Prediction device, prediction system, prediction method, and program

JPWO2023181231A5Active Publication Date: 2024-12-03NEC CORP
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
JP2024509555
Authority / Receiving Office
JP · JP
Patent Type
Applications
Current Assignee / Owner
Priority Date
2022-03-24
Filing Date
2022-03-24
Publication Date
2024-12-03
Estimated Expiration
2042-03-24

AI Technical Summary

Technical Problem

Existing power generation prediction technologies for solar power generation have high processing loads, necessitating continuous accurate predictions for charge/discharge control, which is resource-intensive.

Method used

A prediction device and method that acquire and correct the transition of predicted power supply amounts using sky images only when predetermined conditions are met, reducing processing load by performing highly accurate predictions only when necessary.

Benefits of technology

This approach reduces processing load while maintaining accurate power generation predictions, enabling efficient demand and supply balancing in solar power systems.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

A prediction device (10) comprises an acquisition unit (110) and a prediction unit (130). The acquisition unit (110) acquires the transition of a prediction value of a first power supply amount. The first power supply amount is the amount of supplied power to a supply target from a power supply source including at least one piece of solar power generation equipment. When a prescribed condition is satisfied, the prediction unit (130) uses an image to correct or re-predict the transition of the prediction value of the first power supply amount. The image has been obtained by imaging the sky in the surroundings of the solar power generation equipment.
Need to check novelty before this filing date? Find Prior Art

Description

Prediction device, prediction system, prediction method, and recording medium

[0001] The present invention relates to a prediction device, a prediction system, a prediction method, and a recording medium.

[0002] Since the amount of power generated by solar power generation facilities depends on weather and other factors, power generation prediction technology has been developed.

[0003] Patent Literature 1 describes a method for predicting fluctuations in the amount of power generated by a solar panel from an image of the sky captured by a camera, and also describes a method for mitigating fluctuations in the power input to a power conversion unit by discharging or charging a power storage unit when the predicted fluctuations are equal to or greater than a predetermined value.

[0004] Patent Document 2 describes a method for predicting the power generated by a power generation facility and controlling the discharge of a storage battery facility so that the power generation capacity of the facility does not exceed the interconnection capacity of the commercial power grid. It also describes a method for predicting the amount of solar radiation using an image that directly shows the relationship between the position of the sun and the position of clouds in the entire sky.

[0005] JP 2015-42102 A JP 2019-161863 A

[0006] The techniques disclosed in Patent Documents 1 and 2 described above require constant and highly accurate power generation prediction for charge / discharge control, which results in a heavy processing load for the prediction.

[0007] In view of the above-mentioned problems, an example of an object of the present invention is to provide a prediction device, a prediction system, a prediction method, and a recording medium that have a low processing load for predicting the amount of power generated by photovoltaic power generation.

[0008] According to one aspect of the present invention, there is provided a prediction device comprising: an acquisition means for acquiring a trend in a predicted value of a first power supply amount to a supply target from a power supply source including at least one solar power generation facility; and a prediction means for correcting or re-predicting the trend in the predicted value of the first power supply amount using an image obtained by photographing the sky around the solar power generation facility when a predetermined condition is satisfied.

[0009] According to one aspect of the present invention, there is provided a prediction system comprising: the prediction device described above; and an imaging means for imaging the sky around the solar power generation facility when the prediction device determines that the predetermined condition is satisfied.

[0010] According to one aspect of the present invention, there is provided a prediction method in which one or more computers acquire a trend in a predicted value of a first power supply amount to a supply target from a power supply source including at least one solar power generation facility, and when a predetermined condition is satisfied, correct or re-predict the trend in the predicted value of the first power supply amount using an image obtained by photographing the sky around the solar power generation facility.

[0011] According to one aspect of the present invention, there is provided a recording medium storing a program for causing a computer to execute a prediction method, the prediction method comprising the steps of: acquiring a trend in a predicted value of a first amount of power supply to a supply target from a power supply source including at least one solar power generation facility; and, when a predetermined condition is satisfied, correcting or re-predicting the trend in the predicted value of the first amount of power supply using an image obtained by photographing the sky around the solar power generation facility.

[0012] According to one aspect of the present invention, a prediction device, a prediction system, a prediction method, and a recording medium are provided that impose a low processing load on the amount of power generated by photovoltaic power generation.

[0013] FIG. 1 is a diagram illustrating an overview of a prediction device according to an embodiment. FIG. 2 is a diagram for explaining power supply to a supply target. FIG. 3 is a block diagram illustrating a configuration of a prediction device according to an embodiment. FIG. 4 is a diagram illustrating an example of a graph generated by an output unit. FIG. 5 is a diagram illustrating an example of an image displayed by notification information from the output unit. FIG. 6 is a diagram illustrating a configuration of a prediction system according to an embodiment. FIG. 7 is a diagram illustrating a computer for realizing a prediction device. FIG. 8 is a flowchart illustrating an overview of a prediction method executed by a prediction device according to an embodiment. FIG. 9 is a flowchart illustrating a flow of a prediction method according to an embodiment.

[0014] Hereinafter, embodiments of the present invention will be described with reference to the drawings. In all the drawings, like components are designated by like reference numerals, and the description thereof will be omitted as appropriate.

[0015] (Embodiment) FIG. 1 is a diagram illustrating an overview of a prediction device 10 according to an embodiment. The prediction device 10 includes an acquisition unit 110 and a prediction unit 130. The acquisition unit 110 acquires a trend in a predicted value of a first power supply amount. The first power supply amount is the amount of power supplied to a supply target from a power supply source including at least one photovoltaic power generation facility. When a predetermined condition is satisfied, the prediction unit 130 corrects or re-predicts the trend in the predicted value of the first power supply amount using an image. This image is obtained by photographing the sky around the photovoltaic power generation facility.

[0016] According to the prediction device 10, highly accurate prediction based on a sky image is performed only when necessary, thereby reducing the processing load. A detailed example of the prediction device 10 will be described below.

[0017] FIG. 2 is a diagram illustrating the power supply to a power supply target 50. In recent years, the use of renewable energy has been promoted. Examples of renewable energy include solar energy, wind energy, and geothermal energy. Among these, the installation of solar power generation facilities 52 for utilizing solar energy has progressed. Meanwhile, the amount of power supplied by the solar power generation facilities 52 depends on weather and other factors. When a power user (e.g., a business using power) cannot meet its power demand with the amount of power supplied by the solar power generation facilities 52 alone, the power user receives the shortfall in power from the power company. However, if the power user receives more power than the contracted power with the power company, the power bill increases. Therefore, in order to minimize the power supply from the power company, the power user needs to predict the amount of power supplied by the solar power generation facilities 52 and balance supply and demand. Note that the power company is, for example, a business that provides at least one of power supply (generation) and distribution.

[0018] Furthermore, electricity users may have a target amount of renewable energy usage set based on various policies, systems, etc. This target amount of usage may be constant or may vary over time. It is preferable for electricity users to use the supplied electricity derived from renewable energy as little as possible. Therefore, it is preferable to predict the amount of electricity supply from renewable energy sources, including solar power, and match demand to that amount of electricity supply.

[0019] For these reasons, while it is necessary to predict the amount of power supply from the photovoltaic power generation facility 52, highly accurate predictions are not always required. For example, it is important to make more accurate predictions in situations that require special attention, such as when the discrepancy between the predicted amount of power supply and the actual amount of power supply becomes large, or when the balance between the amount of power supply and the amount of power demand is likely to be disrupted. By doing so, it is possible to implement measures such as necessary demand control.

[0020] An electric power user may obtain renewable energy from their own power generation equipment, such as a solar power generation equipment 52, or may obtain it by purchasing it from another party (for example, another business). Furthermore, the electric power user's own power generation equipment may be located inside or around the facility of the supply target 50, or may be located away from the supply target 50. If the power generation equipment is located inside or around the facility of the supply target 50, the power generation equipment can supply power directly to the supply target 50. If the power generation equipment is located away from the supply target 50, the power generation equipment can supply power to the supply target 50 via a power grid 60 or the like. The power grid 60 is, for example, a transmission and distribution network managed by a power company or the like. The supply target 50 can also receive power from the power company via the power grid 60.

[0021] The supply target 50 is not particularly limited, and may be, for example, one or more facilities, one or more pieces of equipment, one or more buildings, one or more residences, one or more rooms, or one or more floors. The supply target 50 may also be, for example, a factory or production facility. Power is supplied to the supply target 50 from a power supply source 54. Power is also supplied to the supply target 50 from sources other than the power supply source 54, such as a power plant 62, via the grid power network 60. Furthermore, if the amount of power supplied from the power supply source 54 exceeds the power demand of the supply target 50, power is also supplied from the power supply source 54 to the grid power network 60. The power supply source 54 includes at least one solar power generation facility 52. ​​The power supply source 54 may also include one or more power supply facilities other than the solar power generation facility 52. ​​The power supply facility other than the solar power generation facility 52 may be a power generation facility using renewable energy or other power generation facilities. The power supply source 54 may also include power supply facilities related to so-called self-dispatch or an off-site power purchase agreement (PPA). The amount of power supplied by the power supply source 54 is referred to as a first power supply amount. The first power supply amount may indicate the amount of power supplied from the power supply source 54 to the supply target 50. A charging / discharging facility such as a storage battery may be further connected to the supply target 50 and the power supply source 54. Furthermore, the power supply source 54 may also supply power to other targets. In this case, for example, a proportion of the first power supply amount that can be used by the supply target 50 may be set, and this proportion may be used to determine the predetermined conditions described below. Power may be supplied from the power supply source 54 only to the supply target 50.

[0022] FIG. 3 is a block diagram illustrating the configuration of a prediction device 10 according to this embodiment. In the example of FIG. 3, the prediction device 10 further includes a preliminary prediction unit 120, an output unit 150, a performance value acquisition unit 160, a change information acquisition unit 170, an image acquisition unit 180, and a storage unit 140. The preliminary prediction unit 120 generates a transition of a predicted value of the first power supply amount in advance and stores the transition in the storage unit 140. In the example of FIG. 3, the storage unit 140 is included in the prediction device 10, but the storage unit 140 may be provided outside the prediction device 10. Note that the prediction device 10 may not include at least one of the performance value acquisition unit 160 and the change information acquisition unit 170 depending on the content of the predetermined condition to be determined.

[0023] The advance prediction unit 120 predicts the first power supply amount for a predetermined period based on the predetermined time every day at a predetermined time, and generates a transition of the predicted value. For example, the advance prediction unit 120 predicts the first power supply amount for the next day and the day after that, and generates a transition of the predicted value.

[0024] The advance prediction unit 120 includes, for example, a first trained model 122. The first trained model 122 is a trained model based on machine learning, and when input data is input, outputs a trend in the predicted value of the first power supply amount. The input data of the first trained model 122 includes, for example, at least one of weather forecast information, information about the environment surrounding the solar power generation facility 52, and information about the solar power generation facility 52. ​​The weather forecast information includes, for example, at least one of sunrise time, sunset time, temperature trend, solar radiation trend, and cloud cover trend. The weather forecast information includes at least information about the area where the solar power generation facility 52 is installed. The information about the environment of the solar power generation facility 52 is, for example, three-dimensional data of objects (buildings, trees, etc.) surrounding the solar power generation panels provided on the solar power generation facility 52. ​​The information about the solar power generation facility 52 includes at least one of the power generation efficiency of the solar power generation facility 52, the configuration of the solar power generation facility 52, the manufacturer and model number of the solar power generation panels provided on the solar power generation facility 52, and information indicating whether the solar power generation facility 52 is movable. Information about the environment surrounding the solar power generation facility 52 and information about the solar power generation facility 52 are stored in advance in the storage unit 140, and the advance prediction unit 120 can read and use this information. The advance prediction unit 120 can also obtain weather forecast information from, for example, a server of a weather information provider via a communication network. Note that the advance prediction unit 120 may not include the first trained model 122, and may instead predict the first power supply amount based on statistics obtained in advance or rule-based judgment.

[0025] When the power supply source 54 includes a plurality of photovoltaic power generation facilities 52, the predicted value of the first power supply amount is calculated using the predicted value of the power supply amount of each photovoltaic power generation facility 52. ​​When the power supply source 54 includes power supply facilities other than the photovoltaic power generation facility 52, the predicted value of the first power supply amount is calculated as the sum of the predicted values ​​of the power supply amount of each power supply facility. The predicted value of the power supply amount of power supply facilities other than the photovoltaic power generation facility 52 can be calculated using an appropriate existing method. Furthermore, for power supply facilities using energy other than renewable energy, the predicted value of the power supply amount may be a predetermined value.

[0026] The acquisition unit 110 acquires the trend of the predicted value of the first power supply amount. Specifically, the acquisition unit 110 may acquire the trend of the predicted value of the first power supply amount from the current time until a predetermined time (for example, 12 hours or 24 hours later), or may acquire all the trends of the predicted value of the first power supply amount that can be acquired. In the example of FIG. 3 , the acquisition unit 110 can read and acquire the trend of the predicted value of the first power supply amount from the storage unit 140. Alternatively, the acquisition unit 110 may acquire the trend of the predicted value of the first power supply amount directly from the advance prediction unit 120.

[0027] The advance prediction unit 120 also predicts the amount of power demand at the supply target 50 and generates a trend in the predicted value. At the timing of predicting the first power supply amount, the advance prediction unit 120 can predict the amount of power demand at the supply target 50 for the same period as the prediction target period of the first power supply amount. The advance prediction unit 120 stores the generated trend in the predicted value of the power demand in the storage unit 140. The advance prediction unit 120 further includes, for example, a second trained model 124. The second trained model 124 is a trained model based on machine learning, and outputs a trend in the predicted value of the power demand when input data is input. The input data for the second trained model 124 can include, for example, at least one of a temperature trend, information indicating whether the day to be predicted is a holiday, and information indicating the operating status of the supply target 50. The information indicating the operating status of the supply target 50 is, for example, information indicating the production volume of products, the number of people working, business hours, etc.

[0028] The acquisition unit 110 acquires a trend in the predicted value of the power demand at the supply target 50. Specifically, the acquisition unit 110 may acquire a trend in the predicted value of the first power demand from the current time until a predetermined time (for example, 12 hours or 24 hours later), or may acquire all trends in the predicted value of the power demand that can be acquired. In the example of FIG. 3 , the acquisition unit 110 can read and acquire the trend in the predicted value of the power demand from the storage unit 140. Alternatively, the acquisition unit 110 may acquire the trend in the predicted value of the power demand directly from the advance prediction unit 120.

[0029] The prediction device 10 does not need to include the advance prediction unit 120. In that case, the transition of the predicted value of the first power supply amount and the transition of the predicted value of the power demand amount may be generated by a device other than the prediction device 10. The transitions of the generated predicted values ​​are stored in a storage unit accessible by the acquisition unit 110, and the acquisition unit 110 can read and acquire these transitions. Alternatively, the acquisition unit 110 may acquire the transitions of the predicted values ​​directly from the device that generated them. The device that generates the transition of the predicted value of the first power supply amount and the device that generates the transition of the predicted value of the power demand amount may be the same device or may be separate devices.

[0030] FIG. 4 is a diagram showing an example of a graph generated by the output unit 150. FIG. 5 is a diagram showing an example of an image displayed based on the notification information of the output unit 150. FIG. 4 corresponds to an enlarged view of the graph displayed on the left side of FIG. 5. The output unit 150 outputs at least one of notification information and control information based on a comparison result between the predicted value of the first power supply amount acquired by the acquisition unit 110 and the predicted value of the power demand of the supply target 50. The control information is information for controlling at least one of the power supply amount and the power demand of the supply target 50. As will be described later, when the predicted value of the first power supply amount is corrected or re-predicted, the output unit 150 outputs at least one of notification information and control information based on a comparison result between the corrected or re-predicted predicted value of the first power supply amount and the predicted value of the power demand of the supply target 50.

[0031] The output unit 150 can generate notification information such as that shown in Fig. 5 using the transition of the predicted value of the first power supply amount and the transition of the predicted value of the power demand amount acquired by the acquisition unit 110. The notification information is, for example, image data for displaying an image on a display. The notification information is not limited to the example of Fig. 5, and may be audio information or a signal for flashing a light-emitting device.

[0032] The graph in Figure 4 shows trends in power supply and power usage for the supply target 50. In this graph, "photovoltaic power generation" represents the first power supply amount. Note that in the example of Figure 4, the power supply source 54 includes only one or more photovoltaic power generation facilities 52. "Maximum power" represents trends in the power available to the supply target 50, assuming that the maximum available power within the basic scope of the contract with the power company is used. In other words, if the power demand exceeds this line, additional charges will be incurred, making power more expensive. The maximum available power within the basic scope of the contract with the power company is a predetermined value. "Maximum power" is obtained, for example, by adding the maximum available power within the basic scope of the contract with the power company and the supply amount obtained by purchasing power from other suppliers to the first power supply amount. Furthermore, "renewable energy" represents the amount of power supplied from renewable energy to the supply target 50. Power supplied from sources other than the power supply source 54 may also include power derived from renewable energy. The sum of such a power supply amount derived from renewable energy and the power supply amount from the solar power generation facility 52 corresponds to "renewable energy." If the first power supply amount includes a power supply amount other than that derived from renewable energy, the amount of power derived from renewable energy can be calculated by subtracting the power supply amount other than that derived from renewable energy from the first power supply amount. Alternatively, the amount of power supply derived from renewable energy may be calculated by adding the amount of power supply from power generation facilities that generate electricity from renewable energy, such as the solar power generation facility 52, to the amount of power supplied to the supply target 50.

[0033] The power supplied from sources other than the power supply source 54 may include power generated by a power company, as well as power related to so-called self-dispatching or off-site PPA. As described above, power supply facilities related to self-dispatching or off-site PPA may be included in the power supply source 54. In other words, when the supply target 50 can receive power supply from multiple power supply facilities related to self-dispatching or off-site PPA, some of the multiple power supply facilities may be included in the power supply source 54, and the remaining some may not be included in the power supply source 54.

[0034] "Power usage" indicates the amount of power demand at the supply target 50. In Fig. 4, the dashed line indicates the current time. For each line in the graph, the left side of the dashed line indicates the actual value at the supply target 50, and the right side indicates the predicted value.

[0035] In addition to the graph, the left side of Fig. 5 displays the current power usage and the amount of power supplied from sources other than the power supply source 54 ("purchased power"). The right side of Fig. 5 also shows the difference between the renewable energy usage target and the amount of power demand. This visualization allows the user to understand the current situation and outlook for the balance between power supply and demand.

[0036] The right side of FIG. 5 also shows a period for which power saving is determined necessary as a result of comparing the predicted value of the first power supply amount with the predicted value of the power demand amount, labeled "Power Saving Required Time." For example, the output unit 150 identifies a period for which power saving is necessary when the predicted value of power demand exceeds the aforementioned upper power limit. A star mark is added to the graph on the left side of FIG. 5 at the position corresponding to that period. The right side of FIG. 5 also presents proposals for control to reduce the power demand amount. For example, these proposals include reducing the lighting level in some areas, turning off power for battery charging for a specified period, lowering the heating temperature settings in some areas, and stopping the operation of some of the equipment. The amount of power saved by these control (setting changes) is also displayed. The user selects (clicks or touches) the "OK" button displayed on this screen as an approval operation. Control information is then output from the output unit 150, and the control displayed on the screen is executed. If there is no period in which it has been determined that power saving is necessary, the output unit 150 may output notification information indicating a message such as "There is no problem with the balance." Note that the output unit 150 may automatically output the control information without receiving an approval operation.

[0037] The control information is not particularly limited, but may be, for example, information for controlling at least one of lighting, air conditioning, refrigeration equipment, freezing equipment, boilers, hot water supply equipment, sanitary equipment, compressors, wastewater treatment equipment, and water and sewerage equipment. The control information may also be information for controlling power receiving and transforming equipment, such as shutting off unnecessary transformers, or information for controlling equipment for recovering and utilizing exhaust heat.

[0038] The output unit 150 can determine the contents of the notification information and control information to be output based on the result of comparing the predicted value of the first power supply amount with the predicted value of the power demand amount. For example, the output unit 150 calculates the required reduction in power consumption by subtracting the predicted value of the power demand amount from the predicted value of the first power supply amount. Meanwhile, the storage unit 140 stores in advance reference information in which the reduction amount is associated with at least one of the notification information and the control information. The storage unit 140 identifies at least one of the notification information and the control information corresponding to the reduction amount calculated in the reference information as information to be output.

[0039] For example, when the acquisition unit 110 acquires the transition of the predicted value of the first power supply amount and the transition of the predicted value of the power demand amount, the output unit 150 uses the transition of the acquired predicted values ​​to identify a period in which power saving is necessary and further identifies at least one of the notification information and the control information to be output. Alternatively, the output unit 150 may identify a period in which power saving is necessary at a predetermined cycle or at a predetermined timing and identify at least one of the notification information and the control information to be output. Furthermore, the output unit 150 may determine a period in which power saving is necessary when the transition of the predicted value of the first power supply amount is corrected or re-predicted, and identify at least one of the notification information and the control information to be output. When the output unit 150 identifies at least one of the notification information and the control information to be output, it outputs the identified information.

[0040] <Predetermined Condition> In the prediction device 10 according to this embodiment, the prediction unit 130 corrects or re-predicts the transition of the predicted value of the first power supply amount using an image obtained by photographing the sky when a predetermined condition is satisfied. That is, the prediction unit 130 corrects or re-predicts the transition of the predicted value of the first power supply amount when the predetermined condition is satisfied. If the predetermined condition is not satisfied, the prediction unit 130 does not correct or re-predict the transition of the predicted value of the first power supply amount. Therefore, the processing load is reduced compared to when prediction is always performed using an image. Examples of the predetermined condition are described below. The predetermined condition includes multiple conditions, and when at least one of the multiple conditions is satisfied, the predetermined condition is determined to be satisfied. For example, when at least one of the multiple conditions described below is satisfied, the prediction unit 130 corrects or re-predicts the transition of the predicted value of the first power supply amount using an image. However, the predetermined condition is not limited to the following example and may be other conditions. The prediction unit 130 corrects or re-predicts the transition of the predicted value of the first power supply amount when at least the predetermined condition changes from not being satisfied to being satisfied. The prediction unit 130 may correct or re-predict the trend of the predicted value of the first power supply amount only when the state changes from a state where the specified condition is not satisfied to a state where it is satisfied, or may further correct or re-predict the trend of the predicted value of the first power supply amount while the specified condition is satisfied.

[0041] <<Condition Related to Predicted Value>> The predetermined condition may include a condition related to the predicted value of the first power supply amount. For example, the predetermined condition includes at least one of condition (1) and condition (2) described below.

[0042] Condition (1) is that the ratio of the predicted value of the power demand amount at the supply target to the predicted value of the first power supply amount satisfies a predetermined first standard. The prediction unit 130 calculates the value of "predicted value of the power demand amount at the supply target / predicted value of the first power supply amount" at each time using the transition of the predicted value of the first power supply amount and the transition of the predicted value of the power demand amount acquired by the acquisition unit 110. Then, the prediction unit 130 calculates the calculated value as a predetermined standard value S 1 The timing at which the reference value S1 The following values ​​are used to determine the reference value S 1 Identify the time when the value exceeds the reference value S. 1 is, for example, 0.8 or more and 1.2 or less. Condition (1) is satisfied at the specified time. At the specified time, the prediction unit 130 acquires an image of the sky and corrects or re-predicts the transition of the predicted value of the first power supply amount. A more accurate predicted value of the first power supply amount can be obtained prior to the timing when it is necessary to reduce the power demand amount. Consequently, appropriate measures can be taken.

[0043] Condition (2) is that the difference between the actual value and the predicted value of the first power supply amount satisfies a predetermined second standard. The prediction unit 130 compares the predicted value of the first power supply amount for the current time acquired by the acquisition unit 110 with the current actual value of the first power supply amount (actual value). The actual value acquisition unit 160 of the prediction device 10 can acquire the actual value of the first power supply amount of the power supply source 54 from the power supply source 54 at a predetermined cycle (for example, every minute). The comparison between the predicted value and the actual value of the first power supply amount is performed every time an actual value is acquired. The prediction unit 130 calculates the magnitude of the difference between the predicted value of the first power supply amount and the actual value of the first power supply amount. The prediction unit 130 calculates the magnitude of the calculated difference between the predicted value of the first power supply amount and the actual value of the first power supply amount. The prediction unit 130 compares the magnitude of the calculated difference with a predetermined standard value S 2 It is determined whether the magnitude of the calculated difference is equal to or greater than a predetermined reference value S 2 If the difference is equal to or greater than the predetermined reference value S 2 If it is not, it is determined that condition (2) is not satisfied. If it is determined that condition (2) is satisfied, the prediction unit 130 acquires an image of the sky and corrects or re-predicts the transition of the predicted value of the first power supply amount. In this way, when the accuracy of the predicted value becomes insufficient, a more accurate predicted value of the first power supply amount can be obtained.

[0044] In addition, the specified condition may include, as a condition regarding the predicted value of the first power supply amount, a condition that the timing is such that the ratio of the predicted value of the power demand amount at the supply target 50 to the predicted value of the above-mentioned upper limit power satisfies a specified first criterion.

[0045] <<Conditions Related to the Grid Power Network>> The predetermined conditions may include conditions related to at least one of the power supply from the grid power network 60 to the supply target 50 and the power supply from the power supply source 54 to the grid power network 60. For example, the predetermined conditions include at least one of the conditions (3) and (4).

[0046] Condition (3) is that information indicating that the second power supply amount, which is the amount of power supplied from the power grid 60 to the supply target 50, should be changed is received. The information indicating that the second power supply amount should be changed is transmitted, for example, from a power company and acquired by the change information acquisition unit 170 via a communication network. For example, when the power demand amount in the power grid 60 approaches or is expected to approach the supply upper limit, the power company transmits information requesting power users to reduce the amount of power supplied from the power grid 60 to the supply target 50, i.e., information indicating that the second power supply amount should be changed. An example of the information indicating that the second power supply amount should be changed is a so-called "downward demand response (DR)." When such information is received, it is necessary to suppress the power demand amount at the supply target 50. Note that the second power supply amount may refer to the amount of power supplied from sources other than the power supply source 54 or the amount of power supplied from the power company's power generation facility. When the change information acquisition unit 170 receives information indicating that the second power supply amount should be changed, the prediction unit 130 determines that the condition (3) is satisfied, acquires an image of the sky, and corrects or re-predicts the transition of the predicted value of the first power supply amount, thereby enabling appropriate power saving based on a more accurate predicted value of the first power supply amount.

[0047] Condition (4) is that information indicating that the third power supply amount, which is the amount of power supplied from the power supply source 54 to the grid power network 60, should be changed is received. The information indicating that the third power supply amount should be changed is transmitted, for example, from a power company and acquired by the change information acquisition unit 170 via a communication network. For example, when the amount of power supplied in the grid power network 60 significantly exceeds the amount of power demand, or is predicted to significantly exceed the amount of power demand, the power company transmits information requesting power users to reduce the amount of power supplied from the power supply source 54 to the grid power network 60, i.e., information indicating that the third power supply amount should be changed. An example of information indicating that the third power supply amount should be changed is a so-called "increased demand rate." Receiving such information indicates that the power demand at the supply target 50 needs to be increased. When the change information acquisition unit 170 receives information indicating that the third power supply amount should be changed, the prediction unit 130 determines that condition (4) is satisfied, acquires an image of the sky, and corrects or re-predicts the trend of the predicted value of the first power supply amount. This enables appropriate adjustment of power usage based on a more accurate predicted value of the first power supply amount.

[0048] <<Conditions Related to Renewable Energy>> The predetermined conditions may include conditions related to the supply of power from renewable energy to the supply target 50. For example, the predetermined conditions include condition (5).

[0049] Condition (5) is that the difference between the amount of power supplied from renewable energy to the supply target 50 and the amount of power demanded by the supply target 50, or the integrated value of this difference, satisfies the third criterion. It is preferable for power users to use as much of their power consumption as possible from renewable energy sources. The prediction unit 130 calculates the trend in the amount of power supplied from renewable energy sources among the amount of power supplied to the supply target 50. The amount of power supplied from renewable energy sources is as described in relation to FIG. 4 . It is preferable for the power demand of the supply target 50 to be as close as possible to the amount of power supplied from renewable energy sources. The amount of power supplied from renewable energy sources includes at least the amount of power supplied by the solar power generation facility 52 of the power supply source 54. The amount of power supplied from renewable energy sources may further include the amount of power supplied from other renewable energy power generation facilities included in the power supply source 54. Furthermore, the amount of power supplied from renewable energy sources may also include the amount of power supplied from renewable energy sources purchased from other businesses. The amount of power supply from renewable energy may also include the amount of power supply from solar power generation facilities 52 located away from the supply target 50. The prediction unit 130 can acquire and add up the trends in these amounts of power supply, thereby calculating the trend in the amount of power supply from renewable energy to the supply target 50.

[0050] The prediction unit 130 calculates the magnitude of the difference between the calculated amount of power supply by renewable energy and the amount of power demand in the supply target 50 each time these actual values ​​are acquired. Then, the prediction unit 130 calculates the magnitude of the calculated difference when the magnitude of the calculated difference is equal to or exceeds a predetermined reference value S 3 It is determined whether the magnitude of the calculated difference is equal to or greater than a predetermined reference value S 3 If the difference is equal to or greater than the predetermined reference value S 3 If the predicted value is equal to or greater than the predetermined value, an image of the sky is acquired, and the transition of the predicted value of the first power supply amount is corrected or re-predicted. By doing so, it is possible to appropriately adjust the power consumption based on the more accurate predicted value of the first power supply amount.

[0051] Alternatively, the prediction unit 130 calculates the difference between the amount of power supply from renewable energy and the amount of power demand in the supply target 50 by subtracting one from the other. Then, the prediction unit 130 further calculates an integrated value of the difference for a predetermined period ending at the current time. Then, the absolute value of the calculated integrated value is calculated to be equal to or smaller than a predetermined reference value S 4 It is determined whether the absolute value of the calculated integrated value is equal to or greater than a predetermined reference value S 4 If the absolute value of the calculated integrated value is equal to or greater than a predetermined reference value S 4 If the predicted value is equal to or greater than the predetermined value, an image of the sky is acquired, and the transition of the predicted value of the first power supply amount is corrected or re-predicted. By doing so, it is possible to appropriately adjust the power consumption based on the more accurate predicted value of the first power supply amount.

[0052] <Prediction> FIG. 6 is a diagram illustrating an example of the configuration of a prediction system 30 according to this embodiment. The prediction system 30 includes a prediction device 10 and a photographing unit 20 that captures an image of the sky around the solar power generation facility 52. ​​The photographing unit 20 is, for example, a camera. The photographing unit 20 may include a fisheye lens. The photographing direction of the photographing unit 20 may be variable. The photographing unit 20 is preferably installed near the solar power generation facility 52. ​​For example, the distance between the photographing unit 20 and the solar power generation panels of the solar power generation facility 52 is preferably within 100 m. The prediction device 10 is connected to the photographing unit 20 via a wired or wireless connection. When the prediction device 10 determines that a predetermined condition is satisfied, the photographing unit 20 captures an image of the sky around the solar power generation facility 52. ​​Specifically, when the prediction unit 130 determines that the predetermined condition is satisfied, the image acquisition unit 180 outputs a control signal to the photographing unit 20 to capture an image of the sky. When the photographing unit 20 receives the control signal from the image acquiring unit 180, it captures an image of the sky around the photovoltaic power generation facility 52 at that time. When the photographing unit 20 captures an image of the sky, the image acquiring unit 180 acquires the image from the photographing unit 20. Furthermore, using the image acquired by the image acquiring unit 180, the prediction unit 130 corrects or re-predicts the transition of the predicted value of the first power supply amount.

[0053] The capturing range of the sky image around the solar power generation facility 52 is not particularly limited, but for example, the sky image preferably includes the sky directly above the solar power generation facility 52. ​​Furthermore, the sky image preferably includes the sky in the direction of the sun. Furthermore, if the orientation of the image capturing unit 20 is variable, the image acquiring unit 180 may acquire wind information for the area in which the solar power generation facility 52 is located from a server of a weather information provider or the like via a communication network, and control the orientation of the image capturing unit 20 so as to capture an image of the sky in the upwind direction as viewed from the solar power generation facility 52.

[0054] The prediction unit 130 corrects or re-predicts the trend of the predicted value of the first power supply amount using an image of the sky around the solar power generation facility 52. ​​The prediction unit 130 may also correct or re-predict the trend of the predicted value of the first power supply amount using information on date and time or the position of the sun. When using date and time, the prediction unit 130 can identify the position (orientation) of the sun as seen from the solar power generation facility 52 using the date and time and known position information of the solar power generation facility 52. ​​The prediction unit 130 may obtain information indicating the position of the sun from a server of a weather information provider or the like via a communication network. Since the power generation efficiency of the solar power generation facility 52 may change depending on the direction of the sun and the influence of buildings and the like around the solar power generation facility 52 also changes, the prediction accuracy can be improved by using information on date and time or the position of the sun.

[0055] The prediction unit 130 may further use at least one of wind information and temperature to correct or re-predict the transition of the predicted value of the first power supply amount. The wind information includes, for example, information indicating the direction and strength of the wind. The wind information may be an actual measurement value or a predicted value. The movement of clouds blocking sunlight varies depending on the direction and strength of the wind. Furthermore, the power generation efficiency of the solar power generation facility 52 varies depending on the temperature. Therefore, by using at least one of wind information and temperature, the prediction accuracy can be improved.

[0056] Furthermore, when the image of the sky includes an image of the sky upwind from the solar power generation facility 52, the prediction unit 130 preferably corrects or re-predicts the transition of the predicted value of the first power supply amount using wind information. This makes it possible to make a prediction that takes into account changes in cloud conditions due to the influence of wind.

[0057] The prediction unit 130 includes, for example, a third trained model 132. The third trained model 132 is a trained model based on machine learning, and when input data is input, outputs a trend in the predicted value of the first power supply amount. The prediction unit 130 replaces the trend in the predicted value of the first power supply amount for a corresponding period acquired by the acquisition unit 110 with the trend in the predicted value of the first power supply amount output from the third trained model 132. The input data of the third trained model 132 includes at least an image obtained by photographing the sky around the solar power generation facility 52. ​​The input data to the third trained model 132 may further include at least one of date and time, solar position information, wind information, and temperature. The input data to the third trained model 132 may or may not further include at least a portion of the trend in the predicted value of the first power supply amount acquired by the acquisition unit 110. The third trained model 132 outputs the trend in the predicted value of the first power supply amount within a predetermined period based on the current time. The start point of the predetermined period may be the current time, or may be a time T s It may be later. s The end point of the predetermined period is, for example, 5 minutes or more and 45 minutes or less. e Later. T e For example, the predetermined period may be a period during which it is determined that demand suppression or the like is necessary, such as a period during which the difference between the power supply amount predicted by the first trained model and the power demand amount predicted by the second trained model exceeds a threshold value.

[0058] The output unit 150 generates and outputs at least one of notification information and control information using the updated predicted value of the first power supply amount, thereby enabling more appropriate notification and control of power usage.

[0059] The hardware configuration of the prediction device 10 will be described below. Each functional component of the prediction device 10 may be realized by hardware that realizes the functional component (e.g., a hardwired electronic circuit, etc.), or by a combination of hardware and software (e.g., a combination of an electronic circuit and a program that controls it). Below, a case where each functional component of the prediction device 10 is realized by a combination of hardware and software will be further described.

[0060] 7 is a diagram illustrating a computer 1000 for implementing the prediction device 10. The computer 1000 is any computer. For example, the computer 1000 is a system on chip (SoC), a personal computer (PC), a server machine, a tablet terminal, a smartphone, or the like. The computer 1000 may be a dedicated computer designed to implement the prediction device 10, or may be a general-purpose computer.

[0061] The computer 1000 includes a bus 1020, a processor 1040, a memory 1060, a storage device 1080, an input / output interface 1100, and a network interface 1120. The bus 1020 is a data transmission path through which the processor 1040, the memory 1060, the storage device 1080, the input / output interface 1100, and the network interface 1120 transmit and receive data to and from each other. However, the method of connecting the processor 1040 and other components to each other is not limited to bus connection. The processor 1040 may be any of various processors, such as a central processing unit (CPU), a graphics processing unit (GPU), or a field-programmable gate array (FPGA). The memory 1060 is a main storage device implemented using a random access memory (RAM) or the like. The storage device 1080 is an auxiliary storage device implemented using a hard disk, a solid state drive (SSD), a memory card, a read-only memory (ROM), or the like.

[0062] The input / output interface 1100 is an interface for connecting the computer 1000 to an input / output device. For example, the input / output interface 1100 is connected to an input device such as a keyboard and an output device such as a display.

[0063] The network interface 1120 is an interface for connecting the computer 1000 to a network. This communication network is, for example, a LAN (Local Area Network) or a WAN (Wide Area Network). The network interface 1120 may be connected to the network via a wireless connection or a wired connection.

[0064] The storage device 1080 stores program modules that realize each functional component of the prediction device 10. The processor 1040 reads each of these program modules into the memory 1060 and executes them to realize the function corresponding to each program module. Furthermore, when the storage unit 140 is provided inside the prediction device 10, the storage unit 140 is realized using the storage device 1080, for example.

[0065] FIG. 8 is a flowchart showing an outline of the prediction method executed by the prediction device 10 according to this embodiment. In the prediction method according to this embodiment, one or more computers perform an acquisition step S10 and a prediction step S20. In the acquisition step S10, a trend in the predicted value of the first power supply amount is acquired. The first power supply amount is the amount of power supplied to the supply target 50 by a power supply source 54 including at least one photovoltaic power generation facility 52. ​​In the prediction step S20, when a predetermined condition is satisfied, the trend in the predicted value of the first power supply amount is corrected or re-predicted. This correction or re-prediction is performed using an image obtained by photographing the sky around the photovoltaic power generation facility 52.

[0066] 9 is a flowchart illustrating the flow of the prediction method according to this embodiment. When the processing of the prediction device 10 starts, the acquisition unit 110 acquires the transition of the predicted value of the first power supply amount and the transition of the predicted value of the power demand amount (S110). If the prediction device 10 includes the advance prediction unit 120, in S110 the advance prediction unit 120 generates the transition of the predicted value of the first power supply amount and the transition of the predicted value of the power demand amount, and the acquisition unit 110 acquires them.

[0067] Next, in S120, the prediction unit 130 acquires the latest actual values ​​of the first power supply amount and the power demand amount. Then, in S130, the prediction unit 130 determines whether a predetermined condition is satisfied. If the predetermined condition is satisfied (Yes in S130), the prediction unit 130 updates the trend of the predicted value of the first power supply amount (S140). Next, in S150, the output unit 150 identifies and outputs information to be output. If the predetermined condition is not satisfied (No in S130), S140 is not performed and S150 is performed. After S150, it is determined whether a termination condition is satisfied (S160). If the termination condition is satisfied (Yes in S160), the processing of the prediction device 10 is terminated. An example of a case in which the termination condition is satisfied is when an operation to terminate the processing of the prediction device 10 is performed on the prediction device 10.

[0068] If the termination condition is not satisfied (No in S160), it is determined whether or not it is time for the acquisition unit 110 to acquire new trends in the predicted value of the first power supply amount and the predicted value of the power demand amount (S170). For example, when a predetermined time has elapsed since the acquisition unit 110 last acquired these trends, or when a predetermined time has arrived, it is determined that it is time for the acquisition unit 110 to acquire new trends in the predicted value of the first power supply amount and the predicted value of the power demand amount. If it is determined that it is time for the acquisition unit 110 to acquire new trends in the predicted value of the first power supply amount and the predicted value of the power demand amount (Yes in S170), the processing returns to S110. If it is not determined that it is time for the acquisition unit 110 to acquire new trends in the predicted value of the first power supply amount and the predicted value of the power demand amount (No in S170), the processing returns to S120.

[0069] As described above, according to this embodiment, when a predetermined condition is satisfied, the prediction unit 130 corrects or re-predicts the transition of the predicted value of the first power supply amount using an image of the sky. Therefore, highly accurate predictions based on an image of the sky are made only when necessary, thereby reducing the processing load.

[0070] Although the embodiments of the present invention have been described above with reference to the drawings, these are merely examples of the present invention, and various other configurations can also be adopted.

[0071] In addition, although the flowcharts used in the above description show multiple steps (processes) in a sequential order, the order of steps executed in each embodiment is not limited to the order shown. In each embodiment, the order of the steps shown in the drawings can be changed as long as it does not cause any problems in terms of content. Furthermore, the above-described embodiments can be combined as long as the content is not contradictory.

[0072] Some or all of the above embodiments can be described as, but are not limited to, the following supplementary notes. 1-1. A prediction device comprising: an acquisition means for acquiring a trend in a predicted value of a first power supply amount to a supply target from a power supply source including at least one photovoltaic power generation facility; and a prediction means for correcting or re-predicting the trend in the predicted value of the first power supply amount using an image obtained by photographing the sky around the photovoltaic power generation facility when a predetermined condition is met. 1-2. In the prediction device described in 1-1., the predetermined condition includes at least one of condition (1) and condition (2), where condition (1) is that the ratio of the predicted value of the power demand amount at the supply target to the predicted value of the first power supply amount satisfies a predetermined first criterion, and condition (2) is that the difference between the actual value and the predicted value of the first power supply amount satisfies a predetermined second criterion. 1-3. 1-1. or 1-2. 1-4. The prediction device described in 1-3., wherein the predetermined condition includes a condition related to at least one of power supply from the grid power network to the supply target and power supply from the power supply source to the grid power network. 1-4. The prediction device described in 1-3., wherein the predetermined condition includes at least one of condition (3) and condition (4), wherein condition (3) is receiving information indicating that a second power supply amount, which is the amount of power supply from the grid power network to the supply target, should be changed, and condition (4) is receiving information indicating that a third power supply amount, which is the amount of power supply from the power supply source to the grid power network, should be changed. 1-5. The prediction device described in any one of 1-1. to 1-4., wherein the predetermined condition includes a condition related to power supply from renewable energy to the supply target. 1-6. 1-5. 1-7. The prediction device described in any one of 1-1. to 1-6., wherein the prediction means further uses information on date and time or the position of the sun to correct or re-predict a transition in the predicted value of the first power supply amount.1-8. The prediction device described in any one of 1-1. to 1-7., wherein the prediction means further uses at least one of wind information and air temperature to correct or re-predict the transition of the predicted value of the first power supply amount. 1-9. The prediction device described in any one of 1-1. to 1-8., wherein the prediction means uses wind information to correct or re-predict the transition of the predicted value of the first power supply amount, and wherein the image includes an image of the sky upwind from the solar power generation facility. 1-10. The prediction device described in any one of 1-1. to 1-9., further comprising output means for outputting at least one of notification information and control information based on a comparison result between the corrected or re-predicted predicted value of the first power supply amount and the predicted value of power demand at the supply target, and the control information is information for controlling at least one of the power supply amount and the power demand to the supply target. 2-1. 1-1. to 1-10. and an imaging means for imaging the sky around the solar power generation facility when the prediction device determines that the predetermined condition is satisfied.

[0073] 3-1. A prediction method in which one or more computers acquire a trend in a predicted value of a first power supply amount to a supply target from a power supply source including at least one photovoltaic power generation facility, and when a predetermined condition is met, correct or re-predict the trend in the predicted value of the first power supply amount using an image obtained by photographing the sky around the photovoltaic power generation facility. 3-2. A prediction method according to 3-1., wherein the predetermined condition includes at least one of condition (1) and condition (2), wherein condition (1) is that a ratio of a predicted value of power demand at the supply target to the predicted value of the first power supply amount satisfies a predetermined first criterion, and condition (2) is that a difference between an actual value and a predicted value of the first power supply amount satisfies a predetermined second criterion. 3-3. A prediction method according to 3-1. or 3-2., wherein the predetermined condition includes a condition related to at least one of power supply from a grid power network to the supply target and power supply from the power supply source to the grid power network. 3-4. A prediction method according to 3-3. 3-5. A prediction method according to any one of 3-1. to 3-4., wherein the predetermined condition includes a condition related to the supply of power from renewable energy to the supply target. 3-6. A prediction method according to 3-5., wherein the predetermined condition includes a condition related to the supply of power from renewable energy to the supply target. 3-7. A prediction method according to 3-1. to 3-6., wherein the predetermined condition includes a condition related to the difference between the amount of power supplied from renewable energy to the supply target and the amount of power demand at the supply target, or an integrated value of the difference, satisfies a third criterion. 3-8. A prediction method according to 3-1. to 3-6., wherein the predetermined condition includes a condition related to the supply of power from renewable energy to the supply target and the amount of power demand at the supply target. wherein the one or more computers further use information on date and time or the position of the sun to correct or re-predict the transition of the predicted value of the first power supply amount.3-8. The prediction method according to any one of 3-1. to 3-7., wherein the one or more computers further use at least one of wind information and air temperature to correct or re-predict a transition in the predicted value of the first power supply amount. 3-9. The prediction method according to any one of 3-1. to 3-8., wherein the one or more computers use wind information to correct or re-predict a transition in the predicted value of the first power supply amount, and wherein the image includes an image of the sky upwind from the solar power generation facility. 3-10. The prediction method according to any one of 3-1. to 3-9., wherein the one or more computers further output at least one of notification information and control information based on a comparison result between the corrected or re-predicted predicted value of the first power supply amount and a predicted value of power demand at the supply target, and wherein the control information is information for controlling at least one of the power supply amount and the power demand to the supply target.

[0074] 4-1. A recording medium storing a program for causing a computer to execute a prediction method, the prediction method being a prediction method in which the computer: acquires a trend in a predicted value of a first power supply amount to a supply target from a power supply source including at least one photovoltaic power generation facility; and, when a predetermined condition is met, corrects or re-predicts the trend in the predicted value of the first power supply amount using an image obtained by photographing the sky around the photovoltaic power generation facility. 4-2. A recording medium described in 4-1., the predetermined condition includes at least one of condition (1) and condition (2), condition (1) being that the ratio of the predicted value of the power demand at the supply target to the predicted value of the first power supply amount satisfies a predetermined first criterion, and condition (2) being that the difference between the actual value and the predicted value of the first power supply amount satisfies a predetermined second criterion. 4-3. A recording medium according to 4-1. or 4-2. 4-4. A recording medium described in 4-3., wherein the predetermined condition includes a condition related to at least one of power supply from the grid power network to the supply target and power supply from the power supply source to the grid power network. 4-4. A recording medium described in 4-3., wherein the predetermined condition includes at least one of condition (3) and condition (4), wherein condition (3) is receiving information indicating that a second power supply amount, which is the amount of power supply from the grid power network to the supply target, should be changed, and condition (4) is receiving information indicating that a third power supply amount, which is the amount of power supply from the power supply source to the grid power network, should be changed. 4-5. A recording medium described in any one of 4-1. to 4-4., wherein the predetermined condition includes a condition related to power supply from renewable energy to the supply target. 4-6. 4-5. The recording medium described in the above, wherein the predetermined condition includes a condition (5), and the condition (5) is that the difference between the amount of electricity supplied to the supply target from renewable energy and the amount of electricity demanded by the supply target, or the integrated value of the difference, satisfies a third criterion.4-7. A recording medium described in any one of 4-1. to 4-6., wherein in the prediction method, the computer further uses information on date and time or the position of the sun to correct or re-predict the transition of the predicted value of the first power supply amount. 4-8. A recording medium described in any one of 4-1. to 4-7., wherein in the prediction method, the computer further uses at least one of wind information and air temperature to correct or re-predict the transition of the predicted value of the first power supply amount. 4-9. A recording medium described in any one of 4-1. to 4-8., wherein in the prediction method, the computer uses wind information to correct or re-predict the transition of the predicted value of the first power supply amount, and the image includes an image of the sky upwind from the solar power generation facility. 4-10. A recording medium described in any one of 4-1. to 4-9. wherein in the prediction method, the computer further outputs at least one of notification information and control information based on a comparison result between the corrected or re-predicted predicted value of the first power supply amount and the predicted value of the power demand amount at the supply target, and the control information is information for controlling at least one of the power supply amount and the power demand amount to the supply target.

[0075] 5-1. A program causing a computer to execute a prediction method, the prediction method comprising: the computer acquiring a trend in a predicted value of a first power supply amount to a supply target from a power supply source including at least one solar power generation facility; and, when a predetermined condition is met, correcting or re-predicting the trend in the predicted value of the first power supply amount using an image obtained by photographing the sky around the solar power generation facility. 5-2. A program in accordance with 5-1., wherein the predetermined condition includes at least one of condition (1) and condition (2), wherein condition (1) is that a ratio of a predicted value of power demand at the supply target to the predicted value of the first power supply amount satisfies a predetermined first criterion, and condition (2) is that a difference between an actual value and a predicted value of the first power supply amount satisfies a predetermined second criterion. 5-3. A program in accordance with 5-1. or 5-2., wherein the predetermined condition includes a condition related to at least one of power supply from a grid power network to the supply target and power supply from the power supply source to the grid power network. 5-4. A program according to 5-3., wherein the predetermined condition includes at least one of condition (3) and condition (4), wherein condition (3) is that information indicating that a second power supply amount, which is the amount of power supplied from the grid power network to the supply target, should be changed, has been received, and condition (4) is that information indicating that a third power supply amount, which is the amount of power supplied from the power supply source to the grid power network, should be changed, has been received. 5-5. A program according to any one of 5-1. to 5-4., wherein the predetermined condition includes a condition related to the supply of power from renewable energy to the supply target. 5-6. A program according to 5-5., wherein the predetermined condition includes condition (5), wherein condition (5) is that the difference between the amount of power supplied from renewable energy to the supply target and the amount of power demand at the supply target, or an integrated value of the difference, satisfies a third criterion.5-7. The program described in any one of 5-1. to 5-6., wherein in the prediction method, the computer further uses information on date and time or the position of the sun to correct or re-predict the transition of the predicted value of the first power supply amount. 5-8. The program described in any one of 5-1. to 5-7., wherein in the prediction method, the computer further uses at least one of wind information and air temperature to correct or re-predict the transition of the predicted value of the first power supply amount. 5-9. The program described in any one of 5-1. to 5-8., wherein in the prediction method, the computer uses wind information to correct or re-predict the transition of the predicted value of the first power supply amount, and the image includes an image of the sky upwind from the solar power generation facility. 5-10. The program described in any one of 5-1. to 5-9. In the program described in any one of the above, in the prediction method, the computer further outputs at least one of notification information and control information based on a comparison result between the corrected or re-predicted predicted value of the first power supply amount and the predicted value of the power demand amount at the supply target, and the control information is information for controlling at least one of the power supply amount and the power demand amount to the supply target.

[0076] REFERENCE SIGNS LIST 10 Prediction device 20 Photography unit 30 Prediction system 50 Supply target 52 Photovoltaic power generation facility 54 Power supply source 60 Power grid 62 Power plant 110 Acquisition unit 120 Advance prediction unit 122 First trained model 124 Second trained model 130 Prediction unit 132 Third trained model 140 Memory unit 150 Output unit 160 Actual value acquisition unit 170 Change information acquisition unit 180 Image acquisition unit 1000 Computer

Claims

1. An acquisition means for acquiring a transition of a predicted value of a first power supply amount to a supply target of a power supply source including at least one photovoltaic power generation facility; and a prediction means for correcting or re-predicting a transition of the predicted value of the first power supply amount by using an image obtained by photographing the sky around the photovoltaic power generation facility when a predetermined condition is satisfied. Prediction device.

2. 2. The prediction device according to claim 1, The predetermined condition includes at least one of condition (1) and condition (2), The condition (1) is that a ratio of a predicted value of the power demand amount in the supply target to a predicted value of the first power supply amount satisfies a predetermined first criterion; The condition (2) is that the difference between the actual value and the predicted value of the first power supply amount satisfies a predetermined second criterion. Prediction device.

3. 3. The prediction device according to claim 1, The predetermined condition includes a condition related to at least one of power supply from the power grid to the supply target and power supply from the power supply source to the power grid. Prediction device.

4. 4. The prediction device according to claim 3, The predetermined conditions include at least one of conditions (3) and (4), The condition (3) is that information indicating that a second power supply amount, which is the amount of power supply from the grid power network to the supply target, should be changed has been received; The fourth condition is that information indicating that a third power supply amount, which is the amount of power supplied from the power supply source to the power grid, should be changed is received. Prediction device.

5. The prediction device according to claim 1 , The predetermined conditions include conditions regarding the supply of electricity from renewable energy to the supply target. Prediction device.

6. 6. The prediction device according to claim 5, The predetermined conditions include condition (5), The condition (5) is that the difference between the amount of electricity supplied to the supply target from renewable energy and the amount of electricity demanded by the supply target, or the integrated value of the difference, satisfies a third criterion. Prediction device.

7. The prediction device according to claim 1 , The prediction means further uses information on date and time or the position of the sun to correct or re-predict the transition of the predicted value of the first power supply amount. Prediction device.

8. The prediction device according to claim 1 , The prediction means further uses at least one of wind information and air temperature to correct or re-predict a transition of the predicted value of the first power supply amount. Prediction device.

9. The prediction device according to any one of claims 1 to 8, The prediction means uses wind information to correct or re-predict a transition of the predicted value of the first power supply amount, The image includes an image of the sky upwind from the solar power generation facility. Prediction device.

10. The prediction device according to claim 1 , an output unit that outputs at least one of notification information and control information based on a comparison result between the corrected or re-predicted predicted value of the first power supply amount and a predicted value of the power demand amount in the power supply target; The control information is information for controlling at least one of the amount of power supply to the supply target and the amount of power demand. Prediction device.

11. A prediction device according to any one of claims 1 to 10; and an imaging means for imaging the sky around the photovoltaic power generation facility when it is determined in the prediction device that the predetermined condition is satisfied. Prediction system.

12. One or more computers Obtaining a transition of a predicted value of a first power supply amount to a supply target of a power supply source including at least one solar power generation facility; When a predetermined condition is satisfied, a transition of the predicted value of the first power supply amount is corrected or re-predicted using an image obtained by photographing the sky around the photovoltaic power generation facility. Forecasting methods.

13. A program for causing a computer to execute a prediction method, The prediction method includes the steps of: Obtaining a transition of a predicted value of a first power supply amount to a supply target of a power supply source including at least one solar power generation facility; When a predetermined condition is satisfied, a prediction method corrects or re-predicts the transition of the predicted value of the first power supply amount using an image obtained by photographing the sky around the photovoltaic power generation facility. program.