Solar power generation management methods, solar power generation management programs, and solar power generation management systems

Abstract

To predict, with high accuracy, the state of snow accretion on a photovoltaic power generation panel.SOLUTION: A photovoltaic power generation management method of an embodiment includes: a snow accumulation prediction step of, based on weather prediction data and photographed image data of a predetermined region including an installation place of a photovoltaic power generation panel for which the state of snow accretion is predicted, predicting the state of snow accretion in the predetermined region at a prediction target time and outputting a result of snow accumulation prediction; and a snow accretion prediction step of, based on the result of snow accumulation prediction, predicting the state of snow accretion on the photovoltaic power generation panel at the prediction target time and outputting a result of snow accretion prediction.SELECTED DRAWING: Figure 5

Description

Technical Field

[0001] Embodiments of the present invention relate to a photovoltaic power generation management method, a photovoltaic power generation management program, and a photovoltaic power generation management system.

Background Art

[0002] In recent years, due to emergency response measures resulting from the normalization of disasters caused by abnormal weather, and the introduction of renewable energy (renewable energy power generators) as a business strategy for electric power companies and industrial consumers, the correlation between infrastructure-related issues / events and meteorological information has increased. For renewable energy, for example, advanced technology for predicting the power generation amount of photovoltaic power generation facilities (hereinafter also referred to as "PV" (Photovoltaic)) is required. By implementing highly accurate PV power generation amount prediction, it is ultimately possible to contribute to reduction of imbalance costs in renewable energy balancing.

[0003] PV power generation amount prediction is performed using, for example, satellite image data and solar radiation amount data. However, when there is snow accretion on the PV panel, the PV power generation amount decreases, so if snow accretion is not taken into account, a large discrepancy may occur between the predicted value and the actual value of the power generation amount.

[0004] As a countermeasure, for example, there is a technology for predicting the snow accretion state on PV panels mainly using information such as snow depth and the number of elapsed days after the end of snowfall.

Prior Art Documents

Patent Documents

[0005]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0006] However, since snow accumulation on PV panels is influenced by factors other than snow depth and the number of days since the end of snowfall, there is room for improvement in terms of prediction accuracy with the conventional technology described above.

[0007] Therefore, the problem that the present invention aims to solve is to provide a solar power generation management method, a solar power generation management program, and a solar power generation management system that can predict the snow accumulation situation on solar power generation panels with high accuracy. [Means for solving the problem]

[0008] The embodiment of the solar power generation management method includes image data of a predetermined area, including the installation location of the solar power generation panels for which snow accumulation is to be predicted. Snow depth estimation results obtained by estimating the snow depth in the predetermined area at the time of prediction execution. and a snow accumulation prediction step which predicts the snow accumulation status of the predetermined area at the time of prediction based on weather forecast data and outputs a snow accumulation prediction result, and a snow accumulation prediction step which predicts the snow accumulation status of the solar power generation panel at the time of prediction based on the snow accumulation prediction result and outputs a snow accumulation prediction result, The acquisition step involves obtaining the solar power generation amount prediction result from the aforementioned solar power generation panel, and the correction step involves calculating the snow accumulation probability of the snow accumulation prediction result based on the aforementioned weather forecast data for a predetermined period and a predetermined threshold, and correcting the solar power generation amount prediction result based on a correction coefficient using the snow accumulation probability. Includes. [Brief explanation of the drawing]

[0009] [Figure 1] Figure 1 is a graph showing an example of the temporal changes in PV power generation and PV power generation prediction results using conventional technology. [Figure 2] Figure 2 is a block diagram showing the functional configuration of the solar power generation management device according to the first embodiment. [Figure 3] Figure 3 is a schematic diagram showing the first snow accumulation prediction algorithm in the first embodiment. [Figure 4] Figure 4 is a schematic diagram showing the second snow accumulation prediction algorithm in the first embodiment. [Figure 5] Figure 5 is a flowchart showing the processing performed by the solar power generation management device of the first embodiment. [Figure 6] Figure 6 is a block diagram showing the functional configuration of the solar power generation management device according to the second embodiment. [Figure 7]Figure 7 is a flowchart showing the processing performed by the solar power generation management device of the second embodiment. [Modes for carrying out the invention]

[0010] Hereinafter, embodiments (first and second embodiments) of the solar power generation management method, solar power generation management program, and solar power generation management system of the present invention will be described with reference to the drawings.

[0011] (premise) Figure 1 is a graph showing an example of the temporal changes in PV power generation and PV power generation prediction results using conventional technology. Here, the predicted solar power generation result P and the actual output value R1 for a sunny day are shown.

[0012] Normally, the predicted solar power generation on a sunny day looks like the one labeled P in Figure 1. And the actual output value is expected to generally follow this. However, if there is snowfall the day before and snow accumulates on the PV panels (solar power generation panels), the actual output value will not follow the predicted solar power generation result.

[0013] Here, we assume that snow accumulates on the PV panel at time t0, slightly after sunrise, and that this snow gradually disappears from around time t1 due to melting and sliding, until it is completely gone by time t2. While the presence or absence of snow on the PV panel is thought to be related to the presence or absence of snow on the ground, PV panels in Japan are installed with a slope of approximately 10° (this varies depending on latitude), so there is a possibility that snow on the PV panel may slide off. Furthermore, if the weather is good the day after snowfall, the temperature of the PV panel may rise, causing snow melting or snow sliding.

[0014] In the example in Figure 1, the actual output value, as indicated by the sign R1, is zero between times t0 and t1, starts to increase from time t1, and catches up with the predicted solar power generation result P at time t2. In other words, during the period from time t0 to just before time t2, the actual output value R1 is not proportional to the amount of solar radiation due to snow accumulation on the PV panel.

[0015] Note that there is a method of installing a dedicated camera or sensor to recognize the snow accretion status on a PV panel, but this has problems in terms of labor and cost. Accordingly, there is a need to predict the snow accretion status on a PV panel with high accuracy without installing such a dedicated camera or sensor.

[0016] However, as described above, the conventional technology that predicts the snow accretion status on a PV panel mainly using information such as snow depth and the number of elapsed days after the end of snowfall leaves room for improvement in terms of prediction accuracy. Therefore, in the following, a technology capable of predicting the snow accretion status on a PV panel with higher accuracy than before will be described.

[0017] Note that in the present specification, the state of snow adhering to a PV panel is referred to as "snow accretion", and the state of snow accumulated on other areas such as the ground is referred to as "snow cover".

[0018] (First Embodiment) Figure 2 is a block diagram showing the functional configuration of a photovoltaic power generation management device 1 according to the first embodiment. The photovoltaic power generation management device 1 is a computer device, and includes a processing unit 2, a storage unit 3, an input unit 4, an output unit 5, and a communication unit 6.

[0019] Note that part of the functions of the photovoltaic power generation management device 1 can also be implemented by, for example, cloud computing, but in the present embodiment, for the sake of concise description, the photovoltaic power generation management device 1 is described as a single computer device.

[0020] The storage unit 3 is implemented by, for example, a RAM (Random Access Memory), a ROM (Read Only Memory), a flash memory, an HDD (Hard Disk Drive), an SSD (Solid State Drive), or the like. The storage unit 3 stores, for example, the operation program of the processing unit 2, various data, and various calculation results from the processing unit 2.

[0021] Specifically, the memory unit 3 stores, for example, image data (such as satellite image data) of a predetermined area including the installation location of the PV panel (hereinafter also simply referred to as "installation location"), weather forecast data for that predetermined area (temperature, humidity, solar radiation, etc.), threshold determination data, machine learning data, location information of the installation location, and time characteristic information related to the prediction target time (information on characteristics such as season, time of day (morning, noon, night, etc.) (temperature, humidity, solar radiation, etc.)).

[0022] The processing unit 2 is implemented, for example, by a CPU (Central Processing Unit). The processing unit 2 operates based on the operation program and various data stored in the memory unit 3, and as functional units, it implements the acquisition unit 21, the snow depth prediction unit 22, and the snow accumulation prediction unit 23.

[0023] The acquisition unit 21 acquires various types of information from the storage unit 3 and external devices.

[0024] The snow accumulation prediction unit 22 predicts the snow accumulation in a predetermined area (hereinafter also simply referred to as the "predetermined area") that includes the installation locations of the solar power generation panels for which snow accumulation is to be predicted, based on captured image data (e.g., satellite image data) and weather forecast data, and outputs a snow accumulation prediction result.

[0025] For example, the snow depth prediction unit 22 estimates the snow depth in a predetermined area at the time of prediction execution based on satellite image data, and uses threshold determination or machine learning based on the snow depth estimation result and weather forecast data to predict the snow depth in a predetermined area at the target time, and outputs information on snow depth or presence or absence of snow as a snow depth prediction result.

[0026] Here, we will explain the estimation of snow cover conditions in a given area at the time of prediction based on satellite image data. By performing image synthesis processing (e.g., RGB (Red, Green, Blue) synthesis processing) on ​​satellite image data, it becomes possible to determine not only the presence or absence of snow on the ground, but also vegetation and clouds in the upper atmosphere. For example, snow cover has a high reflectivity to visible light and a high absorption rate to near-infrared light, while clouds have a high reflectivity to both visible light and near-infrared light. These characteristics allow for the distinction between snow cover and clouds.

[0027] The image synthesis techniques (recipes) used could include, for example, those that produce a composite result of visible light or those that distinguish between snow and fog. However, it is also possible to use a combination of multiple techniques.

[0028] Next, we will explain the snow depth estimation results for a given area at the time of prediction (e.g., the present) and the prediction of the snow depth in that area at the target time (e.g., the next day) based on weather forecast data. Two types of methods can be used for this prediction, for example, thresholding and machine learning.

[0029] In threshold determination, parameters are used to determine the threshold. In machine learning, for example, random forests, regression analysis, and logistic regression are used. Snow depth prediction results can be of two types, for example, snow depth and presence or absence of snow, but it is also possible to set a threshold for snow depth and classify it as either presence or absence of snow.

[0030] When comparing thresholding and machine learning, it is considered that machine learning is more accurate when the amount of training data for the snow depth prediction model is sufficiently large, while thresholding is more accurate when the amount of training data is sufficiently small. This is because if the amount of training data is insufficient, the snow depth prediction model may overfit. Therefore, the amount of training data at that boundary is defined as the first predetermined value.

[0031] The snow depth prediction unit 22, for example, uses threshold determination if the amount of training data for the machine learning snow depth prediction model is less than a first predetermined value, and uses machine learning if the amount of training data for the machine learning snow depth prediction model is equal to or greater than the first predetermined value. Specifically, for example, for locations where data is difficult to obtain, threshold determination is used if the amount of training data is less than the first predetermined value. Furthermore, it is possible to perform snow depth prediction using threshold determination when the amount of training data is less than the first predetermined value immediately after system installation, and then switch the prediction method to machine learning after a predetermined period has elapsed and the amount of training data is equal to or greater than the first predetermined value. In this way, threshold determination and machine learning can be appropriately used depending on the amount of training data.

[0032] Furthermore, the scope of the snowfall prediction model can be determined arbitrarily. In other words, for example, it could be a nationwide uniform model, or it could be an area-specific model. Here, the area can be a division of administrative units such as cities, towns, villages, prefectures, or regions, or it can be arbitrarily set by the implementer.

[0033] For example, a company that operates multiple solar power plants within a prefecture can create a snowfall prediction model for all of its power plants in the prefecture. Alternatively, the same company can specify area divisions within the prefecture and create snowfall prediction models for each area. However, if areas are arbitrarily defined, weather forecast data for at least one location in that area must be available.

[0034] The snow accumulation prediction unit 23 predicts the snow accumulation status of the solar power generation panels at the target time based on the snow accumulation prediction results, and outputs the snow accumulation prediction results.

[0035] For example, the snow accumulation prediction unit 23 predicts the snow accumulation status of the solar power generation panels at the target time based on at least one of the following: snow accumulation prediction results, weather forecast data, location information of the installation site, and time characteristic information related to the target time of prediction, using threshold determination or machine learning, and outputs information on the presence or absence of snow accumulation or the probability of snow accumulation as a snow accumulation prediction result.

[0036] The following describes how to output information on the presence or absence of snow accumulation using Figure 3, and how to output information on the probability of snow accumulation using Figure 4. Figure 3 is a schematic diagram showing the first snow accumulation prediction algorithm in the first embodiment. Figure 4 is a schematic diagram showing the second snow accumulation prediction algorithm in the first embodiment.

[0037] In Figure 3, first, in step S11, it is determined whether or not there is snow on the ground at the predicted time. If yes, the process proceeds to step S12; if no, it proceeds to step S13, where it is determined that there is no snow and the process ends.

[0038] In step S12, it is determined whether or not there is snow accumulation in the most recent time (h hours). If yes, the process proceeds to step S14; otherwise, it proceeds to step S15, where it is determined that there is no snow accumulation and the process ends.

[0039] In step S14, it is determined whether the average temperature over the most recent h hours is below the temperature threshold T. If yes, the process proceeds to step S16; otherwise, it proceeds to step S17, where it is determined that there is no snow accumulation and the process ends.

[0040] In step S16, it is determined whether the average solar radiation over the most recent h hours is less than the solar radiation threshold R. If Yes, the process proceeds to step S18, where it is determined that snow accumulation is present and the process ends. If No, the process proceeds to step S19, where it is determined that there is no snow accumulation and the process ends.

[0041] Next, in Figure 4, in step S21, it is determined whether or not there is snow on the ground at the predicted time. If yes, the process proceeds to step S22; if no, the process proceeds to step S23, where the probability of snow accumulation is determined to be 0%, and the process ends.

[0042] In step S22, it is determined whether or not there is snow accumulation in the most recent time (h hours). If Yes, the process proceeds to step S24; otherwise, it proceeds to step S25, where the probability of snow accumulation is determined to be 10%, and the process ends.

[0043] In step S24, it is determined whether the average temperature over the most recent h hours is below the temperature threshold T. If yes, the process proceeds to step S26; otherwise, it proceeds to step S27, where the probability of snow accumulation is determined to be 50%, and the process ends.

[0044] In step S26, it is determined whether the average solar radiation over the most recent h hours is less than the solar radiation threshold R. If yes, the process proceeds to step S28, where the probability of snow accumulation is determined to be 100%, and the process ends. If no, the process proceeds to step S29, where the probability of snow accumulation is determined to be 80%, and the process ends.

[0045] In this way, snow accumulation predictions can be determined using a simple binary system of "yes" or "no" snow accumulation, or they can be determined as a probability.

[0046] Note that the judgment conditions used in Figures 3 and 4 are examples and are not limited to these. In addition, location information such as the latitude and longitude of the equipment to be predicted, equipment information such as the tilt angle and material of the PV panel, and time characteristic information such as the time of day and season may also be used. This information is related to the melting and sliding of snow on the PV panel and is therefore effective in improving the accuracy of snow accumulation prediction. Furthermore, the probability values ​​of the judgment results shown in Figure 4 can also be set arbitrarily, and similar to the judgment conditions, it is possible to vary the probability values ​​for each piece of equipment to be predicted using location information, equipment information, time characteristic information, etc.

[0047] Returning to the explanation of Figure 2, the snow accumulation prediction unit 23 uses threshold determination if the amount of training data for the snow accumulation learning model in machine learning is less than a second predetermined value, and uses machine learning if the amount of training data for the snow accumulation learning model in machine learning is equal to or greater than the second predetermined value. Similar to the snow depth prediction case, if the amount of training data is insufficient, the model may overfit. Therefore, the amount of training data at that boundary is set as the second predetermined value, and threshold determination and machine learning are used appropriately.

[0048] Input unit 4 is a means of information input by the user, such as a mouse, keyboard, or touch panel.

[0049] Output unit 5 is a display device such as an LCD (Liquid Crystal Display) or an audio output device such as a speaker.

[0050] The communication unit 6 is a communication interface for communicating (sending and receiving) various types of information with external devices and various sensors within the power system.

[0051] Figure 5 is a flowchart showing the processing performed by the solar power generation management device 1 of the first embodiment. First, in step S1, the snow depth prediction unit 22 estimates the current snow depth in a predetermined area (at the time of prediction execution) based on satellite image data.

[0052] Next, in step S2, the snow depth prediction unit 22 predicts the snow depth in a predetermined area at the target time using threshold determination or machine learning, based on the snow depth estimation result from step S1 and weather forecast data, and outputs the snow depth prediction result. For example, using the estimation result from step S1, it performs a snow depth prediction for the location in question and obtains snow depth prediction results for a predetermined number of days (for example, about one week) from the current day.

[0053] Next, in step S3, the snow accumulation prediction unit 23 predicts the snow accumulation status of the solar power generation panels at the target time using threshold determination or machine learning, based on at least one of the following: the snow accumulation prediction result from step S2, weather forecast data, location information of the installation site, and time characteristic information related to the target time of prediction, and outputs the snow accumulation prediction result.

[0054] In this way, according to the solar power generation management device 1 of the first embodiment, as described above, first, the snow accumulation status of a predetermined area at the present time (the time of prediction execution) is estimated based on satellite image data, etc., then the snow accumulation status of the predetermined area at the target time is predicted based on the snow accumulation estimation result and weather forecast data, etc., and then the snow accumulation status of the solar power generation panels at the target time is predicted based on the snow accumulation prediction result. Through this stepwise processing, the snow accumulation status on PV panels can be predicted with higher accuracy than conventional methods, even without installing dedicated cameras or sensors.

[0055] Furthermore, the snow accumulation prediction results can be used for various subsequent information processing. As an example, the case in which the solar power generation prediction results are corrected based on the snow accumulation prediction results will be described in the following second embodiment.

[0056] (Second Embodiment) A second embodiment will now be described. Details similar to those in the first embodiment will be omitted as appropriate. Figure 6 is a block diagram showing the functional configuration of the solar power generation management device 1 in the second embodiment. Compared to Figure 2, a correction unit 24 has been added to the processing unit 2.

[0057] The acquisition unit 21 acquires the predicted solar power generation amount from the PV panels. This predicted solar power generation amount may be generated by the solar power generation management device 1, or it may be acquired from another computer device.

[0058] The correction unit 24 corrects the solar power generation forecast result based on the snow accumulation forecast result. For example, the correction unit 24 corrects the solar power generation forecast result based on a correction coefficient corresponding to the snow accumulation forecast result.

[0059] The following details the correction of solar power generation forecast results. For example, one possible method is to multiply the power output forecast result by a correction coefficient c_r (0≦c_r≦1) for times when snow accumulation is determined to be present.

[0060] Furthermore, there are two possible methods for calculating this correction coefficient: using a constant value or using a variable value. If a constant value is used, one possible method is to pre-determine the value of the correction coefficient for any area and any period and then provide it to the system.

[0061] When using variable values, possible methods for applying these variations include using time-based information (seasonal, monthly, weekly, daily, or hourly), using the probability of snow accumulation from snow accumulation forecasts, or using location information such as the latitude and longitude of the solar power generation facility being corrected.

[0062] When using the snow accumulation probability from the snow accumulation forecast results, one possible calculation method is to set the correction coefficient to 0.9 if the snow accumulation probability is 10%, and to 0.8 if the snow accumulation probability is 20%. In other words, if the snow accumulation probability is SP and the correction coefficient is CF, the correction coefficient can be calculated using the following equation (1). CF = (100 - SP) × 0.01 ... (1)

[0063] Figure 7 is a flowchart showing the processing performed by the solar power generation management device 1 of the second embodiment. Steps S1 to S3 are the same as steps S1 to S3 in Figure 5.

[0064] After step S3, in step S4, the correction unit 24 determines whether or not correction of the PV power generation forecast result is necessary. If yes, the process proceeds to step S5; otherwise, the process ends.

[0065] In step S5, the correction unit 24 corrects the solar power generation forecast result based on a correction coefficient corresponding to the snow accumulation forecast result.

[0066] In this way, according to the second embodiment, by correcting the solar power generation prediction result based on the snow accumulation prediction result, a highly accurate solar power generation prediction result can be obtained.

[0067] While embodiments of the present invention have been described above, these embodiments are presented as examples only and are not intended to limit the scope of the invention. These 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 variations are included in the scope and spirit of the invention, as well as in the claims and their equivalents.

[0068] Furthermore, the program executed by the solar power generation management device 1 of this embodiment can be provided as an installable or executable file recorded on a recording medium readable by a computer device, such as a CD (Compact Disc)-ROM (Read Only Memory), flexible disk (FD), CD-R (Recordable), or DVD (Digital Versatile Disk). Alternatively, the program may be provided or distributed via a network such as the Internet.

[0069] The snow accumulation prediction for PV panels and the correction of the solar power generation prediction results may be performed by the same device (program), or they may be performed by separate devices (programs). In the latter case, for example, as in the first embodiment, the solar power generation management device 1 may perform a snow accumulation prediction for PV panels and transmit the snow accumulation prediction result to another computer device, which may then perform a solar power generation prediction and a correction of the solar power generation prediction result using the snow accumulation prediction result.

[0070] Furthermore, the image data of a designated area, including the installation location of the PV panels, is not limited to satellite image data, but may also include other data such as aerial photographs. [Explanation of Symbols]

[0071] 1...Solar power generation management device, 2...Processing unit, 3...Storage unit, 4...Input unit, 5...Output unit, 6...Communication unit, 21...Acquisition unit, 22...Snow depth prediction unit, 23...Snow accumulation prediction unit, 24...Correction unit

Claims

1. A snow accumulation prediction step involves estimating the snow accumulation status of a predetermined area at the time of prediction execution from captured image data of the area including the installation location of solar power generation panels to be predicted, and predicting the snow accumulation status of the predetermined area at the time of prediction based on weather forecast data, and outputting a snow accumulation prediction result. A snow accumulation prediction step that predicts the snow accumulation status of the solar power generation panels at the predicted time based on the snow accumulation prediction result and outputs a snow accumulation prediction result, The acquisition step involves obtaining the predicted solar power generation amount from the aforementioned solar power generation panels, A solar power generation management method comprising: a correction step of calculating the probability of snow accumulation in the snow accumulation forecast result based on the weather forecast data for a predetermined period and a preset threshold, and correcting the solar power generation forecast result based on a correction coefficient using the snow accumulation probability.

2. The solar power generation management method according to claim 1, wherein the snow accumulation prediction step estimates the snow accumulation status of the predetermined area at the time of prediction execution based on the captured image data, predicts the snow accumulation status of the predetermined area at the target time using threshold determination or machine learning based on the snow accumulation estimation result and the weather forecast data, and outputs information on snow depth or presence or absence of snow as the snow accumulation prediction result.

3. The solar power generation management method according to claim 1, wherein the snowfall prediction step uses satellite image data as the captured image data.

4. The solar power generation management method according to claim 1, wherein the snow accumulation prediction step predicts the snow accumulation status of the solar power generation panels at the predicted time, based on at least one of the following in addition to the snow accumulation prediction result: the weather forecast data, the location information of the installation site, and the time characteristic information relating to the prediction target time, using threshold determination or machine learning, and outputs information on the presence or absence of snow accumulation or the probability of snow accumulation as the snow accumulation prediction result.

5. The aforementioned snowfall prediction step is: If the amount of training data for the snow depth prediction model in the aforementioned machine learning is less than a first predetermined value, the threshold determination is used. The solar power generation management method according to claim 2, wherein if the amount of training data for the snow depth prediction model in the machine learning is equal to or greater than a first predetermined value, the machine learning method is used.

6. The aforementioned snow accumulation prediction step is: If the amount of training data for the snow accumulation learning model in the aforementioned machine learning is less than the second predetermined value, the threshold determination is used. The solar power generation management method according to claim 4, wherein if the amount of training data for the snow accumulation learning model in the machine learning is equal to or greater than a second predetermined value, the machine learning method is used.

7. Computers, A snow depth prediction unit predicts the snow depth of a predetermined area at a target time based on photographic image data of the area including the installation location of solar power generation panels to be predicted, estimates the snow depth of the predetermined area at the time of prediction execution, and weather forecast data, and outputs a snow depth prediction result. A snow accumulation prediction unit predicts the snow accumulation status of the solar power generation panels at the predicted time based on the snow accumulation prediction result and outputs a snow accumulation prediction result. An acquisition unit that acquires the predicted amount of solar power generated by the aforementioned solar power generation panels, A solar power generation management program that functions as a correction unit that calculates the probability of snow accumulation in the snow accumulation forecast result based on the weather forecast data for a predetermined period and a preset threshold, and corrects the solar power generation forecast result based on a correction coefficient using the snow accumulation probability.

8. A snow depth prediction unit predicts the snow depth of a predetermined area at a target time based on photographic image data of the area including the installation location of solar power generation panels to be predicted, estimates the snow depth of the predetermined area at the time of prediction execution, and weather forecast data, and outputs a snow depth prediction result. A snow accumulation prediction unit predicts the snow accumulation status of the solar power generation panels at the predicted time based on the snow accumulation prediction result and outputs a snow accumulation prediction result. An acquisition unit that acquires the predicted amount of solar power generated by the aforementioned solar power generation panels, A solar power generation management system comprising: a correction unit that calculates the probability of snow accumulation in the snow accumulation forecast result based on the weather forecast data for a predetermined period and a preset threshold, and corrects the solar power generation forecast result based on a correction coefficient using the snow accumulation probability.

9. A snow accumulation prediction step that estimates the snow accumulation in a predetermined area at the time of prediction execution based on image data of a predetermined area including the installation location of solar power generation panels to be predicted for snow accumulation, and predicts the snow accumulation in the predetermined area at the time of prediction using the estimated snow accumulation result and weather forecast data, using threshold judgment if the amount of training data for the snow accumulation prediction model in machine learning is less than a first predetermined value, and using machine learning if the amount of training data for the snow accumulation prediction model in machine learning is equal to or greater than the first predetermined value, and outputs information on snow depth or presence or absence of snow as a snow accumulation prediction result, A solar power generation management method comprising: a snow accumulation prediction step, which predicts the snow accumulation status of the solar power generation panels at the predicted time based on the snow accumulation prediction result, and outputs a snow accumulation prediction result.

10. A snow accumulation prediction step that predicts the snow accumulation in a predetermined area at a target time based on image data of a predetermined area including the installation location of solar power generation panels to be predicted for snow accumulation, and weather forecast data, and outputs a snow accumulation prediction result. A solar power generation management method comprising: a snow accumulation prediction step in which, in addition to the snow accumulation prediction result, based on at least one of the weather forecast data, the location information of the installation site, and the time characteristic information relating to the prediction target time, a threshold determination is used if the amount of training data for the snow accumulation learning model in machine learning is less than a second predetermined value, and machine learning is used if the amount of training data for the snow accumulation learning model in machine learning is equal to or greater than the second predetermined value, the snow accumulation status of the solar power generation panels at the prediction target time is predicted, and information on the presence or absence of snow accumulation or the probability of snow accumulation is output as a snow accumulation prediction result.

Images