WO2019021438A1 - Solar power generation amount prediction device, solar power generation amount prediction system, prediction method, and program - Google Patents

Abstract

A power generation amount record acquisition unit (4) acquires, from a power amount measurement device (3), a power generation amount record for the amount of power generated by a solar power generation system (2). An extraterrestrial solar radiation amount calculation unit (7) calculates the extraterrestrial solar radiation amount on a designated day and in a designated time period at the location at which the solar power generation system (2) is installed. A weather information acquisition unit (10) acquires weather forecast information for each time period of a day for which a prediction is to be made. A power generation amount record classification unit (8) classifies power generation amount coefficients, obtained by dividing a power generation amount by an extraterrestrial solar radiation amount, into a given number of power generation amount record groups associated with weather attributes. A predicted power generation amount calculation unit (11) calculates a predicted power generation amount on the basis of the extraterrestrial solar radiation amount in each time period for which a prediction is to be made, and the power generation amount record group associated with a weather attribute corresponding to the weather forecast information.

Description

Photovoltaic power generation prediction device, photovoltaic power generation prediction system, prediction method and program

The present invention relates to a photovoltaic power generation amount prediction device that predicts the power generation amount of a solar power generation system, a solar power generation amount prediction system, a prediction method, and a program.

The electricity generated by the solar power generation system can be stored as energy by storing the electricity in a storage battery, storing heat using a heat pump water heater, or the like. The stored energy can be used at a different timing from power generation. For example, by storing hot water in the heat pump water heater using electricity generated by the solar power generation system during the day, it is possible to prepare for the hot water supply demand in the evening. In order to store heat just enough for hot water supply demand, it is necessary to accurately predict the power generation amount of the photovoltaic power generation system.

In addition, the electricity generated by the solar power generation system can be sold to the distribution system. However, since the photovoltaic power generation system can not control the amount of power generation and the power generation time, when a large amount of photovoltaic power generation systems are introduced, it becomes difficult to manage the distribution voltage. For this reason, power companies are considering reducing the output of photovoltaic power generation systems. The power company needs to accurately predict the power generation amount of the solar power generation system in order to determine whether the power reduction should be performed or to determine the suppression amount in the case of the suppression.

Patent Document 1 defines the ratio of the actual amount of solar radiation in one day to the maximum value of the theoretical amount of solar radiation as the degree of clearness of the day, and based on the past weather information and the degree of past clearance, A solar radiation amount prediction method is disclosed that predicts the degree of clearness from the meteorological forecast data of and the solar radiation amount. Patent Document 1 also discloses a solar power generation output prediction method and a solar power generation output prediction system that predicts a power generation output generated by solar power generation based on the amount of solar radiation predicted by this method.

JP 2013-164286 A

In the photovoltaic power generation output prediction system disclosed in Patent Document 1, in order to predict the degree of clearness from the weather forecast data, it is necessary to accumulate past weather information and the degree of past clearness. Therefore, the amount of data to be stored increases as the accumulation period is extended to improve the prediction accuracy.

In view of the above circumstances, it is an object of the present invention to provide a solar power generation amount prediction apparatus, a solar power generation amount prediction system, and a prediction method capable of predicting the power generation amount of a solar power generation system without accumulating past weather data. To provide the program.

The solar power generation amount predicting apparatus according to the present invention comprises: a power generation amount actual acquisition means, a power generation amount actual storage means, an extrasolar solar radiation amount calculation means, a weather forecast acquisition means, a power generation actual result classification means, a predicted power generation amount Calculating means. The power generation amount performance acquiring means acquires the power generation amount of the photovoltaic power generation system as a power generation amount performance together with an acquisition date and an acquisition time zone. The power generation result storage means stores the power generation results acquired by the power generation result acquisition means. The extra-atmospheric solar radiation amount calculation means calculates the extra-atmospheric solar radiation amount at the installation place of the solar power generation system. The weather forecast acquiring means acquires weather forecast information including the installation location of the solar power generation system in a target area. The generated power result classification means causes the extracorporeal solar radiation amount calculation means to calculate the extracorporeal solar radiation amount in the acquisition date and the acquisition time zone included in the generated power amount results accumulated by the generated power amount result storage means, and is included in the generated power amount results The power generation coefficient calculated based on the power generation amount and the calculated extrasolar radiation amount is classified into a power generation amount performance group associated with the meteorological attribute. The forecasted power generation amount calculation means, the power generation actual result group corresponding to the weather attribute corresponding to the obtained weather forecast information and weather forecast information including the day and the time zone to be predicted acquired by the weather forecast acquiring means The predicted power generation amount is calculated on the basis of the extrasolar radiation amount in the day and the time zone to be predicted, which is calculated by the extrasolar radiation amount calculation means.

According to the present invention, by grouping power generation coefficients into groups and associating them with weather attributes, it is possible to predict the power generation amount of the solar power generation system without accumulating past weather data.

Block diagram showing a functional configuration of the solar power generation amount predicting apparatus according to the first embodiment A diagram showing an example of a power generation amount result stored in the power generation amount result storage unit in the first embodiment Flow chart showing an example of processing of the power generation performance result classification unit according to the first embodiment Flowchart showing an example of processing of the predicted power generation amount calculation unit in the first embodiment The figure which shows an example of the limitation method of the period in the process which acquires a power generation amount coefficient from the power generation amount performance group of a predicted power generation amount calculation part in Embodiment 1. The figure which shows an example of the process of regression analysis of the amount of external solar radiation and the actual amount of power generation in Embodiment 1 The block diagram which shows the structural example which shares a solar power generation amount prediction apparatus with a server and a terminal based on Embodiment 1 The block diagram which shows the structural example which shares a solar power generation amount prediction apparatus with a server and a terminal based on Embodiment 1 Block diagram showing a functional configuration of a solar power generation amount predicting apparatus according to a second embodiment A diagram showing an example of a power generation amount result stored in the power generation amount result storage unit in the second embodiment A diagram showing an example of predicted power generation amount accumulated in the predicted power generation amount storage unit in the second embodiment Flow chart showing an example of processing of the extraction unit in the second embodiment The figure which shows an example of the processing which extracts an exclusion candidate from a prediction deviation list in a 2nd embodiment. The figure which shows the example of hardware constitutions of the solar power generation amount prediction apparatus which concerns on Embodiment 1.

Hereinafter, a solar power generation amount predicting apparatus according to a mode for carrying out the present invention will be described with reference to the drawings.

Embodiment 1
The functional configuration of the solar power generation amount predicting apparatus 1 will be described with reference to FIG. The solar power generation amount forecasting device 1 is a power generation amount actual acquisition unit 4 that acquires the power generation amount actual of the solar power generation system 2 from the power amount measurement device 3, a power generation actual result storage unit 5 that accumulates the power generation amount actual, For the amount of extrasolar radiation calculation unit 7 that calculates the amount of extrasolar radiation from the position information source 6 indicating the installation location of the photovoltaic system 2, and the amount of power generation and the amount of extra solar radiation accumulated in the power generation result storage unit 5 Based on the power generation amount coefficient to be described later is calculated, the power generation amount performance classification unit 8 that classifies the calculated power generation amount coefficient into a power generation amount result group associated with the meteorological attribute, and acquires weather forecast information from the weather information source 9 The power generation result group corresponding to the weather attribute corresponding to the weather forecast information and the weather forecast acquisition unit 10 to be identified is specified, and the predicted power generation is performed based on the power generation coefficient and the extrasolar radiation amount belonging to the power generation result group Predicted power generation to calculate quantity It includes a calculating unit 11, a.

The power amount measuring device 3 is a device that measures the amount of power generated by the photovoltaic power generation system 2 during a fixed time zone, and is, for example, a smart meter. The constant time zone is, for example, a time zone divided at regular time intervals such as 30 minutes, 1 hour, etc., starting from midnight. Note that a plurality of solar power generation systems 2 and corresponding power amount measurement devices 3 may exist for one solar power generation amount prediction device 1.

The position information source 6 is an information source indicating a place where the solar power generation system 2 is installed, and includes information represented by latitude and longitude. The position information source 6 is, for example, a GPS (Global Positioning System) terminal provided in the photovoltaic power generation system 2. In addition, a storage device that stores the longitude and latitude obtained from the address of the installation place of the solar power generation system 2 may be used as the position information source 6. Further, the position information source 6 may not be an information source that accurately indicates the installation location of the photovoltaic power generation system 2. For example, a storage device that stores the latitude and longitude of a point representing a city or a town where the solar power generation system 2 is installed may be used as the position information source 6. Further, the position information source 6 may be a storage device provided in the solar power generation amount predicting apparatus 1 and storing information indicating a place where the solar power generation system 2 is installed.

The weather information source 9 is a source providing weather forecast information in the area where the solar power generation system 2 is installed, and is, for example, a weather information providing server of the Japan Meteorological Agency. In addition, the weather information source 9 may be an information source provided by a private weather information provision service, or may be a device for a weather forecaster to input weather data.

The power generation result acquisition unit 4 acquires information indicating the power generation amount in each time zone of the measurement date from the power amount measuring device 3 as a power generation result, and stores the acquired power generation result in the power generation result storage unit 5. The power generation amount performance acquiring unit 4 acquires the power generation amount performance every predetermined time such as 0 minutes and 30 minutes every hour. Further, when there are a plurality of solar power generation systems 2 and corresponding power amount measuring devices 3, the power generation performance is acquired for each of the solar power generation systems 2.

The power generation result storage unit 5 stores the power generation result acquired by the power generation result acquisition unit 4 in the power generation result table for each measurement date and for each measurement time zone. An example of the power generation amount result stored in the power generation amount result storage unit 5 will be described with reference to FIG. The power generation result table includes a date field 21, a time zone field 22, a solar power generation system ID field 23 and a power generation field 24. In this example, it is assumed that the time zone is divided every 30 minutes.

The date field 21 stores information indicating the date when the power generation amount is obtained. The information indicating the date is, for example, a character string by concatenating numbers indicating the year, month, and day. For example, when the measurement date is July 1, 2017, the character string stored in the date field is "20170701".

The time zone field 22 stores information indicating the time zone in which the amount of power generation has been measured. The information indicating the time zone is, for example, a character string by connecting numbers indicating the hour and minute of the end of the time zone. For example, if the measured time zone is more than 7 o'clock and 7:30 in the 24-hour notation, the character string stored in the time zone field is "0730". The method of expressing the time zone is not limited to this. Instead of the end, the time zone may be expressed by the start. In addition, "0" for more than 0:00 to 0:30, "1" for more than 0:30 to 1:00, ..., more than 23:30 to 24:00 The time zone may be represented by a numerical value representing the order of "47" and so on. Any character string may be used as long as it can uniquely identify each time zone.

The photovoltaic power generation system ID is stored in the photovoltaic power generation system ID field 23. The solar power generation system ID is represented by a numerical value, a character string, or the like, and is unique identification information provided for each solar power generation system 2. The solar power generation system ID may be sequentially numbered with respect to the solar power generation system 2 to be predicted by the solar power generation amount prediction apparatus 1 or may be a contract number of the power company. For example, when there are three solar power generation systems to be predicted by the solar power generation amount prediction apparatus, the solar power generation system ID corresponding thereto can be “0001”, “0002”, and “0003”. Further, in the case where power generation results are accumulated only for one solar power generation system 2, the solar power generation system ID field 23 may be omitted.

The power generation amount field 24 stores information indicating the power generation amount in the time zone in which the power generation amount is measured. When the time to measure the amount of power deviates, for example, when the amount of power for 31 minutes is measured, the power generation amount corrected in 30 minutes is generated by performing proportional distribution with the power generation amount measured in the preceding and subsequent time zones. Store in the quantity field 24. Also, for example, when the power amount measuring device 3 can measure the power generation amount only in a one-hour cycle, halve each of the measured power generation amounts as the power generation amount of each 30 minutes, the power generation amount of the time zone before and after The power generation amount obtained by, for example, calculating the power generation amount for 30 minutes by a data interpolation method such as Lagrange interpolation, or the like is stored in the power generation amount field 24.

In order to reduce the data storage capacity by deleting the oldest power generation record when the power generation record is newly accumulated after the power generation record accumulated in the power generation record storage 5 reaches a certain number. May be

The extrasolar radiation calculation unit 7 calculates the month and day and the time zone to be calculated based on the latitude and longitude indicated by the position information source 6 and the date and time period for which the extrasolar radiation is calculated. In the above, the amount of extrasolar radiation at the place where the solar power generation system 2 is installed is calculated. The amount of extrasolar radiation is the energy per unit area received by a plane perpendicular to sunlight outside the atmosphere approximately 8 km above the measurement point. Since the amount of extrasolar radiation is a value outside the atmosphere, it does not depend on the weather, but depends only on the position and the date and time. In response to the acquisition request from the power generation actual result classification unit 8 or the predicted power generation calculation unit 11 described later, the outside air radiation calculation unit 7 outputs the calculated outside air radiation to these. The amount of extrasolar radiation is calculated by the calculation procedure described below.

The latitude φ ₀ and the longitude λ ₀ of the installation location of the photovoltaic power generation system 2 indicated by the position information source 6 are both expressed by the frequency method. The latitude is expressed as positive on the north and negative on the south, and as positive on the longitude as east, and negative on the west. First, the latitude φ ₀ and the longitude λ ₀ expressed by the frequency method are converted to φ and λ expressed in radians.
φ [rad] = φ ₀ [degree] × π / 180
λ [rad] = λ ₀ [degree] × π / 180

Next, for the date to be calculated, the number of days elapsed from January 1 to the date to be calculated is DN. The intermediate variable θ ₀ can be introduced to obtain the sun declination δ, the earth-centered sun distance r and the equalization time E _q on the day to be calculated.
θ ₀ [rad] = 2π (DN [day] -1) / 365
δ [rad] = 0.006918-0.399912cos (θ 0) + 0.070257sin (θ 0) -0.006758cos (2θ 0) + 0.000907sin (2θ 0) -0.002697 (3θ 0) + 0.001480sin ( 3θ ₀ )
r [astronomical unit] = 1 / {1.000110 + 0.034221 cos (θ ₀ ) + 0.001280 sin (θ ₀ ) + 0.000719 cos (2θ ₀ ) + 0.000077 sin (2θ ₀ )} ^0.5
E _q [rad] = 0.000075 + 0.001868 cos (θ ₀ ) −0.032077 sin (θ ₀ ) −0.014615 cos (2θ ₀ ) −0.040849 sin (2θ ₀ )

Next, the time representative of the time zone to be calculated is determined. As the time representing the time zone, the beginning, the end, or an intermediate time between the beginning and the end of the time zone can be adopted. The time representing the time zone is assumed to be HH hours and MM minutes. The notation of time is Japan Standard Time. An intermediate variable JST can be introduced to obtain the solar time angle h at HH, MM minutes.
JST [hour] = HH [hour] + MM [minute] / 60
h [rad] = (JST-12) π / 12 + (λ-135π / 180) + E _q

Since the latitude φ, the sun declination δ, and the time angle h are determined, the sun elevation angle α at the date and time to be calculated can be determined.
α [rad] = arc sin {sin (φ) sin (δ) + cos (φ) cos (δ) cos (h)}

Then, from the geocentric sun distance r and the elevation angle α, it is possible to calculate the amount of extraterrestrial solar radiation Q at the date and time to be calculated.
Q [W / m ² ] = 1367 [W / m ² ] × (1 / r) ² × sin (α)

The extrasolar radiation calculation unit 7 may calculate and output the extrasolar radiation at every request from the actual power generation classification unit 8 or the predicted power generation calculation unit 11, or the extrasolar radiation may be output in advance. It is good also as what calculates and accumulates, and acquires and outputs the amount of extraterrestrial solar radiation accumulated when requested. In the latter case, a memory is provided in the extrasolar radiation calculation unit 7, and the extraneous radiation in each time zone is calculated for one year in advance when the position information source 6 is updated, and the calculated extraneous radiation is calculated. It can be realized by storing everything in memory. Further, the amount of extrasolar radiation may be acquired by connecting to an external device that calculates the amount of extra solar radiation according to the date and time and the input of position information. That is, the calculation of the amount of extrasolar radiation includes not only the sequential calculation of the amount of extra solar radiation but also the acquisition of the amount of extra solar radiation by some means.

When the date to be calculated is February 29, it may be treated as February 28 or March 1. Further, the average of the amount of extrasolar radiation on February 28 and the amount of extrasolar radiation on March 1 may be taken as the amount of extrasolar radiation on February 29.

The power generation actual result classification unit 8 first calculates a power generation amount coefficient shown below for each measurement date and for each measurement time zone, with respect to the power generation amount actual accumulated in the power generation actual result storage unit 5. Assuming that the amount of generated power indicated by the actual amount of generated power is P, and the extrasolar radiation amount in the measured day and time zone corresponding to P is Q, the power generation coefficient K can be calculated by the following formula.
K = P / Q
The power generation amount performance classification unit 8 acquires the amount of extrasolar radiation from the atmospheric radiation calculation unit 7. When the weather is fine, the power generation coefficient is large because sunlight does not attenuate much in the atmosphere. Moreover, when the solar power generation amount prediction apparatus 1 predicts the power generation amount of a plurality of solar power generation systems 2, the power generation amount P is the rating of the solar power generation system 2 for the power generation amount performance of one solar power generation system 2. The normalized power generation coefficient may be determined by dividing by the output. The regression analysis described later can be performed collectively on the plurality of photovoltaic power generation systems 2 by the normalized power generation amount coefficient.

Next, the power generation amount performance classification unit 8 associates the power generation amount coefficient calculated for each measurement day and for each measurement time zone with the measured date and time period, according to the magnitude of the power generation amount coefficient It is classified into the power generation result group. A plurality of power generation result groups exist, and weather attributes corresponding to the magnitude of the power generation coefficient are associated. The number of groups to be classified is, for example, the same as the number of weather attributes obtained by weather forecasting. The weather attribute is an attribute related to the weather, such as "fine", "cloudy", "rain", etc., which is correlated with the magnitude of the power generation coefficient. The power generation coefficient is classified into a power generation result group according to the magnitude of the power generation coefficient by a clustering method such as the k-means method or the k-means ++ method. The power generation coefficient is classified into, for example, three groups, a large power generation coefficient group, a medium power group, and a small power generation coefficient. Then, by associating these groups with weather attributes such as "fine", "cloudy", and "rain", it is possible to associate the magnitude relationship of the power generation coefficient with the weather attribute. The number of weather attributes and the number of power generation result groups corresponding to them may not be three. For example, the weather attributes may be four of "fine", "slightly cloudy", "cloudy", and "rain", and the number of corresponding power generation result groups may be four.

The generated power result classification unit 8 may execute classification once a day, for example, at 23 o'clock, or may execute classification at the timing when the predicted generated power calculation unit 11 described later starts prediction of the generated power. . When the classification is performed, for example, all existing data classified into the power generation result group are erased before the classification is performed.

The weather forecast acquiring unit 10 acquires, from the weather information source 9, weather forecast information related to an area including the installation location of the solar power generation system 2 indicated by the position information source 6. The weather forecast information is, for example, information indicating a weather forecast in each time zone from now to 24 hours ahead. The weather forecast acquiring unit 10 may acquire the weather forecast information at the timing when the information indicated by the weather information source 9 is updated, and the weather information may be acquired each time an acquisition request from the predicted power generation calculator 11 described later comes. The latest weather forecast information may be obtained from the source 9. The timing at which the weather information source 9 is updated is, for example, when the weather information providing server of the Japan Meteorological Agency distributes new weather forecast information.

The predicted power generation amount calculation unit 11 first obtains, from the weather forecast acquisition unit 10, weather forecast information on each time zone of the forecast target day, and specifies a power generation result group associated with the weather attribute related to the weather forecast information. Do. For example, when the weather forecast information in a certain time zone is "fine", the power generation amount performance group associated with the "fine" weather attribute is specified. In addition, when the weather forecast information is information showing the transition of the weather such as "finely cloudy", "cloudy and fine", "finely cloudy" and "pokyly cloudy" by replacing with the prevailing weather attribute. "Sunny" may be considered "cloudy". Here, for example, it is assumed that the update interval of the weather forecast is every three hours, and the time slot to be the target of the power generation amount prediction is every thirty minutes. In this case, three hours corresponding to the weather forecast update interval are divided into six time zones every 30 minutes, and weather attributes are assigned to each time zone. For example, when the weather indicated by the weather forecast information is "smoothly cloudy", the weather attributes for the six time zones are assigned to "smoothly", "sunnyly", "cloudy", "sunnyly", "sunnyly", "cloudy" in order, for example. Similarly, when the weather indicated by the weather forecast information is “cloudy then fine”, for example, “cloudy”, “cloudy”, “cloudy”, “fine”, “fine”, “fine” are assigned in order.

Next, the predicted power generation amount calculation unit 11 calculates a predicted power generation amount by calculating a regression coefficient by regression analysis described later, and multiplying the calculated regression coefficient and the extrasolar radiation amount for the day and time zone of the prediction target. calculate. The predicted power generation amount calculation unit 11 acquires the extra-atmospheric radiation amount from the extra-atmospheric radiation amount calculation unit 7. The predicted power generation amount calculation unit 11 may predict the power generation amount of each time zone of the next day, for example, once a day, for example, at 23:00, or each time of the next day at the timing when the weather forecast acquisition unit 10 acquires weather forecast information. You may forecast the amount of power generation of the band. Since the predicted power generation amount can be calculated by multiplying the regression coefficient by the extrasolar radiation amount, it can be said that the regression coefficient is a predicted power generation coefficient.

The regression coefficient is determined by regression analysis of the amount of extrasolar radiation and the amount of power generation calculated from the power generation coefficient. As shown in FIG. 6, the amount of extrasolar radiation stored in the array IV and the amount of generated power stored in the array DV are subjected to regression analysis, and a straight line passing through the origin is defined as a regression line. The slope of this regression line is the regression coefficient. The amount of power generation may be calculated by the product of the power generation amount coefficient and the amount of extrasolar radiation. Further, the regression coefficient may more easily adopt the average value of the power generation coefficient belonging to the power generation result group. Further, as a regression curve, a straight line not passing through the origin, a logistic curve, or the like may be adopted to perform regression analysis. The functions of the array IV and the array DV are realized, for example, by providing the storage device to the predicted power generation amount calculation unit 11.

As a method of regression analysis, weighted least squares may be used. For example, it is conceivable to use a weighted least squares method in which the weight is increased as the day difference from the prediction target day is smaller, or as the prediction target day and the month and day are closer, or as the weather situation such as temperature is closer. For example, when the accumulation period of the power generation results is long, it is conceivable that a high building is constructed near the photovoltaic system 2 at a certain point in the accumulation period. In that case, a large difference occurs in the amount of solar radiation with respect to the photovoltaic system 2 before and after the building is built. Therefore, by performing regression analysis using a weighted least squares method in which the weight is increased as the day difference from the forecasted date decreases, the influence of the actual amount of power generation before a high building is built can be reduced. It becomes possible, and it becomes possible to predict the amount of power generation corresponding to the time series change.

The functional configuration of the solar power generation amount predicting apparatus 1 has been described above. The generated power result acquiring unit 4 functions as the generated power result acquiring means described in the claims. The power generation result storage unit 5 functions as a power generation result storage unit described in the claims. The extrasolar radiation calculating unit 7 functions as an extraneous radiation calculating unit described in the claims. The weather forecast acquisition unit 10 functions as a weather forecast acquisition unit described in the claims. The generated power result classification unit 8 functions as a generated power result classification unit described in the claims. The predicted power generation amount calculation unit 11 functions as a predicted power generation amount calculation unit described in the claims.

Next, a hardware configuration example of the solar power generation amount predicting apparatus 1 will be described with reference to FIG. The solar power generation amount predicting apparatus 1 includes a processor 1001 that executes processing, a memory 1002 that stores information, and an interface 1003 that transmits and receives information.

The memory 1002 is, for example, a storage device such as a dynamic random access memory (DRAM), a flash memory, or a magnetic disk drive. The memory 1002 implements the function of the power generation result storage unit 5.

The processor 1001 is, for example, a CPU (Central Processing Unit) that executes a program stored in the memory 1002. The processor 1001 realizes the respective functions of the power generation actual performance acquiring unit 4, the extrasolar radiation calculating unit 7, the power generation actual performance classifying unit 8, the weather forecast acquiring unit 10, and the predicted power generation calculating unit 11.

The interface 1003 is, for example, various I / O (Input / Output) ports. By connecting the interface 1003 to the power amount measuring device 3 and the weather information source 9, the power generation result acquiring unit 4 can acquire the power generation result, and the weather forecast acquiring unit 10 can acquire the weather forecast information.

Although only one processor 1001 is shown in FIG. 14, a plurality of processors may cooperate to realize the above function. Also, although only one memory 1002 is similarly illustrated in FIG. 14, a plurality of memories may cooperate to realize the above function.

For example, the first processor realizes the functions of the power generation actual result acquiring unit 4 and the power generation actual result classifying unit 8, and the second processor calculates the extrasolar solar radiation amount calculating unit 7, the weather forecast acquiring unit 10, and the predicted power generation amount calculation It may be configured to realize the function of the unit 11. At this time, the first processor and the second processor may not be mounted on one device. For example, the server on the cloud may include the first processor, and the client terminal connected to the server via the network may include the second processor. In this case, the server comprises the first memory, and the client terminal comprises the second memory.

The central part for realizing the functions of the solar power generation amount predicting apparatus 1 can be realized by using a normal computer system, not by a dedicated system. For example, a dedicated program for realizing the above functions is stored and distributed in a computer readable recording medium. Then, by installing a dedicated program on a computer and executing it, the above function can be realized by a normal computer system. In addition, part of the above functions may be shared by an operating system (OS), and the above functions may be realized by cooperation of the OS and a dedicated program. In this case, only the dedicated program may be stored in the recording medium.

FIG. 14 is an example, and the solar power generation amount predicting apparatus 1 can be configured by other than the above hardware configuration. For example, the function of the solar power generation amount predicting apparatus 1 may be realized by a dedicated signal processing circuit without using a normal computer system.

Next, the operation of the solar power generation amount predicting apparatus 1 will be described with reference to the drawings. The operation of the solar power generation amount prediction apparatus 1 is roughly classified into the classification of the power generation amount coefficient by the power generation amount actual classification unit 8 and the prediction of the power generation amount by the predicted power generation amount calculation unit 11. As described above, the power generation result acquisition unit 4 acquires the power generation result at regular intervals, and the power generation result storage unit 5 stores the acquired power result.

First, the classification of the power generation coefficient will be described with reference to FIG. The power generation actual result classification unit 8 acquires the power generation actual result from the power generation actual result storage unit 5 (step S101). The power generation performance to be acquired may be all of the power generation performance stored in the power generation performance storage unit 5 or may be a power generation performance for a predetermined period such as the last 15 days. In addition, the amount of power generation may be a target for a certain period of time such as 9-16 o'clock, or the amount of power generation may be a certain number of data such as the latest 500 pieces. However, since the difference of the power generation amount by meteorological attribute does not appear or it is hard to appear in the time zone where the extrasolar radiation amount is 0 or very small, such as night time, the power generation performance target for these time zones is not acquired. Let N be the number of data of the actual power generation amount acquired in step S101.

The power generation actual result classification unit 8 calculates a power generation amount coefficient by loop processing for all the acquired N power generation actual results (steps S102 to S105). In the loop, the extrasolar radiation amount corresponding to the measured day and time zone, which is included in the i-th power generation amount result, is acquired from the extrasolar radiation amount calculating unit 7 (step S103). The power generation amount coefficient is calculated by dividing the power generation amount included in the i-th power generation amount result by the acquired extrasolar radiation amount (step S104). When the loop processing is completed, the power generation amount performance classification unit 8 classifies the calculated power generation amount coefficient into a power generation amount performance group associated with the meteorological attribute (step S106).

The weather attribute corresponding to the power generation coefficient can be estimated by classifying the power generation coefficient into the power generation actual result group associated with the weather attribute. Therefore, it is not necessary to store past weather information, and the amount of data to be stored can be reduced.

Next, the prediction of the power generation amount will be described with reference to FIG. The predicted power generation amount calculation unit 11 predicts the power generation amount of each time zone by loop processing for all time zones to be predicted (steps S201 to S212). In the operation shown in FIG. 4, it is assumed that one day is divided into 48 time zones. Hereinafter, in the process from step S202 to S211, it demonstrates as what predicts the electric power generation amount prediction with respect to the T-th time slot | zone. However, T is a natural number from 0 to 47, and counts from the 0th. Also, the T-th time zone is simply described as a time zone T.

The predicted power generation amount calculation unit 11 acquires weather forecast information corresponding to the time zone T from the weather forecast acquisition unit 10 (step S202). Next, the predicted power generation amount calculation unit 11 specifies a power generation amount performance group associated with the weather attribute corresponding to the acquired weather forecast information (step S203). The predicted power generation amount calculation unit 11 acquires a power generation amount coefficient for the same time zone T among the power generation amount coefficients belonging to the specified power generation amount actual group (step S204). The number of acquired power generation amount coefficients is M.

The predicted power generation amount calculation unit 11 creates data for the above-described regression analysis by loop processing on all of the acquired M power generation amount coefficients (steps S205 to S208). The predicted power generation amount calculation unit 11 acquires the extrasolar radiation corresponding to the power generation coefficient to be processed from the extrasolar radiation calculation unit 7, and adds it to the array IV for storing the extrasolar radiation (step) S206). The predicted power generation amount calculation unit 11 acquires a power generation result corresponding to the power generation coefficient to be processed from the power generation result storage unit 5, and adds it to the array DV for storing the power generation result (step S207). Note that the power generation result may be acquired by multiplying the power generation amount coefficient by the extrasolar radiation amount.

The predicted power generation amount calculation unit 11 performs regression analysis of the extrasolar radiation amount stored in the array IV and the power generation amount stored in the array DV to obtain a regression coefficient (step S209). As mentioned above, it can be said that the regression coefficient is a predicted power generation coefficient. The predicted power generation amount calculation unit 11 acquires the outside air radiation amount corresponding to the time zone T from the outside air radiation amount calculation unit 7 (step S210). The predicted power generation amount calculation unit 11 calculates a predicted power generation amount corresponding to the time zone T by multiplying the calculated regression coefficient and the extrasolar radiation amount (step S211).

Note that if the power generation result group acquired in step S203 includes data of a large date that is different from the date of the forecasted date, the forecasting date is likely to differ from the power generation tendency, and therefore the forecasting accuracy is Deteriorate. Therefore, as for the power generation amount coefficient acquired in step S204, only one for a certain period may be acquired. For example, as shown in FIG. 5, it is conceivable to acquire the power generation amount coefficient limited to data within the last 15 days, and within the same month and day of the past year and 15 days before and after that.

The configuration and operation of the solar power generation amount predicting apparatus 1 have been described above. The photovoltaic power generation amount prediction apparatus 1 classifies the power generation amount coefficient into the power generation amount performance group based on the weather information, and specifies the power generation amount performance group to be referred to in the power generation amount prediction based on the weather information. Weather information is only used for classification and identification, and no weather information is stored. Therefore, according to the photovoltaic power generation amount forecasting device 1, without storing the weather information in the past, the installation position information of the photovoltaic power generation system 2, the past power generation amount performance, the extrasolar radiation amount, and the forecasted date The predicted power generation amount of each time zone on the prediction target day can be calculated on the basis of the weather forecast information in. Since there is no need to accumulate past weather information, the required data storage capacity can be reduced.

Each function of the solar power generation amount predicting apparatus 1 may be realized by a server on a network to calculate predicted power generation amounts of a plurality of solar power generation systems 2 connected to the network. Alternatively, each function of the solar power generation amount predicting apparatus 1 can be realized by a terminal installed for each of the solar power generation systems 2, and the predicted power generation amount of the solar power generation system 2 can be calculated. When each function of the photovoltaic power generation amount prediction apparatus 1 is realized by a server on a network, it may be configured by one server, or may be distributed to a plurality of servers. Furthermore, one function may be distributed to a plurality of servers. When all functions of the photovoltaic power generation prediction device 1 are realized by one server or one terminal, they are also the photovoltaic power generation prediction device 1.

The function of the solar power generation amount predicting apparatus 1 may be shared by the server and the terminal. Hereinafter, about the component which is common in the structure of the solar power generation amount prediction apparatus 1, the same code | symbol is attached | subjected and description is abbreviate | omitted. The solar power generation amount prediction apparatus 1A shown in FIG. 7 includes a server 71A and a terminal 72A.

The terminal 72A acquires the power generation result from the atmospheric radiation acquisition unit 10, the weather forecast acquiring unit 10, the power amount measuring device 3, and transmits the power generation result communication unit 73A to the server 71A, and the server 71A. A generated power result acquisition unit 4A that acquires a generated power result from the generated power result storage unit 5A, a generated power result classification unit 8, and a predicted generated power calculation unit 11.

The server 71A includes a power generation result receiving unit 74A that receives the power generation result from the terminal 72A, and a power generation result storage unit 5A that stores the power generation result received by the power generation result receiving unit 74A.

The terminal 72A is, for example, a HEMS (Home Energy Management System) terminal that manages energy in the home. A plurality of terminals 72A may be connected to one server 71A. According to this configuration, it is possible to reduce the data storage capacity required for the terminal 72A while accumulating a large amount of power generation results in the server 71A.

Moreover, as shown in FIG. 8, the function sharing of the solar power generation amount prediction apparatus 1 can also be made into the structure different from FIG. The solar power generation amount prediction device 1B illustrated in FIG. 8 includes a server 71B and a terminal 72B.

The terminal 72B receives the actual power generation result from the extraneous solar radiation amount calculation unit 7, the power amount measurement device 3, and outputs the generated power result communication unit 73B to the generated power coefficient calculation unit 84B, and the generated power result communication unit 73B. A power generation amount coefficient calculation unit 84B that calculates the power generation amount coefficient based on the acquired power generation amount results and the extrasolar radiation amount obtained from the extrasolar radiation amount calculation unit 7 and sends it to the server 71B, and the predicted power generation amount from the server 71B The power generation coefficient acquisition unit 89B acquires the coefficient and outputs it to the predicted power generation calculation unit 11B, the predicted power generation coefficient acquired from the power generation coefficient acquisition unit 89B, and the extrasolar radiation amount acquired from the extrasolar radiation calculation unit 7 And a predicted power generation amount calculation unit 11B that calculates a predicted power generation amount by multiplying

The server 71B includes the weather forecast acquisition unit 10, a power generation coefficient reception unit 85B that receives the power generation coefficient from the terminal 72B, and a power generation coefficient storage unit 86B that stores the power generation coefficient received by the power generation coefficient reception unit 85B. Based on the weather forecast information acquired from the weather forecast acquiring unit 10, the power generation factor classifying unit 87B that acquires the power generation factor from the power generation factor storage unit 86B and classifies it into the power generation factor group associated with the meteorological attribute. Then, the power generation coefficient group is selected, regression analysis is performed on the power generation coefficient belonging to the selected power generation coefficient group, and the predicted power generation coefficient in each time zone of the prediction target date is calculated and transmitted to the terminal 72B And a power generation amount coefficient calculation unit 88B.

With this configuration, the server 71B is responsible for a part of the processing necessary for the power generation amount prediction. Therefore, in addition to being able to reduce the data storage capacity necessary for the terminal 72B, the processing amount of the terminal 72B can also be reduced.

Second Embodiment
The functional configuration of the solar power generation amount prediction device 1C will be described with reference to FIG. The solar power generation amount prediction apparatus 1C is used for the extraneous solar radiation amount calculation unit 7, the weather forecast acquisition unit 10, the power generation amount actual acquisition unit 4, the predicted power generation amount calculation unit 11, and the power generation amount prediction Of the power generation results stored in the power generation result storage unit 5C and the power generation result storage unit 5C, the values of which are updated by the power generation result updating unit 94C. The power generation actual result classification unit 8C that classifies only the value including the value indicating use for the amount prediction into the power generation amount actual performance group, and the predicted power generation amount calculated by the predicted power generation amount calculation unit 11 together with the predicted day and time Accuracy of power generation prediction based on the comparison result of predicted actual storage unit 92C accumulated as actual results, power generation actual stored in generated power actual results storage 5C, and predicted actuals stored in predicted actual storage 92C Amount of power generation that causes An extraction unit 93C that extracts results, and a power generation result update unit 94C that updates the power generation result storage unit 5C to make the power generation result extracted by the extraction unit 93C unusable for prediction of the power generation. Prepare.

The power generation result table storing the power generation results stored in the power generation result storage unit 5C is, as shown in FIG. 10, a date field 21, a time zone field 22, a solar power generation system ID field 23, and a power generation amount. In addition to the field 24, a prediction use flag field 101C is included. The predicted use flag field 101C stores a value indicating whether the power generation actual value is used for power generation prediction. Although the initial value of the value stored in the predicted usage flag field 101C is "used", it is updated by the power generation actual result updating unit 94C.

The classification of the power generation amount coefficient by the power generation amount performance classification unit 8C is substantially the same as in the first embodiment, but only for the power generation amount results for which the value stored in the predicted use flag field 101C is "use". Calculation of the power generation coefficient and classification of the power generation coefficient are performed. The predicted power generation amount calculation unit 11 uses these power generation results by not performing calculation and classification of the power generation coefficient with respect to the power generation results for which the value stored in the predicted use flag field 101C is “do not use”. It will not be.

The prediction result storage unit 92C associates the predicted power generation amount calculated by the predicted power generation amount calculation unit 11 with the predicted date and time zone, and stores it as a predicted result in the prediction result table. As shown in FIG. 11, the forecasted performance table includes a date field 111C, a time zone field 112C, a photovoltaic system ID field 113C, and a predicted power generation field 114C.

The extraction unit 93C acquires the power generation result from the power generation result storage unit 5C, and acquires the prediction result from the prediction result storage unit 92C. The extraction unit 93C compares the acquired predicted actual results with the acquired actual power generation results, and determines the accuracy of the generated power prediction based on the difference between the predicted generated power and the measured generated power in the same time zone on the same day. Extract the amount of power generation that is the cause of lowering. The operation of the extraction unit 93C will be described later. The operation of the extraction unit 93C is performed, for example, at a timing at which the predicted power generation amount calculation unit 11 tries to predict a specific day as a prediction target day.

Of the power generation results stored in the power generation result storage unit 5C, the power generation result updating unit 94C changes the value of the predicted use flag field 101C corresponding to the power generation result extracted by the extracting unit 93C to "do not use". Make updates.

The configuration of the solar power generation amount prediction device 1C has been described above. The prediction result storage unit 92C functions as a prediction result storage unit described in the claims. The extraction unit 93C functions as an extraction unit described in the claims. The power generation result updating unit 94C functions as a power generation result updating unit described in the claims.

Next, the operation of the extraction unit 93C will be described with reference to FIGS. 5, 12 and 13 while giving a specific example. As an example, assume that the forecast target date is July 1, 2017, and the forecast time zone is a time zone represented by “0800”.

First, the extraction unit 93C acquires the power generation result to be used for power generation prediction from the power generation result storage unit 5C (step S301). For example, as shown in FIG. 5, the power generation results used for power generation prediction are data within the last 15 days, and within the same month of the past 3 years and around 15 days thereafter.

Next, the extraction unit 93C creates a prediction deviation list by loop processing on each of the acquired power generation amount results (steps S302 to S306). First, the extraction unit 93C extracts the measured day and time zone included in the power generation amount record to be subjected to the loop processing. The prediction result including the same prediction date and prediction time zone as the extracted date and time zone is acquired from the prediction history storage unit 92C (step S303). In this example, the forecast results for the “0800” time zone within the last 15 days and within the same month and day of the past 3 years and the 15 days before and after that. Next, the extraction unit 93C determines whether there is a certain difference or more between the amount of power generation included in the actual amount of power generation and the predicted amount of power generation included in the predicted actual result (step S304). The predetermined difference or more is, for example, the case where the difference between the power generation amount and the predicted power generation amount is 20% or more of the power generation amount. In this example, it is assumed that there is a predetermined difference on June 17, 2017, on the same day, on the same day, on the same day, on the same day, on the same day, on the same day. If there is a difference of a certain level or more (step S304: Yes), a set of the prediction date and the prediction time zone is added to the prediction deviation list 131C (step S305), and the process proceeds to step S306. If there is no difference greater than or equal to a predetermined amount (step S304: No), the process proceeds to step S306 without adding to the prediction deviation list 131C. In this example, as shown in FIG. 13, “0800” time zones of June 17, 2017, 20, 28, and 30 of 2017 are added to the misprediction list 131C.

Then, the extraction unit 93C extracts the power generation actual value used when the power generation amount prediction for each prediction date and the prediction time slot in the prediction deviation list 131C is performed as the exclusion candidate. In this example, the actual power generation amount in the “0800” time zone of the day indicated by the thick line in FIG. 13 is extracted as the exclusion candidate. Then, the excluded candidates extracted a predetermined number of times or more are extracted as a power generation amount performance which causes the accuracy of the power generation amount prediction to be lowered (step S307). The predetermined number of times or more is, for example, three or more times. In addition, for example, the number of days of 10% or more of the total number of days of the power generation actual value used for the power generation prediction may be used. In this example, as shown in FIG. 13, the power generation performance in the “0800” time zone of each day from June 13, 2017 to the same year, January 19, 2017 is extracted three or more times as an exclusion candidate. Therefore, these power generation results are extracted as power generation results that cause the accuracy of the power generation prediction to be lowered.

The generated power result updating unit 94C compares the value of the predicted use flag field 101C corresponding to the power generation result with respect to the power generation result that is the cause of lowering the accuracy of power generation prediction extracted by the operation of the extracting unit 93C. Update to "do not use".

In the above description, the value of the predicted use flag field 101C is “use” or “do not use”, but may be represented by a numerical value of reliability instead of these two values. For example, the difference between the power generation amount and the predicted power generation amount is made into a point, and the weight of the power generation amount performance is calculated based on the point total for each exclusion candidate as a weight when performing regression analysis by the weighted least squares method. You may use. Alternatively, a weighted least squares regression analysis may be performed, using the number of times of extraction as an exclusion candidate as a weight.

Since the solar power generation amount prediction apparatus 1C can perform prediction by excluding the power generation amount performance causing the misprediction, the prediction accuracy is improved.

The present invention is capable of various embodiments and modifications without departing from the broad spirit and scope. In addition, the embodiment described above is for describing the present invention, and does not limit the scope of the present invention. That is, the scope of the present invention is indicated not by the embodiments but by the claims. And, various modifications applied within the scope of the claims and the meaning of the invention are considered to be within the scope of the present invention.

The present invention is suitable for prediction of the amount of power generation in a solar power generation system.

DESCRIPTION OF SYMBOLS 1, 1A, 1B, 1C Photovoltaic power generation forecasting device, 2 photovoltaic power generation system, 3 electric energy measuring device, 4, 4A generated power result acquisition part, 5, 5A, 5C , 7 extraterrestrial solar radiation amount calculation unit, 8, 8C power generation result classification unit, 9 weather information source, 10 weather forecast acquisition unit, 11 11B predicted power generation calculation unit, 21 date field, 22 time zone field, 23 Solar power generation system ID field, 24 power generation field, 71A, 71B server, 72A, 72B terminal, 73A, 73B power generation result communication unit, 74A power generation result reception unit, 84B power generation coefficient calculation unit, 85B power generation coefficient reception Part, 86 B power generation factor storage part, 87 B power generation factor classification part, 88 B predicted power generation factor calculation part, 89 B power generation factor acquisition part, 2C forecast result storage unit, 93C extraction unit, 94C power generation result update unit, 101C forecast use flag field, 111C date field, 112C time zone field, 113C solar power generation system ID field, 114C forecast power generation field, 131C forecast Outlier list, 1001 processor, 1002 memory, 1003 interface.

Claims (10)

Power generation actual result acquisition means for acquiring the amount of power generation of the photovoltaic power generation system as a power generation actual with the acquisition date and the acquisition time zone,
Power generation result storage means for storing the power generation results acquired by the power generation result acquisition means;
Extra-atmosphere solar radiation amount calculating means for calculating the amount of extra solar radiation at the installation place of the solar power generation system;
Weather forecast acquiring means for acquiring weather forecast information including in a target area the installation location of the solar power generation system;
The outside air irradiance calculation means is made to calculate the outside air irradiance in the acquisition date and the acquisition time zone included in the power generation result accumulated by the power generation result storage means, and the power generation amount included in the power generation result and the Power generation actual result classification means for classifying the power generation amount coefficient calculated based on the calculated amount of extraterrestrial solar radiation into the power generation amount actual result group associated with the meteorological attribute;
Weather forecast information including the day and time zone to be predicted acquired by the weather forecast acquiring means, the power generation amount performance group associated with the weather attribute according to the acquired weather forecast information, and Predicted power generation amount calculating means for calculating a predicted power generation amount on the basis of the extraneous solar radiation amount in the day and the time zone to be predicted, which is calculated by the external air radiation amount calculating means;
Photovoltaic power generation forecasting device provided with The power generation actual result classification means classifies the power generation coefficient according to the magnitude of the power generation coefficient.
The solar power generation amount prediction apparatus according to claim 1. The predicted power generation amount calculating means calculates the predicted power generation amount based on the predicted power generation coefficient which is an average value of the power generation amount coefficient belonging to the power generation amount actual group and the extrasolar radiation amount in the day and time zone to be predicted. calculate,
The solar power generation amount prediction apparatus according to claim 1. The predicted power generation amount calculating means calculates the power generation amount included in the power generation amount result corresponding to the power generation amount coefficient belonging to the power generation amount result group and the extraneous solar radiation amount in the acquisition date and the acquisition time zone included in the power generation amount result. The predicted power generation amount is calculated based on the predicted power generation amount coefficient obtained by regression analysis, and the extrasolar radiation amount in the day and time zone to be predicted.
The solar power generation amount prediction apparatus according to claim 1. Predicted performance storage means for storing the predicted power generation amount calculated by the predicted power generation amount calculation means as a predicted performance together with a predicted date and a predicted time zone;
The amount of power generation actual accumulated by the power generation actual result storage means is compared with the predicted actual result accumulated by the predicted actual result storage means, and between the amount of power generation included in the power generation actual and the predicted amount of power included in the predicted actual result Extracting means for acquiring predicted actual results on a day and a time zone in which there is a difference of at least a predetermined level, and extracting the actual amount of generated power used at least in the predicted amount of generated power concerning the acquired predicted actual results;
Power generation actual result updating means for updating the power generation actual results accumulated by the power generation actual result storage means so that the power generation actual results extracted by the extraction means can not be used for power generation amount prediction;
The solar power generation amount prediction device according to any one of claims 1 to 4, further comprising: Predicted performance storage means for storing the predicted power generation amount calculated by the predicted power generation amount calculation means as a predicted performance together with a predicted date and a predicted time zone;
The amount of power generation obtained by comparing the amount of power generation performance used in the power generation amount prediction according to the predicted performance obtained by comparing the power generation amount performance stored in the power generation amount performance storage means with the predicted performance stored in the predicted performance storage means An extraction unit that extracts a string after linking the difference between the power generation amount of the actual result and the predicted power generation amount of the predicted actual result as a weight,
Power generation record updating means for updating the power generation record stored in the power generation record storage means, associating the power generation record extracted by the extraction means with the weight, and storing the power generation record in the power generation record storage means;
And further
The predicted power generation amount calculating means calculates a predicted power generation amount by a weighted least squares method using the weight.
The photovoltaic power generation amount prediction device according to any one of claims 1 to 4. Power generation actual result acquisition means for acquiring the amount of power generation of the photovoltaic power generation system as a power generation actual with the acquisition date and the acquisition time zone,
Extra-atmosphere solar radiation amount calculating means for calculating the amount of extra solar radiation at the installation place of the solar power generation system;
Of the power generation coefficients calculated based on the power generation amount included in the power generation amount actual result and the calculated extrasolar radiation amount and classified into the power generation amount result group associated with the meteorological attribute, the day to be a forecast target And a power generation amount coefficient classified into a power generation amount performance group associated with a weather attribute corresponding to weather forecast information including a time zone, and the day to be predicted, calculated by the external solar radiation amount calculation means And predicted power generation amount calculating means for calculating the predicted power generation amount based on the external solar radiation amount in the time zone, and
Photovoltaic power generation forecasting device provided with Solar power system,
The photovoltaic power generation amount prediction device according to any one of claims 1 to 7,
Equipped with
The power generation result acquiring means acquires a power generation amount from the solar power generation system.
Solar power forecasting system. A power generation result storage step of accumulating the results of the power generation amount of the photovoltaic power generation system as the power generation amount results,
A weather forecast acquisition step for acquiring weather forecast information;
The power generation coefficient calculated based on the power generation amount included in the power generation amount results and the extrasolar radiation amount in the day and time zone when the power generation amount was acquired is classified into a power generation amount result group associated with the meteorological attribute Power generation actual result classification step,
Weather forecast information including the day and time zone to be predicted, acquired in the weather forecast acquiring step, the power generation result group associated with the weather attribute according to the acquired weather forecast information, and A predicted power generation amount calculation step of calculating a predicted power generation amount based on the extrasolar radiation amount in the day and the time zone to be predicted;
Forecasting method. On the computer
A power generation result storage step of accumulating the results of the power generation amount of the photovoltaic power generation system as the power generation amount results,
A weather forecast acquisition step for acquiring weather forecast information;
The power generation coefficient calculated based on the power generation amount included in the power generation amount results and the extrasolar radiation amount in the day and time zone when the power generation amount was acquired is classified into a power generation amount result group associated with the meteorological attribute Power generation actual result classification step,
Weather forecast information including the day and time zone to be predicted, acquired in the weather forecast acquiring step, the power generation result group associated with the weather attribute according to the acquired weather forecast information, and A predicted power generation amount calculation step of calculating a predicted power generation amount based on the extrasolar radiation amount in the day and the time zone to be predicted;
A program that runs

Images