstable, so measurement data lower than a preset reference value may be excluded. It is also useful to perform performance evaluation by excluding data under conditions that affect the performance calculation but should not be included in the performance calculation, such as during output suppression.

[0060] Although some embodiments of the present invention have been described, these embodiments are presented as examples and are not intended to limit the scope of the invention. These novel embodiments can be implemented in various other forms, and various omissions, substitutions, and modifications can be made without departing from the spirit of the invention. These embodiments and their modifications are included in the scope and spirit of the invention, and are included in the scope of the invention and its equivalents described in the claims. [Explanation of symbols]

[0061] 100: Solar power plant 110:PV panel 120: Junction box 130: Intermediate substation 131: PCS package 132: PCS 133: Intermediate transformer 140: Extra-high voltage substation 141: High-voltage switchgear 142: Extra-high voltage transformer 143: Extra-high voltage switchgear 150: Power measuring equipment 160:Environmental measuring equipment 161:Pyranometer 162: Thermometer 163: Wind gauge 200: Photovoltaic power generation performance evaluation device 210: Storage section 220: Data acquisition section 230: Data processing unit 240: Output section 300: Commercial line 400: Learning model database

Claims

1. A storage unit that stores a learning model created by machine learning using weather conditions as explanatory variables and a result of simulating the power generation performance of a photovoltaic power plant based on the weather conditions as a target variable; and A data acquisition unit that acquires meteorological data measured at the solar power plant and power generation data measured by a plurality of types of power equipment installed at the solar power plant; a data processing unit that evaluates a power generation performance of the photovoltaic power plant based on a comparison result between a first power generation performance calculated by inputting the weather data into the learning model and a second power generation performance derived from the power generation data; and A photovoltaic power generation performance evaluation device comprising:

2. The photovoltaic power generation performance evaluation device according to claim 1 , wherein the learning model is created using dummy weather data.

3. The learning model is created for each type of the electric power device, The photovoltaic power generation performance evaluation device according to claim 1 , wherein the data processing unit compares the first power generation performance with the second power generation performance for each type of the electric power device.

4. The photovoltaic power generation performance evaluation device according to claim 1 , wherein the data processing unit evaluates the power generation performance of the photovoltaic power plant based on system performance coefficients of a plurality of the power devices.

5. The photovoltaic power generation performance evaluation device according to claim 1 , wherein the data processing unit updates the learning model in accordance with a comparison result between the first power generation performance and the second power generation performance.

6. 3. The photovoltaic power generation performance evaluation device according to claim 1, wherein data on the power consumed in the photovoltaic power plant is input as an explanatory variable of the learning model.

7. The photovoltaic power generation performance evaluation device according to claim 1 , further comprising an output unit that outputs an evaluation result of the data processing unit.

8. The photovoltaic power generation performance evaluation device according to claim 7 , wherein, when the evaluation result of the data processing unit indicates an abnormality in the photovoltaic power plant, the output unit notifies a manager of the abnormality.

9. A learning model created by machine learning is stored in advance, the learning model having weather conditions as explanatory variables and a result of simulating the power generation performance of the photovoltaic power plant based on the weather conditions as a target variable; Acquire weather data measured at the solar power plant and power generation data measured by a plurality of types of power equipment installed at the solar power plant; evaluating the power generation performance of the solar power plant based on a comparison result between a first power generation performance calculated by inputting the weather data into the learning model and a second power generation performance derived from the power generation data; Photovoltaic power generation performance evaluation method.

10. A process of storing a learning model created by machine learning using weather conditions as explanatory variables and a result of simulating the power generation performance of a photovoltaic power plant based on the weather conditions as a target variable; and A process of acquiring meteorological data measured at the solar power plant and power generation data measured by a plurality of types of power equipment installed at the solar power plant; A process of evaluating a power generation performance of the photovoltaic power plant based on a comparison result between a first power generation performance calculated by inputting the weather data into the learning model and a second power generation performance derived from the power generation data; A program for causing a computer to execute the following.

Images