Model generation device, generated amount of power estimation device, information processing method, and program

Abstract

To enable the power output of a photovoltaic power generation device to be more accurately estimated.SOLUTION: A model generation device 2 comprises: a data acquisition unit 21 for acquiring meteorological data 50 including the performance value 43 of generated amount of power of a photovoltaic power generation device 4, output limiting information 44, and a physical quantity related to meteorology; a determination unit 22 for determining whether the performance value 43 of generated amount of power is limited, on the basis of the output limiting information 44 and / or the performance value 43 of generated amount of power; an analysis data generation unit 23 for generating data 60 for analysis in which the performance value 43 of the generated amount of power having been determined as being not limited by the determination unit 22 is associated with the meteorological data 50; and a model generation unit 24 for generating a model 70 for letting a computer function to calculate the generated amount of power on the basis of the inputted physical quantity related to meteorology by regression analysis or machine learning of the data 60 for analysis on the basis of a prescribed algorithm.SELECTED DRAWING: Figure 2

Description

Technical Field

[0001] The present invention relates to a model generation device, a power generation amount estimation device, an information processing method, and a program. For example, the present invention relates to a model generation device, an information processing method, and a program for generating a model for estimating a power generation amount indicating the magnitude of power output from a photovoltaic power generation device, and a power generation amount estimation device, an information processing method, and a program for estimating the power generation amount of a photovoltaic power generation device using the above model.

Background Art

[0002] With the spread of photovoltaic power generation devices installed in general consumers and the like, the need to accurately estimate the power generation amount indicating the magnitude of power output from the photovoltaic power generation device is increasing.

[0003] Here, the power generation amount refers to the power generation output or the amount of generated electric power per unit time.

[0004] As a conventional technique for estimating the power generation output of a photovoltaic power generation device, for example, by associating the actual value of the power generation output by a past photovoltaic power generation device with the actual value of the solar irradiance and performing machine learning, a model for estimating the power generation output of the photovoltaic power generation device from the solar irradiance is generated, and the power generation output of the photovoltaic power generation device is estimated from the solar irradiance using the model. This technique is disclosed in Patent Document 1.

Prior Art Documents

Patent Documents

[0005]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0006] Incidentally, it is generally known that the power generation output of a photovoltaic power generation device changes according to conditions related to weather such as solar irradiance and temperature. On the other hand, the power generation output of a photovoltaic power generation device may be restricted by factors different from the above-mentioned weather-related conditions. For example, in a photovoltaic power generation device including a solar panel and a power conversion device (power conditioner, hereinafter also referred to as "PCS (Power Conditioning System)") connected to the solar panel, when the output power of the solar panel is greater than the rated output of the PCS (hereinafter also referred to as "overload"), the power (power generation output) output from the photovoltaic power generation device via the PCS is restricted to a magnitude equal to or less than the rated output of the PCS. Further, for example, in self-delivery and connect & manage, etc., when the power generation output of the photovoltaic power generation device is suppressed by an output suppression command signal, the power (power generation output) output from the photovoltaic power generation device via the PCS is restricted based on the value specified by the output suppression command signal.

[0007] In this way, when regression analysis or machine learning is performed using, as analysis data (teacher data), a data set in which the actual performance data of the power generation output measured in a situation where the power generation output of the photovoltaic power generation device is restricted is associated with weather data such as solar irradiance and temperature at that time, the estimation accuracy of the power generation output of the photovoltaic power generation device by the model generated by the regression analysis or machine learning may be lowered. For example, under conditions where the solar irradiance exceeds a predetermined magnitude, there is a possibility that the error between the estimated value of the power generation output of the photovoltaic power generation device and the original power generation output will increase.

[0008] The present invention has been made in view of the above-mentioned problems, and an object thereof is to enable more accurate estimation of the power generation amount of a photovoltaic power generation device.

Means for Solving the Problems

[0009] The model generation device according to a representative embodiment of the present invention includes: a data acquisition unit that acquires power generation performance data including an actual value of the amount of power output from a photovoltaic power generation device and output limit information regarding the limit of the amount of power generation, and a physical quantity related to weather corresponding to the power generation performance data; a determination unit that determines whether or not the actual value of the amount of power generation is subject to a limit based on at least one of the output limit information and the actual value of the amount of power generation included in the power generation performance data acquired by the data acquisition unit; an analysis data generation unit that generates analysis data in which the actual value of the amount of power generation determined by the determination unit not to be subject to a limit is associated with the physical quantity related to weather; a storage unit that stores the analysis data generated by the analysis data generation unit; and a model generation unit that generates a model for causing a computer to function to calculate the amount of power generation based on the input physical quantity related to weather by performing regression analysis or machine learning on the analysis data stored in the storage unit based on a predetermined algorithm.

Advantages of the Invention

[0010] According to the model generation device of the present invention, it is possible to more accurately estimate the amount of power generation of a photovoltaic power generation device.

Brief Description of the Drawings

[0011]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10A

Figure 10B

Figure 11A

Figure 11B

Mode for Carrying Out the Invention

[0012] 1. Outline of the Embodiment First, an outline of a typical embodiment of the invention disclosed in the present application will be described. In the following description, as an example, the reference signs in the drawings corresponding to the components of the invention are described with parentheses.

[0013] [(1) The model generation device (2) according to a representative embodiment of the present invention includes actual power generation data (42) including an actual value (43) of the amount of power output from the photovoltaic power generation device (4) and output limit information (44) regarding the limit of the amount of power generation, and physical quantities (50, 51, 52) related to the weather corresponding to the actual power generation data. A data acquisition unit (21) for acquiring, a determination unit (22) for determining whether or not the actual value of the power generation amount is limited based on at least one of the output limit information and the actual value of the power generation amount included in the actual power generation data acquired by the data acquisition unit, and the physical quantity related to the weather And an analysis data generation unit (23) for generating analysis data (60) in which the actual value of the power generation amount determined not to be limited by the determination unit is associated, and a storage unit (25) for storing the analysis data generated by the analysis data generation unit. And a model generation unit (24) for generating a model (70) for causing a computer to function so as to calculate the power generation amount based on the input physical quantity related to the weather by performing regression analysis or machine learning on the analysis data stored in the storage unit based on a predetermined algorithm. It is characterized by comprising.]

[0014] [(2) In the model generation device, the photovoltaic power generation device includes a solar panel (40) and a power conversion device (41) that converts the power generated by the solar panel into AC power and outputs it. The output limit information includes a rated output value (45) indicating the magnitude of the rated output of the power conversion device. The determination unit may determine whether or not the actual value of the power generation amount is limited based on a comparison result between the rated output value and the actual value of the power generation amount or a comparison result between a value obtained by converting the rated output value into the amount of power generated per unit time and the actual value of the power generation amount.]

[0015] [(3) In the model generation device according to the above (2), the determination unit determines that the actual value of the power generation amount is not limited when the actual value of the power generation amount is smaller than a value obtained by multiplying the rated output value or a value obtained by converting the rated output value into the amount of power generated per unit time by a predetermined coefficient K1 (0 <K1 ≦ 1). It may be.]

[0016] 〔4〕In the model generation device according to any one of 〔1〕to 〔3〕above, when the output suppression command value (46) that designates the suppression amount of the generated power output from the photovoltaic power generation device is not included in the power generation performance data as the output limit information, the determination unit may determine that the actual value of the generated power is not restricted.

[0017] 〔5〕In the model generation device according to 〔4〕above, when the actual value of the generated power is smaller than the value obtained by multiplying the generated power after suppression based on the output suppression command value by a predetermined coefficient K2 (0 < K2 ≤ 1), the determination unit may determine that the actual value of the generated power is not restricted.

[0018] 〔6〕In the model generation device according to 〔4〕above, the output limit information further includes execution necessity information indicating the necessity of executing output suppression. When the execution necessity information indicates that it is not necessary to execute output suppression, the determination unit may determine that the actual value of the generated power is not restricted regardless of the output suppression command value.

[0019] 〔7〕In the model generation device according to any one of 〔1〕to 〔6〕above, the physical quantity related to the weather includes the solar radiation amount (51). When the solar radiation amount is smaller than a predetermined reference value, the determination unit may determine that the actual value of the generated power is not restricted.

[0020] 〔8〕The power generation amount estimation device (3) according to a representative embodiment of the present invention includes an input data acquisition unit (31) that acquires input data (80) including physical quantities related to weather, and a plurality of analysis data generated by associating the physical quantities related to the weather with the magnitude of the power output from the photovoltaic power generation device. A storage unit (36) that stores a model (70) for causing a computer to function so as to calculate the power generation amount based on the input physical quantities related to the weather, which is generated by performing regression analysis or machine learning on a predetermined algorithm; and based on the model stored in the storage unit, an estimation unit (32) that estimates the power generation amount corresponding to the physical quantities related to the weather included in the input data acquired by the input data acquisition unit. The estimation unit includes a power generation amount calculation unit (33) that calculates an estimated value of the power generation amount by inputting the physical quantities related to the weather included in the input data into the model, a determination unit (34) that determines whether or not the power generation amount is restricted under the condition of estimating the power generation amount of the photovoltaic power generation device, and when it is determined by the determination unit that the power generation amount is not restricted, the estimated value of the power generation amount calculated based on the model is output as the estimation result of the power generation amount, and when it is determined by the determination unit that the power generation amount is restricted, a value obtained by correcting the estimated value of the power generation amount calculated based on the model so as to be smaller is output as the estimation result of the power generation amount.

[0021] 〔9〕In the power generation amount estimation device according to 〔8〕 above, the photovoltaic power generation device includes a solar panel (40) and a power conversion device (41) that converts the power generated by the solar panel into AC power and outputs it. The storage unit stores a rated output value (45) indicating the magnitude of the rated output of the power conversion device. The determination unit determines that the power generation amount of the photovoltaic power generation device is not restricted under the condition of estimating the power generation amount when the estimated value of the power generation amount is smaller than the rated output value or the value obtained by converting the rated output value into the power generation amount per unit time. When the estimated value of the power generation amount is larger than the rated output value or the value obtained by converting the rated output value into the power generation amount per unit time, the determination unit determines that the power generation amount of the photovoltaic power generation device is restricted under the condition of estimating the power generation amount. When the estimated value of the power generation amount is larger than the rated output value or the value obtained by converting the rated output value into the power generation amount per unit time, and thus it is determined that the power generation amount is restricted, the output unit corrects the estimated value of the power generation amount so as not to exceed the rated output value or the value obtained by converting the rated output value into the power generation amount per unit time, and may output the corrected value as the estimation result of the power generation amount of the photovoltaic power generation device.

[0022] 〔10〕In the power generation amount estimation device according to 〔8〕 or 〔9〕 above, the determination unit determines that the power generation amount of the photovoltaic power generation device is not restricted under the condition of estimating the power generation amount when an output suppression command value (46) for designating the suppression amount of the power generation amount output from the photovoltaic power generation device is not included in the input data. When the output suppression command value is included in the input data, the determination unit determines that the power generation amount of the photovoltaic power generation device is restricted under the condition of estimating the power generation amount. When the output suppression command value is included in the input data, and thus it is determined that the power generation amount is restricted, the output unit corrects the estimated value of the power generation amount so as not to exceed the suppressed power generation amount based on the output suppression command value, and may output the corrected value as the estimation result of the power generation amount of the photovoltaic power generation device.

[0023] 〔11〕The information processing method according to a representative embodiment of the present invention includes: a first step (S11) of acquiring power generation performance data including an actual value of the amount of power generated indicating the magnitude of the power output from a solar power generation device and output limit information regarding the limit of the amount of power generated, and a physical quantity related to the weather corresponding to the power generation performance data; a second step (S12) of determining whether or not the actual value of the amount of power generated is subject to a limit based on at least one of the output limit information and the actual value of the amount of power generated included in the power generation performance data acquired in the first step; a third step (S13) of generating analysis data in which the actual value of the amount of power generated determined not to be subject to a limit in the second step is associated with the physical quantity related to the weather; a fourth step (S13, S14) of storing the analysis data generated in the third step; and a fifth step (S2, S3) of generating a model for causing a computer to function so as to calculate the amount of power generated based on the input physical quantity related to the weather by performing regression analysis or machine learning on the analysis data stored in the fourth step based on a predetermined algorithm.

[0024] 〔12〕The program according to a representative embodiment of the present invention causes the computer to execute the first step to the fifth step in the information processing method described in the above

[11] .

[0025] 〔13〕The information processing method according to a representative embodiment of the present invention includes a first step (S41) of acquiring input data including physical quantities related to weather, and a plurality of analysis data generated by associating the physical quantities related to the weather with the amount of power output from a photovoltaic power generation device, and performing regression analysis or machine learning on the basis of a predetermined algorithm. Based on a model for causing a computer to function so as to calculate the amount of power generation based on the input physical quantities related to the weather, a second step (S42 to S46) of estimating the amount of power generation corresponding to the physical quantities related to the weather included in the input data acquired in the first step is included. The second step includes a third step (S42) of calculating an estimated value of the amount of power generation by inputting the physical quantities related to the weather included in the input data into the model, a fourth step (S43) of determining whether or not the amount of power generation is restricted under the condition of estimating the amount of power generation of the photovoltaic power generation device, a fifth step (S46) of outputting, as an estimation result of the amount of power generation, the estimated value of the amount of power generation calculated based on the model when it is determined in the fourth step that the amount of power generation is not restricted, and a sixth step (S44, S45) of outputting, as an estimation result of the amount of power generation, a value obtained by correcting the estimated value of the amount of power generation calculated based on the model so as to be smaller when it is determined in the fourth step that the amount of power generation is restricted.

[0026] 〔14〕The program according to a representative embodiment of the present invention causes a computer to execute the first step to the sixth step in the information processing method described in the above 〔13〕.

[0027] 2. Specific Examples of Embodiments Hereinafter, specific examples of embodiments of the present invention will be described with reference to the drawings. In the following description, the same reference numerals are assigned to the common components in each embodiment, and the repeated description will be omitted.

[0028] ≪Embodiment≫ FIG. 1 is a diagram showing the configuration of the power generation amount estimation system 1 according to the embodiment.

[0029] The power generation amount estimation system 1 shown in FIG. 1 generates analysis data associating the actual value of the power generation amount indicating the magnitude of the power output from the photovoltaic power generation device 4 with the physical quantity related to the weather during power generation, and generates a model by performing regression analysis or machine learning on the analysis data. Using the model, the power generation amount of the photovoltaic power generation device 4 is estimated from the estimated value of the physical quantity related to the input weather.

[0030] As described above, in the present embodiment, the power generation amount refers to the power generation output or the power generation amount per unit time. The unit of time refers to an arbitrary unit of time such as 30 minutes or 1 hour.

[0031] As shown in FIG. 1, the power generation amount estimation system 1 is configured to be connectable to, for example, the network 6. The power generation amount estimation system 1, the photovoltaic power generation device 4, the server 5 that provides weather data, and the power management device 7 can communicate with each other via the network 6.

[0032] When the power management device 7 manages the power in the solar power generation plant installation customer, the network 6 is, for example, a LAN (Local Area Network). When the power management device 7 manages a power supply network including a plurality of customers and power generation plants, it is a wide area network (WAN: Wide Area Network) represented by, for example, the Internet. When the network 6 is a customer premise communication network such as a LAN, it is further connected to a wide area network (WAN) represented by, for example, the Internet through a router or the like, and an external institution that provides weather data (a related institution of the Meteorological Agency or a weather information providing institution such as a private weather information service) and information (weather data, output suppression commands, etc.) from an external institution that issues an output suppression command (for example, a transmission and distribution system operator such as a general transmission and distribution utility or a distribution utility). Even when the network 6 is a communication network that connects a plurality of customers and power generation plants such as a WAN, information from the above external institutions is received through a wider communication network. The weather data may be provided from an external weather information providing service, or may be based on the data measured by a weather observation device installed near the solar power generation device. Any weather data is received and stored in the server 5 that stores the weather data, and is transmitted to the power management device 7 and the power generation amount estimation system 7 as needed. The output suppression command from the external institution is once received by the power management device 7, and the format is converted and sent to the solar power generation device 4 and the power generation amount estimation system 7 as an output suppression command signal. In FIG. 1, only the output suppression command signal is shown as being transmitted through a separate route, but it may be transmitted through the network 6 in the same way as other data. Of course, each data may use the same communication route, or may be a separate route (a separate communication system).

[0033] FIG. 1 shows, as an example, a case where one photovoltaic power generation device 4 is connected to a power management device 7 and a power generation amount estimation system 1 via a network 6. However, a plurality of photovoltaic power generation devices 4 may be connected to the power management device 7 and the power generation amount estimation system 1 via the network 6. Even when a plurality of photovoltaic power generation devices 4 are connected to the power management device and the power generation amount estimation system 1 via the network 6, the power generation amount estimation system 1 collects and learns data on the power generation performance of the photovoltaic power generation device 4 for each photovoltaic power generation device 4, and estimates the power generation amount for each photovoltaic power generation device 4 using the model generated for each photovoltaic power generation device 4. Details of the power generation amount estimation system 1 will be described later.

[0034] The photovoltaic power generation device (PV: photovoltaics) 4 is installed, for example, within the site of a consumer. The photovoltaic power generation device 4 includes, for example, a solar panel 40 and a power conversion device 41. The solar panel 40 is a device that generates electricity by converting the energy of light into electrical energy.

[0035] The power conversion device 41 is a device that converts the power generated by the solar panel 40 into AC power and outputs it. The power conversion device 41 is, for example, a power conditioner (PCS). Hereinafter, the power conversion device 41 may be referred to as "PCS41". PCS41 is configured to be able to output the converted AC power to at least one of the load of the consumer and the power grid.

[0036] The power generation output of the photovoltaic power generation device 4 may be restricted. As described above, the power generation output of the photovoltaic power generation device 4 may be restricted by the rated output of PCS41 and an output suppression command signal. The output suppression command signal may be created based on a power generation output suppression command from an external organization (for example, an operator of a power transmission and distribution system such as a general power transmission and distribution company or a distribution company), or may be created independently by an internal power management device.

[0037] The PCS 41 has a rated output that indicates the magnitude of the power that can be output. The PCS 41 outputs power so as not to exceed its rated output. For example, in the case of overloading where the output power of the solar panel 40 is greater than the rated output of the PCS 41, the PCS 41 outputs the power generated by the solar panel with the rated output of the PCS 41 as the upper limit value. For example, when the output power of the solar panel 40 is 900 kW and the rated output of the PCS 41 is 700 kW, the maximum value of the power output from the PCS 41 is 700 kW. Thus, when the solar power generation device 4 is overloaded, the power generation output of the solar power generation device 4 is limited to be equal to or less than the rated output of the PCS 41. Hereinafter, the information (data) indicating the value of the rated output of the PCS 41 is referred to as the "PCS rated output value 45".

[0038] Also, the PCS 41 suppresses the power generation output of the solar power generation device 4 in accordance with, for example, an output suppression command signal transmitted from the solar power generation plant operator or the power management device 7 of the power supply network. The output suppression command signal includes an output suppression command value that specifies the suppression amount of the power generation amount output from the solar power generation device. For example, the output suppression command value is a value that specifies the suppression amount of the power output from the solar power generation device 4 or the power generation amount per unit time. Specifically, the output suppression command value may be, for example, a value that specifies the upper limit value of the power output from the solar power generation device 4, or a value that specifies the ratio (for example, 50% or the like) to be suppressed with respect to the power that can be output from the solar power generation device 4 (for example, the rated output of the PCS 41). Note that the output suppression command signal may be a signal that specifies the time period during which the power generation output of the solar power generation device 4 should be suppressed and the output suppression command value.

[0039] When the PCS 41 receives an output suppression command signal, it calculates the upper limit value of the power generation output that can be output based on the output suppression command value included in the output suppression command signal, and outputs the generated power of the solar panel 40 so as not to exceed the upper limit value. Thereby, the power generation output of the solar power generation device 4 becomes equal to or less than the upper limit value determined by the output suppression command value. Hereinafter, the information (data) indicating the output suppression command value is denoted as "output suppression command value 46". Note that the output suppression command value included in the output suppression command signal may be information specifying the upper limit value of the amount of power generated per unit time that can be output, rather than information specifying the upper limit value of the power generation output that can be output. In this case, the PCS 41 outputs the generated power of the solar panel 40 so that the amount of power generated per unit time does not exceed the specified upper limit value.

[0040] Note that the output suppression command signal may further include execution necessity information indicating whether or not execution of output suppression is necessary, in addition to the output suppression command value 46. The execution necessity information is information including a value indicating that execution of output suppression is necessary (for example, "1") or a value indicating that execution of output suppression is unnecessary (for example, "0"). For example, when a value (for example, "1") indicating that execution of output suppression is necessary is set as the execution necessity information, the PCS 41 outputs the generated power of the solar power generation device 4 so as not to exceed the upper limit value based on the output suppression command value 46. On the other hand, when a value (for example, "0") indicating that execution of output suppression is unnecessary is set as the execution necessity information, the PCS 41 does not suppress the generated power of the solar power generation device 4 regardless of the output suppression command value 46.

[0041] In addition to the function of controlling the output of the electric power generated by the solar panel 40 described above, the PCS 41 has a function of measuring the power generation output of the solar power generation device 4. For example, the PCS 41 measures the power generation output of the solar power generation device 4, that is, the electric power output from the PCS 41, every unit time (for example, every 30 seconds or every minute), and sets the measured value (power generation output) as the actual value 43 of the power generation amount. The PCS 41 stores the actual value 43 of the power generation amount in a storage device (not shown) inside the PCS 41, and transmits it to the power management device 7, the power generation amount estimation system 1, etc. via the network 6. The function of measuring the power generation output of the solar power generation device 4 may be realized by a voltage measurement device, a current measurement device, and a power measurement device installed separately from the PCS 41. The measurement unit may be not for each PCS, but the outputs of several PCSs or the entire solar power generation device may be measured together.

[0042] The transmission of the actual value 43 of the power generation amount by the PCS 41 may be performed, for example, periodically, or may be performed at any time in response to a request from an external information processing device (such as the power management device 7). For example, the PCS 41 periodically transmits the actual value 43 of the power generation amount of the solar power generation device 4 to the power management device 7, and the power management device 7 stores and manages the received actual value 43 of the power generation amount. Then, the power management device 7 may transmit the actual value 43 of the power generation amount of the solar power generation device 4 to the power generation amount estimation system 1 via the network 6 in response to a request from the power generation amount estimation system 1.

[0043] Note that the PCS 41 may store and transmit the output suppression command value 46 (and the necessity of execution information) as the output limit information 44 regarding the restriction of the power generation output of the solar power generation device 4 together with the actual value 43 of the power generation amount. For example, the PCS 41 may associate the actual value 43 of the power generation amount with the output suppression command value 46 (and the necessity of execution information) so that the measured time zones match, store them in a storage device (not shown) inside the PCS 41, and transmit them to the power management device 7.

[0044] The server 5 (weather DBS) is an information processing device (database server) that stores and manages information related to weather. The server 5 stores data on the actual values (estimated actual values) and estimated values (predicted values) of physical quantities related to weather for each unit of time and each region. The above weather data can be obtained from an external weather information providing service. Among the above weather data, the actual values of physical quantities related to weather can also be obtained by measurement with a weather observation device installed near the solar power generation device.

[0045] Here, examples of physical quantities related to weather include solar radiation amount, air temperature, wind speed, wind direction, and humidity. Hereinafter, information on the measured value or estimated value (predicted value) of the solar radiation amount will be denoted as "solar radiation amount 51", and information on the measured value or estimated value (predicted value) of the air temperature will be denoted as "air temperature 52".

[0046] In response to a request from the power generation amount estimation system 1, the server 5 transmits weather data 50 including physical quantities related to weather at a specified date and time in a specified region to the power generation amount estimation system 1 via the network 6. The weather data 50 includes, for example, the solar radiation amount 51 and the air temperature 52. The weather data 50 transmitted from the server 5 may include information on other physical quantities related to weather such as wind speed and wind direction. Here, the weather data 50 may be the actual values of the solar radiation amount 51 and the air temperature 52 measured at a specified date and time in a specified region, or may be the estimated values (predicted values) of the solar radiation amount 51 and the air temperature 52 at a specified date and time in a specified region. Further, the weather data 50 may be the actual values of the solar radiation amount 51 and the air temperature 52 measured at a specified date and time by a weather observation device installed near the solar power generation device.

[0047] The power management device 7 is a device that controls the power supply and demand of the power supply network including the solar power generation device 4 or the power supply and demand in the customer's premises and monitors the solar power generation device 4. The power management device 7 is, for example, an EMS (Energy Management System).

[0048] The power management device 7 controls the output suppression of the solar power generation device 4 to eliminate grid congestion, maintain the stability of the grid, or adjust the supply and demand in entrusted power transmission. Specifically, in response to a command for output suppression of the solar power generation device 4 from the outside, an output suppression command signal is generated and transmitted to the solar power generation device 4 and the power generation amount estimation system 1, so that the power supplied from the solar power generation device 4 to the grid is adjusted, grid congestion can be eliminated, the stability of the grid can be ensured, or the supply and demand in entrusted power transmission can be adjusted. The power management device 7 generates and outputs an output suppression command signal including the above-described output suppression command value 46. Note that the power management device 7 may generate the execution necessity information included in the output suppression command signal.

[0049] The power management device 7 acquires and manages information regarding the power generation performance by the solar power generation device 4. For example, the power management device 7 acquires the actual value 43 of the power generation amount measured per unit time in the solar power generation device 4 via the network 6 and stores it in the storage device in the power management device 7 as power generation performance data 42. For example, the power management device 7 may summarize the actual values of the power generation amount (power generation output) measured in units of 30 seconds or 1 minute in the solar power generation device 4 into actual values of the power generation amount per unit of 30 minutes or 1 hour and manage them as the power generation performance data 42.

[0050] Also, when there is output limit information 44 corresponding to the actual value 43 of the power generation amount, the power management device 7 may manage the output limit information 44 together with the actual value 43 of the power generation amount as the power generation performance data 42. For example, the power management device 7 may store the output limit information 44 including the PCS rated output value 45 of the solar power generation device 4 to be monitored together with the actual value 43 of the power generation amount. Also, when managing by summarizing the actual values of the power generation amount per unit time, the output limit value in the output limit information may be converted into the power amount per the same unit time and associated with the corresponding actual value of the power generation amount per unit time for management. This also applies to the PCS rated output value 45.

[0051] Further, for example, when the power management device 7 stores information regarding the history of output suppression command signals output within its own database, the power management device 7 may store output limit information 44 including the output suppression command value 46 or the value obtained by converting the output suppression command value 46 into the amount of power per unit time and the execution necessity information, associated with the actual power generation value 43 at the time when the actual power generation value 43 was measured, such that the measured time zones match. Further, the power management device 7 may acquire the weather data 50 from the server 5 and store the weather data 50, the actual power generation value 43, and the output limit information 44 or the value obtained by converting the output suppression command value into the amount of power per unit time, associated with each other such that the measured time zones match.

[0052] As shown in FIG. 1, the power generation amount estimation system 1 includes a model generation device 2 that generates a model, and a power generation amount estimation device 3 that estimates the power generation amount of the solar power generation device 4 corresponding to the input physical quantity related to the weather based on the above model.

[0053] The model generation device 2 and the power generation amount estimation device 3 are connected via a predetermined network, and data can be transmitted and received between them. For example, the model generation device 2 and the power generation amount estimation device 3 may be connected to each other by wired or wireless communication via a LAN (Local Area Network), or may be connected to each other via a wide area network (WAN) such as the Internet. When the model generation device 2 and the power generation amount estimation device 3 are individually operated, the model generation device 2 and the power generation amount estimation device 3 may not be connected to each other. The functions of the model generation device 2 and the power generation amount estimation device 3 may operate within the same server or personal computer. Also, the functions may be realized as the same program within the same server or personal computer.

[0054] First, the model generation device 2 will be described.

[0055] FIG. 2 is a diagram showing the functional block configuration of the model generation device 2 according to the embodiment. FIG. 3 is a diagram showing the hardware configuration of the model generation device 2 according to the embodiment.

[0056] The model generation device 2 is realized by an information processing device (computer) such as a server or a personal computer (PC), for example, generates analysis data according to an installed program for generating analysis data, and performs regression analysis or machine learning on the generated analysis data based on a predetermined algorithm to generate a model.

[0057] As shown in FIG. 2, the model generation device 2 includes a data acquisition unit 21, a determination unit 22, an analysis data generation unit 23, a model generation unit 24, and a storage unit 25 as functional blocks for generating a model. These functional blocks are realized by the hardware resources constituting the information processing device shown in FIG. 3 cooperating with software (various programs including the model generation program) installed in the information processing device.

[0058] Specifically, as shown in FIG. 3, the model generation device 2 includes, as hardware resources, for example, an arithmetic device 101, a storage device 102, an input device 103, an I / F (Interface) device 104, an output device 105, and a bus 106. Also, when the functions of the model generation device 2 and the power generation amount estimation device 3 operate in the same server or personal computer, or when the functions are realized as the same program in the same server or personal computer, the model generation device 2 and the power generation amount estimation device 3 also share the hardware resources in FIG. 3.

[0059] The arithmetic unit 101 is composed of a processor such as a CPU (Central Processing Unit) or a DSP (Digital Signal Processor). The storage device 102 has a storage area for storing a program 1021 for causing the arithmetic unit 101 to execute various data processes and data 1022 used in the data processes by the arithmetic unit 101, and is composed of, for example, a ROM (Read Only Memory), a RAM (Random Access Memory), an HDD, and a flash memory.

[0060] The program 1021 stored in the storage device 102 includes a model generation program for causing the computer to function as the model generation device 2, and is, for example, pre-installed in the storage device 102.

[0061] The data 1022 stored in the storage device 102 includes, for example, various parameters used in the data processes by the arithmetic unit 101, calculation results by the arithmetic unit 101, power generation performance data 42, weather data 50, analysis data 60, and the generated model 70.

[0062] Note that the program 1021 and the data 1022 may be distributed via a network, or may be written to a computer-readable storage medium (Non-transitory computer readable medium) such as a CD-ROM or a memory card and distributed.

[0063] The input device 103 is a device that detects the input of information from the outside, and examples thereof include a keyboard, a mouse, a pointing device, a button, or a touch panel. The I / F device 104 is a device that transmits and receives data to and from the outside, and is composed of, for example, a communication control circuit, an input / output port, an antenna, etc. for performing communication by wire or wirelessly.

[0064] The output device 105 is a device that outputs information and the like obtained by data processing by the arithmetic unit 101. Examples of the output device 105 include external storage devices such as SSDs and HDDs, and display devices such as LCDs (Liquid Crystal Displays) and organic ELs (Electro Luminescence). The bus 106 interconnects the arithmetic unit 101, the storage device 102, the input device 103, the I / F device 104, and the output device 105, enabling data transmission and reception between these devices.

[0065] The model generation device 2 causes the arithmetic unit 101 to execute calculations according to the program 1021 stored in the storage device 102, and controls the storage device 102, the input device 103, the I / F device 104, the output device 105, and the bus 106, whereby each functional block (data acquisition unit 21, determination unit 22, analysis data generation unit 23, model generation unit 24, and storage unit 25) of the model generation device 2 shown in FIG. 2 is realized.

[0066] Hereinafter, each functional block of the model generation device 2 will be described in detail.

[0067] The data acquisition unit 21 is a functional unit for acquiring data necessary for generating the model 70. The data acquisition unit 21 acquires necessary data by transmitting and receiving data to and from external devices such as the power management device 7, the solar power generation device 4, and the server 5 via the network 6.

[0068] The data acquisition unit 21 acquires data related to the power generation performance of the solar power generation device 4. Specifically, the data acquisition unit 21 acquires power generation performance data 42 including the actual power generation value 43 and the output limit information 44 of the solar power generation device 4, and weather data 50 corresponding to the power generation performance data 42.

[0069] For example, the data acquisition unit 21 acquires the actual power generation value 43 from the power management device 7 via the network 6 and stores it in the storage unit 25. Note that the data acquisition unit 21 may acquire the actual power generation value 43 from the solar power generation device 4.

[0070] Further, the data acquisition unit 21 acquires output limit information 44. For example, when the power management device 7 manages the output suppression command value 46 and the execution necessity information, the data acquisition unit 21 acquires, via the network 6, the output suppression command value 46 or the value obtained by converting the output suppression command value into the amount of electric power per unit time and the execution necessity information from the power management device 7, and stores them in the storage unit 25 as the output limit information 44.

[0071] Also, when the PCS rated output value 45 of the photovoltaic power generation device 4 which is the generation target of the model is stored in advance in the storage device 102 in the model generation device 2, the data acquisition unit 21 stores, in the storage unit 25 as the output limit information 44, the PCS rated output value 45 read from the storage device 102 or the value obtained by converting the rated output value into the amount of electric power per unit time. Alternatively, when the user operates an input device 103 such as a keyboard of the model generation device 2 to input the PCS rated output value 45 or the value obtained by converting the rated output value into the amount of electric power per unit time, the data acquisition unit 21 stores, in the storage unit 25 as the output limit information 44, the input PCS rated output value 45 or the value obtained by converting the rated output value into the amount of electric power per unit time.

[0072] Furthermore, the data acquisition unit 21 acquires from the server 5 meteorological data 50 including the actual value or the estimated value of the physical quantity related to the weather of the area where the photovoltaic power generation device 4 is installed in the time period when the actual generation amount value 43 of the photovoltaic power generation device 4 included in the generation performance data 42 is acquired (measured). In the present embodiment, it is assumed that the meteorological data 50 includes at least the solar irradiance 51.

[0073] The data acquisition unit 21 associates the acquired power generation performance data 42 including the actual power generation value 43 and the output limit information 44 with the weather data 50 for each time period when the actual power generation value 43 is measured, and stores them in the storage unit 25. Note that if the power generation performance data 42 and the weather data 50 have already been associated and stored for each time period in the power management device 7 and the model generation device 2 has acquired these data sets from the power management device 7, the model generation device 2 does not need to perform the process of associating the data for each time period.

[0074] Note that the method for acquiring the power generation performance data 42 and the weather data 50 is not limited to the above example. For example, the data acquisition unit 21 may read the power generation performance data 42 and the weather data 50 from a storage medium such as a memory card or an external storage device and store them in the storage unit 25.

[0075] The storage unit 25 is a functional unit for storing various data necessary for generating the model 70. For example, the storage unit 25 stores the power generation performance data 42, the weather data 50, the analysis data 60, and the generated model 70, etc. The storage unit 25 is configured to be accessible from an external information processing device, for example. For example, by the power generation estimation device 3 communicating with the model generation device 2, the power generation estimation device 3 can read and acquire the model 70 from the storage unit 25 of the model generation device 2.

[0076] The determination unit 22 is a functional unit for determining whether the actual power generation value 43 included in the power generation performance data 42 acquired by the data acquisition unit 21 is subject to a limit. In other words, the determination unit 22 determines whether the actual power generation value 43 was measured in a situation where the photovoltaic power generation device 4 was subject to an output limit.

[0077] The determination unit 22 determines whether the actual power generation value 43 included in the power generation performance data 42 is subject to a limit based on at least one of the output limit information 44 and the actual power generation value 43 included in the power generation performance data 42 acquired by the data acquisition unit 21. An example of the determination method is shown below.

[0078] For example, the determination unit 22 may determine whether the actual power generation value 43 is restricted based on the comparison result between the PCS rated output value 45 and the actual power generation value 43.

[0079] As described above, when the photovoltaic power generation device 4 is overloaded, the power generation amount of the photovoltaic power generation device 4 does not exceed the PCS rated output value 45 of the PCS 41 or the value obtained by converting the PCS rated output value 45 into the power generation amount per unit time. Therefore, for example, when the actual power generation value 43 is smaller than the value obtained by multiplying the PCS rated output value 45 or the value obtained by converting the PCS rated output value 45 into the power generation amount per unit time by a predetermined coefficient K1, the determination unit 22 determines that the actual power generation value 43 is not restricted. Here, 0 < K1 ≤ 1.

[0080] For example, when K1 = 0.9 and the PCS rated output value = 700 kW, the determination unit 22 determines whether the actual power generation value 43 (here, for example, the actual value of the power generation output) included in the power generation actual data 42 is smaller than 630 kW (= 700 kW × 0.9). When the actual power generation value 43 is smaller than 630 kW, the determination unit 22 determines that the actual power generation value 43 is not restricted. On the other hand, when the actual power generation value 43 is 630 kW or more, the determination unit 22 determines that the actual power generation value 43 is restricted.

[0081] In addition, the determination unit 22 may determine whether the actual power generation value 43 is restricted based on the presence or absence of the output suppression command value 46 in the power generation actual data 42. As described above, when the photovoltaic power generation device 4 performs output suppression according to the output suppression command signal, the power generation amount of the photovoltaic power generation device 4 is restricted by the output suppression command value 46 included in the output suppression command signal. Therefore, when the output suppression command value 46 is not included in the power generation actual data 42, the determination unit 22 determines that the actual power generation value 43 is not restricted.

[0082] Further, when the actual power generation value 43 of the power generation amount included in the power generation performance data 42 is smaller than the value obtained by multiplying the power generation amount after suppression based on the output suppression command value 46 included in the power generation performance data 42 by a predetermined coefficient K2, the determination unit 22 may determine that the actual power generation value 43 of the power generation amount is not subject to output limitation. Here, 0 < K2 ≤ 1.

[0083] For example, when K2 = 0.8, the output suppression command value = the power generation output after suppression = 600 kW, the determination unit 22 determines whether the actual power generation value 43 (here, for example, the actual value of the power generation output) is smaller than 480 kW (= 600 kW × 0.8). When the actual power generation value 43 is smaller than 480 kW, the determination unit 22 determines that the actual power generation value 43 is not subject to limitation. On the other hand, when the actual power generation value 43 included in the power generation performance data 42 is 480 W or more, the determination unit 22 determines that the actual power generation value 43 is subject to limitation.

[0084] In this way, the determination unit 22 determines whether the actual power generation value 43 included in the power generation performance data 42 is subject to output limitation based on at least one of the output limitation information 44 and the actual power generation value 43 included in the power generation performance data 42 acquired by the data acquisition unit 21.

[0085] Note that the determination unit 22 may determine whether the actual power generation value 43 is limited by using the weather data 50 corresponding to the power generation performance data 42 instead of the output limitation information 44 and the actual power generation value 43 included in the power generation performance data 42.

[0086] Specifically, when the solar radiation amount 51 corresponding to the power generation performance data 42 is smaller than a predetermined reference value, the determination unit 22 may determine that the actual power generation value 43 is not limited. Here, the reference value of the solar radiation amount may be determined as follows.

[0087] For example, when the solar radiation amount is 800 W / m 2 and the power generation output of the solar panel 40 exceeds the PCS rated output value, a lower value than 800 W / m 2 such as "700 W / m2 ” is used as the reference value of the solar radiation amount. In this case, when the solar radiation amount 51 corresponding to the actual value 43 of the power generation amount (here, for example, the actual value of the power generation output) is smaller than the reference value “700 W / m 2 ”, the determination unit 22 determines that the actual value 43 of the power generation amount is not restricted. On the other hand, when the solar radiation amount 51 corresponding to the actual value 43 of the power generation amount is equal to or greater than the reference value “700 W / m 2 ”, the determination unit 22 determines that the actual value 43 of the power generation amount is restricted.

[0088] Also, for example, when the power generation amount of the solar panel 40 is restricted by the output suppression command value 46, several sample data (data pairs combining the power generation actual data 42 and the corresponding weather data 50) are used in advance to calculate the correlation between the output suppression command value 46 and the solar radiation amount, and the solar radiation amount when the power generation amount of the solar panel 40 is restricted is estimated. Then, the estimated value of the solar radiation amount is used as the reference value of the solar radiation amount. The determination unit 22 determines that the actual value 43 of the power generation amount is not restricted when the solar radiation amount 51 corresponding to the actual value 43 of the power generation amount is smaller than the above reference value, and determines that the actual value 43 of the power generation amount is restricted when the solar radiation amount 51 corresponding to the actual value 43 of the power generation amount is equal to or greater than the above reference value. At this time, when the model 70 is already stored in the storage unit 25 instead of the correlation between the output suppression command value 46 and the solar radiation amount for estimating the solar radiation amount when the power generation amount of the solar panel 40 is restricted, the model 70 or a simplified version thereof can also be used.

[0089] The analysis data generation unit 23 is a functional unit that generates the analysis data 60 necessary for generating the model 70 for causing the computer to function to calculate the power generation amount of the solar power generation device 4 based on the input physical quantities related to the weather.

[0090] The analysis data generation unit 23 generates analysis data 60 by associating the actual power generation amount value 43 of the solar power generation device 4 with the meteorological data 50. Specifically, the analysis data generation unit 23 selects the power generation performance data 42 to be used as analysis data based on the determination result by the determination unit 22, and generates the analysis data 60 based on the selected power generation performance data 42 and the meteorological data 50 corresponding to the power generation performance data 42.

[0091] More specifically, the analysis data generation unit 23 selects the power generation performance data 42 determined by the determination unit 22 that the actual power generation amount value 43 is not restricted from among the plurality of power generation performance data 42 acquired by the data acquisition unit 21. The analysis data generation unit 23 generates analysis data 60 in which the actual power generation amount value 43 included in the selected power generation performance data 42 is associated with the physical quantities related to the weather (for example, solar radiation amount 51 and temperature 52) included in the meteorological data 50 corresponding to the selected power generation performance data 42.

[0092] For example, when generating data for regression analysis as the analysis data 60, the analysis data generation unit 23 generates analysis data 60 in which the solar radiation amount 51 (or the solar radiation amount 51 and the temperature 52) is used as an explanatory variable and the actual power generation amount value 43 determined not to be restricted is used as a target variable, and stores it in the storage unit 25.

[0093] Also, for example, when generating learning data (analysis data) for machine learning as the analysis data 60, the analysis data generation unit 23 generates the analysis data 60 by labeling the solar radiation amount 51 (or the solar radiation amount 51 and the temperature 52) when the actual power generation amount value 43 determined not to be restricted is measured with the actual power generation amount value 43 as correct answer information, and stores it in the storage unit 25.

[0094] For each piece of power generation performance data 42 acquired by the data acquisition unit 21, the analysis data generation unit 23 sequentially generates analysis data 60 by the above-described method and stores it in the storage unit 25.

[0095] The model generation unit 24 is a functional unit that generates a model 70.

[0096] Here, as described above, the model 70 is a program generated by regression analysis or machine learning based on a predetermined algorithm (technique) for causing a computer to function so as to calculate the power generation amount of the photovoltaic power generation device 4 based on the physical quantities related to the input weather.

[0097] Here, as an algorithm based on regression analysis, at least one of an algorithm based on simple regression analysis, multiple regression analysis, linear regression analysis, polynomial regression analysis, and a combined regression analysis thereof, and an algorithm based on other statistical analysis can be exemplified.

[0098] Further, as a predetermined algorithm based on machine learning, at least one of an algorithm of a Boltzmann machine, a support vector machine, a Bayesian network, sparse regression, a decision tree, statistical estimation using a random forest, a neural network (such as a CNN (Convolutional Neural Network)), reinforcement learning, and deep learning can be exemplified.

[0099] When the model 70 is a regression model based on regression analysis, for example, a function having the solar radiation amount as an explanatory variable and the power generation amount of the photovoltaic power generation device 4 as an objective variable, or a function having the solar radiation amount and the temperature as explanatory variables and the power generation amount of the photovoltaic power generation device 4 as an objective variable can be used as the model 70. The above functions include coefficients of each term and the like.

[0100] In addition, when the meteorological data 50 includes wind force and wind direction, the model 70 may be a regression model that uses, as explanatory variables, the solar radiation amount, the temperature, and the values obtained by quantifying or numericalizing the wind force and wind direction, and uses the power generation amount of the solar power generation device 4 as the objective variable (method of numericalizing the wind direction).

[0101] Also, when the model 70 is a learned model based on machine learning, the model 70 is a program for estimating the power generation amount for causing an information processing device (computer) to function so as to perform an operation based on predetermined learned parameters on the input physical quantities related to the weather and output a value obtained by quantifying the power generation amount of the solar power generation device 4. Specifically, the model 70 includes learned parameters and an inference program. The inference program is a program for incorporating the learned parameters and outputting a certain power generation amount of the solar power generation device 4 with respect to the input physical quantities related to the weather. The learned parameters are parameters that are mechanically adjusted so as to calculate the power generation amount of the solar power generation device 4 under the conditions based on the input physical quantities related to the weather by using the physical quantities related to the weather as inputs to the inference program. For example, in the case of a neural network, the learned parameters are variables such as weighting coefficients.

[0102] The model generation unit 24 generates the model 70 by performing regression analysis or machine learning on the plurality of analysis data 60 generated by the analysis data generation unit 23 based on the above-described predetermined algorithm.

[0103] When the predetermined algorithm is simple regression analysis, the model generation unit 24 calculates, for example, the coefficients of each term of a regression model that uses the solar radiation amount as an explanatory variable and the power generation amount of the solar power generation device 4 as an objective variable by the least squares method, and generates the model 70.

[0104] When the specified algorithm is a neural network, the model generation unit 24 sequentially updates the learned parameters so that the error between the estimated power generation amount calculated based on the physical quantities related to the weather used for learning (solar radiation amount 51 and temperature 52) and the actual power generation amount as the correct data corresponding to the physical quantities related to the weather used for learning becomes small, for example, by the error backpropagation method, and generates the model 70.

[0105] The model 70 generated by the model generation unit 24 is stored in the storage unit 25 and registered in the storage unit 36 of the power generation amount estimation device 3 described later.

[0106] As described above, various models can be adopted as the model 70. In this embodiment, as an example, the model 70 will be described as being represented by a regression model in which the solar radiation amount is an explanatory variable and the power generation amount of the photovoltaic power generation device 4 is an objective variable.

[0107] Note that the instruction to execute the processing related to the above-described model generation by the model generation device 2 and the input of other various information to the model generation device 2 may be realized by the user operating an input device 103 such as a keyboard of the model generation device 2, or may be realized by the user operating an information processing terminal (PC, tablet terminal, smartphone) etc. as an input / output interface and transmitting it from the information processing terminal to the model generation device 2 via the network 6. Further, the model generation device 2 may automatically execute the processing related to model generation at a fixed time of a day or at a time defined as a schedule such as at regular time intervals to update the model. Further, after receiving a command for starting the execution of the processing related to model generation from another information system (server etc.) for the model generation device 2, in the another information system, a command for starting the execution of the processing related to model generation may be automatically transmitted at a time defined as a schedule.

[0108] Next, the power generation amount estimation device 3 will be described.

[0109] FIG. 4 is a diagram showing a functional block configuration of the power generation amount estimation device 3 according to the embodiment. FIG. 5 is a diagram showing a hardware configuration of the power generation amount estimation device 3 according to the embodiment.

[0110] The power generation amount estimation device 3 is, for example, an information processing device (computer) such as a PC (personal computer) or a server, and estimates the power generation amount of the photovoltaic power generation device 4 under the conditions based on the input meteorological data 50 according to the installed power generation amount estimation program (model 70).

[0111] As shown in FIG. 4, the power generation amount estimation device 3 has an input data acquisition unit 31, an estimation unit 32, and a storage unit 36 as functional blocks for estimating the power generation amount of the photovoltaic power generation device 4 based on the input physical quantities related to the weather. These functional blocks are realized by the cooperation of the hardware resources constituting the information processing device shown in FIG. 5 and the software (various programs including the power generation amount estimation program) installed in the information processing device.

[0112] As shown in FIG. 5, the power generation amount estimation device 3 includes, as hardware resources, an arithmetic device 101, a storage device 102, an input device 103, an I / F (Interface) device 104, an output device 105, and a bus 106. These components are the same as the hardware resources constituting the above-described model generation device 2. Also, when the functions of the model generation device 2 and the power generation amount estimation device 3 operate in the same server or personal computer, or when the functions are realized as the same program in the same server or personal computer, the model generation device 2 and the power generation amount estimation device 3 also share the hardware resources of FIG. 5.

[0113] The storage device 102 stores a program 1023 for causing the arithmetic device 101 to execute various data processes as the power generation amount estimation device 3, and data 1024 such as parameters and calculation results used in the data process by the arithmetic device 101.

[0114] Here, the program 1023 includes a model 70 and the like as a program for estimating the power generation amount for causing a computer (processor) to function as the power generation amount estimation device 3, and is, for example, pre-installed in the storage device 102. Further, the data 1024 includes weather data 50, output limit information 44, calculation results of the power generation amount, and the like.

[0115] Note that the program 1023 and the data 1024 may be distributed via a network, or may be written to a computer-readable storage medium (Non-transitory computer readable medium) such as a CD-ROM or a memory card and distributed.

[0116] The power generation amount estimation device 3 controls the storage device 102, the input device 103, the I / F device 104, the output device 105, and the bus 106 by the arithmetic device 101 executing an arithmetic operation according to the program 1023 stored in the storage device 102, whereby each functional block (input data acquisition unit 31, estimation unit 32, and storage unit 36) of the power generation amount estimation device 3 shown in FIG. 4 is realized.

[0117] Hereinafter, each functional block of the power generation amount estimation device 3 will be described in detail.

[0118] The input data acquisition unit 31 is a functional unit that receives instructions from a user or the like and acquires various data necessary for estimating the power generation amount of the solar power generation device 4.

[0119] As described above, the input data acquisition unit 31 transmits and receives data to and from the model generation device 2 in the power generation amount estimation system 1 via a predetermined network such as a LAN or a WAN. Further, the input data acquisition unit 31 can transmit and receive data to and from external information processing devices such as the solar power generation device 4 and the server 5 via the network 6.

[0120] For example, the user instructs the power generation amount estimation device 3 to execute a process of estimating the power generation amount of the solar power generation device 4 (hereinafter, also referred to as the "power generation amount estimation process"). Here, the instruction to execute the power generation amount estimation process and the input of other various information to the power generation amount estimation device 3 may be realized by the user operating an input device 103 such as a keyboard of the power generation amount estimation device 3, or may be realized by the user operating an information processing terminal (PC, tablet terminal, smartphone) as an input / output interface and transmitting it from the information processing terminal to the power generation amount estimation device 3 via the network 6. Also, the power generation amount estimation device 3 may be configured to automatically execute the power generation amount estimation process at a fixed time of day or at a time defined as a schedule such as a certain time interval. Further, after the power generation amount estimation device 3 receives a command to start executing the power generation amount estimation process from another information system (server, etc.), in the other information system, a command to start executing the power generation amount estimation process may be automatically transmitted at a time defined as a schedule.

[0121] At this time, in addition to the instruction to execute the power generation amount estimation process, the user inputs information regarding the conditions for estimating the power generation amount to the power generation amount estimation device 3. For example, the user inputs meteorological data 50A that designates the meteorological conditions at the time of estimating the power generation amount as information regarding the conditions for estimating the power generation amount to the power generation amount estimation device 3. The meteorological data 50A at least includes information on the estimated value (designated value) of a physical quantity related to the weather at the time of estimating the power generation amount, for example, information on the estimated value (designated value) of the solar irradiance (also referred to as the "solar irradiance 51A"). Further, the meteorological data 50A may include information on the estimated value (designated value) of the temperature (hereinafter, also referred to as the "temperature 52A"), or may include information on the wind direction and wind speed and humidity information, etc., similar to the meteorological data 50. That is, the information included in the meteorological data 50A is determined according to the model 70 to be used.

[0122] Also, when considering the output limit of the photovoltaic power generation device 4 during the estimation of the power generation amount, the user operates the input device 103 or the information processing terminal to input the output limit information 44 to the power generation amount estimation device 3 as information regarding the conditions during the estimation of the power generation amount, in addition to the meteorological data 50A. For example, when estimating the power generation amount of the photovoltaic power generation device 4 in a situation where output suppression is being performed according to an output suppression signal, the user operates the input device 103 or the information processing terminal to input the assumed output suppression command value 46 to the power generation amount estimation device 3 as the output limit information 44. In this case, in addition to the output suppression command value 46, the above-described execution necessity information may be input to the power generation amount estimation device 3 as the output limit information 44.

[0123] Also, when the photovoltaic power generation device 4 is overloaded, the user operates the input device 103 or the information processing terminal to input the PCS rated output value 45 to the power generation amount estimation device 3 as the output limit information 44. Note that the PCS rated output value 45 may be stored in advance in the storage unit 36 together with the model 70 as information regarding the photovoltaic power generation device 4.

[0124] Note that when the power generation amount estimation device 3 stores the models 70 corresponding to the plurality of photovoltaic power generation devices 4 respectively, the user operates the input device 103 or the information processing terminal to also input identification information for designating the photovoltaic power generation device 4 to be the estimation target of the power generation amount to the power generation amount estimation device 3.

[0125] When an instruction to execute the above-described power generation amount estimation process is input to the power generation amount estimation device 3, the input data acquisition unit 31 notifies the estimation unit 32 that it has received the instruction. Also, as described above, the input data acquisition unit 31 acquires the meteorological data 50A and the output limit information 44 input to the power generation amount estimation device 3 as input data 80 and stores them in the storage unit 36.

[0126] Note that instead of directly inputting the meteorological data 50A into the power generation amount estimation device 3, the user may input, together with an instruction to execute the power generation amount estimation process, information specifying the area where the photovoltaic power generation device 4 for which the power generation amount is to be estimated is installed or the photovoltaic power generation device 4, and information specifying the time period for which the power generation amount is to be estimated, into the power generation amount estimation device 3. In this case, the input data acquisition unit 31 accesses the server 5 in response to the input of the above instruction and information, and acquires estimated values of physical quantities related to the weather (such as solar radiation amount and temperature) in the area and time period where the photovoltaic power generation device 4 specified by the user is installed from the server 5 or the power management device 7, and may store them in the storage unit 36 as the meteorological data 50A.

[0127] The storage unit 36 is a functional unit that stores various data necessary for the power generation amount estimation process of the photovoltaic power generation device 4 by the power generation amount estimation device 3. For example, in addition to the meteorological data 50A and the output limit information 44 as the input data 80 described above, the storage unit 36 stores the model 70 and the estimation result of the power generation amount, etc.

[0128] As described above, the model 70 is a program for causing an information processing device (computer) to function so as to output a value obtained by quantifying the power generation amount based on the input physical quantities related to the weather. For example, the power generation amount estimation device 3 acquires the model 70 from the model generation device 2 by communicating with the model generation device 2, and stores it in the storage unit 36 in advance.

[0129] The estimation unit 32 is a functional unit that estimates the power generation amount of the photovoltaic power generation device 4 under the specified conditions based on the model 70. The estimation unit 32 determines whether the power generation amount of the photovoltaic power generation device 4 is restricted under the conditions specified by the input data 80 (meteorological data 50A and output limit information 44), and generates an estimation result 92 of the power generation amount of the photovoltaic power generation device 4 based on the determination result.

[0130] Specifically, when the power generation amount of the solar power generation device 4 is not restricted, the estimation unit 32 outputs the estimated value 90 of the power generation amount calculated based on the model 70 as the estimation result 92 of the power generation amount of the solar power generation device 4. On the other hand, when the power generation amount of the solar power generation device 4 is restricted, the estimated value 90 of the power generation amount calculated based on the model 70 is corrected to be smaller, and the corrected value 91 of the power generation amount is output as the estimation result 92 of the power generation amount of the solar power generation device 4.

[0131] For example, as shown in FIG. 4, the estimation unit 32 includes a power generation amount calculation unit 33, a determination unit 34, and an output unit 35.

[0132] The power generation amount calculation unit 33 is a functional unit that calculates the estimated value 90 of the power generation amount of the solar power generation device 4. The power generation amount calculation unit 33 calculates the power generation amount of the solar power generation device 4 corresponding to the meteorological data 50A (for example, solar irradiance 51A and temperature 52) of the input data 80 based on the model 70 stored in the storage unit 36. For example, when receiving a notification from the input data acquisition unit 31 that an instruction to execute the power generation amount estimation process has been input, the power generation amount calculation unit 33 reads out the meteorological data 50A of the input data 80 specified by the above instruction from the storage unit 36. The power generation amount calculation unit 33 calculates the power generation amount of the solar power generation device 4 corresponding to the read meteorological data 50A using the model 70. That is, the power generation amount calculation unit 33 stores the value calculated from the model 70 by inputting the read meteorological data 50A into the model 70 in the storage unit 36 as the estimated value 90 of the power generation amount of the solar power generation device 4.

[0133] The determination unit 34 is a functional unit that determines whether the power generation amount of the solar power generation device 4 is restricted under the condition of estimating the power generation amount of the solar power generation device 4.

[0134] When the PCS rated output value 45 of the solar power generation device 4 is included in the input data 80, the determination unit 34 determines whether the power generation amount of the solar power generation device 4 is restricted based on the comparison result between the PCS rated output value 45 or the value obtained by converting the rated output value into the generated electric energy per unit time and the estimated value 90 of the power generation amount calculated by the power generation amount calculation unit 33.

[0135] For example, when the estimated power generation amount 90 calculated by the power generation amount calculation unit 33 is smaller than the PCS rated output value 45 or the value obtained by converting the rated output value into the power generation amount per unit time, the determination unit 34 determines that the power generation amount of the solar power generation device 4 is not restricted. On the other hand, when the estimated power generation amount 90 calculated by the power generation amount calculation unit 33 is larger than the PCS rated output value 45 or the value obtained by converting the rated output value into the power generation amount per unit time, the determination unit 34 determines that the power generation amount of the solar power generation device 4 is restricted.

[0136] Also, when the output suppression command value 46 is not included in the input data 80, the determination unit 34 determines that the power generation amount of the solar power generation device 4 is not restricted. On the other hand, when the output suppression command value 46 is included in the input data 80, the determination unit 34 may determine that the power generation amount of the solar power generation device 4 is restricted. In addition, when execution necessity information is included in the input data 80 in addition to the output suppression command value 46, the determination unit 34 may determine whether or not the power generation amount of the solar power generation device 4 is restricted based on the execution necessity information. For example, when a value (e.g., "1") indicating that execution of output suppression is necessary is set as the execution necessity information, the determination unit 34 determines that the power generation amount of the solar power generation device 4 is restricted. On the other hand, when a value (e.g., "0") indicating that execution of output suppression is unnecessary is set as the execution necessity information, the determination unit 34 determines that the power generation amount of the solar power generation device 4 is not restricted regardless of the output suppression command value 46.

[0137] The output unit 35 is a functional unit that generates and outputs an estimated result 92 of the power generation amount of the solar power generation device 4 based on the estimated power generation amount 90 calculated by the power generation amount calculation unit 33 and the determination result of the determination unit 34.

[0138] When the determination unit 34 determines that the power generation amount of the solar power generation device 4 is not restricted, the output unit 35 outputs the estimated power generation amount 90 calculated by the power generation amount calculation unit 33 as the estimated result 92 of the power generation amount.

[0139] On the other hand, when it is determined by the determination unit 34 that the power generation amount of the photovoltaic power generation device 4 is restricted, the output unit 35 corrects the estimated value 90 of the power generation amount calculated by the power generation amount calculation unit 33 to be smaller, and outputs the corrected value 91 of the power generation amount as the estimated result 92 of the power generation amount.

[0140] Specifically, when it is determined that the power generation amount of the photovoltaic power generation device 4 is restricted because the estimated value 90 of the power generation amount is larger than the PCS rated output value 45 or the value obtained by converting the rated output value into the power generation amount per unit time, the output unit 35 corrects the estimated value 90 of the power generation amount so as not to exceed the PCS rated output value 45 or the value obtained by converting the rated output value into the power generation amount per unit time, and outputs the corrected value (corrected value 91 of the power generation amount) as the estimated result 92 of the power generation amount.

[0141] More specifically, the output unit 35 corrects the estimated value 90 of the power generation amount to be a value corresponding to the PCS rated output value 45 or the value obtained by converting the rated output value into the power generation amount per unit time, and outputs the corrected value as the estimated result 92 of the power generation amount. For example, when the PCS rated output value 45 is "700 kW" and the estimated value 90 of the power generation output is "850 kW", the output unit 35 sets "700 kW" corresponding to the PCS rated output value 45 as the corrected value 91 of the power generation output. In this case, in addition to the corrected value 91 of the power generation amount, the output unit 35 may include information indicating that the power generation amount is restricted (for example, information indicating that the power generation output is restricted due to the PCS rated output) in the estimated result 92 of the power generation amount and output it.

[0142] Also, when it is determined that the power generation amount of the photovoltaic power generation device 4 is restricted because the output suppression command value 46 is included in the input data 80, the output unit 35 corrects the estimated value 90 of the power generation amount so as not to exceed the power generation amount after suppression based on the output suppression command value 46, and outputs the corrected value (corrected value 91 of the power generation amount) as the estimated result 92 of the power generation amount.

[0143] Specifically, the output unit 35 corrects the estimated power generation amount 90 so as to be a value corresponding to the power generation amount after suppression based on the output suppression command value 46, and outputs the corrected value as the estimated result 92 of the power generation amount. For example, when the output suppression command value 46 is "500 kW" and the estimated value 90 of the power generation output is "850 kW", the output unit 35 sets "500 kW", which corresponds to the power generation output after suppression based on the output suppression command value 46, as the corrected value 91 of the power generation output. At this time, in addition to the corrected value 91 of the power generation amount, the output unit 35 may output information indicating that the power generation amount is limited (for example, information indicating that the power generation amount is limited due to the output suppression command signal) included in the estimated result 92 of the power generation amount.

[0144] Note that even if the output suppression command value 46 is included in the input data 80, the power generation amount estimation device 3 may output the estimated result of the power generation amount assuming that there is no output suppression of the photovoltaic power generation device 4. For example, the output unit 35 may output "estimated power generation amount 90" as the estimated result 92 of the power generation amount as the power generation amount assuming that there is no output suppression of the photovoltaic power generation device 4, together with the corrected value 91 of the power generation amount based on the output suppression command value 46.

[0145] Here, whether or not to consider the output suppression of the photovoltaic power generation device 4 in the power generation amount estimation process may be switchable according to an instruction from the user. For example, when the user inputs an instruction indicating that the output suppression of the photovoltaic power generation device 4 is not considered together with the input data 80 to the power generation amount estimation device 3, the output unit 35 does not perform the above-described power generation amount correction process, and outputs "estimated power generation amount 90" calculated based on the model 70 as the estimated result 92 of the power generation amount. On the other hand, when the user inputs an instruction indicating that the output suppression of the photovoltaic power generation device 4 is to be considered together with the input data 80 to the power generation amount estimation device 3, the output unit 35 executes the above-described power generation amount correction process and outputs the corrected value 91 of the power generation amount based on the output suppression command value 46 as the estimated result 92 of the power generation amount. Note that whether or not to consider the output suppression of the photovoltaic power generation device 4 may be realized by inputting the above-described execution necessity information (1 or 0) to the power generation amount estimation device 3 together with the output suppression command value 46.

[0146] Next, the processing flow in the power generation amount estimation system 1 will be described.

[0147] FIG. 6 is a flowchart showing the processing flow by the power generation amount estimation system 1 according to the embodiment.

[0148] First, as shown in FIG. 6, in the power generation amount estimation system 1, the model generation device 2 generates analysis data 60 (step S1).

[0149] FIG. 7 is a flowchart showing the detailed flow of the generation process of the analysis data 60 (step S1) according to the embodiment.

[0150] In step S1, first, the model generation device 2 acquires data related to the power generation performance of the photovoltaic power generation device 4 (step S11). Specifically, the model generation device 2 acquires the power generation performance data 42 (the actual power generation amount value 43 and the output limit information 44) of the photovoltaic power generation device 4 and the weather data 50 of the weather conditions when the power generation performance data 42 is measured by the above-described method.

[0151] Next, the model generation device 2 determines whether there was a limit on the power generation amount at the time of measurement of the actual power generation amount value 43 of the acquired power generation performance data 42 (step S12). Specifically, the determination unit 22 determines whether there was a limit on the power generation amount based on the output limit information 44 by the above-described method.

[0152] When it is determined that there was a limit on the actual power generation amount value 43 (step S12: YES), the model generation device 2 does not use the power generation performance data 42 including the actual power generation amount value 43 for generating the analysis data and returns to step S11.

[0153] On the other hand, when it is determined that there is no limit to the actual power generation value 43 (step S12: NO), the model generation device 2 generates analysis data 60 based on the actual power generation value 43 (step S13). Specifically, the analysis data generation unit 23 associates the actual power generation value 43 determined by the determination unit 22 to have no limit by the above-described method with the weather data 50 acquired in step S11 so that the measured time zones match, thereby generating the analysis data 60 and storing it in the storage unit 25.

[0154] Next, the model generation device 2 determines whether or not the required number of analysis data 60 has been generated (step S14). For example, the number of analysis data 60 required to generate the model 70 is preset in the model generation device 2, and the model generation device 2 counts the number of generations each time the analysis data 60 is generated. Then, the model generation device 2 determines whether or not the required number of analysis data 60 has been generated by determining whether or not the counted number of generations has reached the preset number of data.

[0155] If the required number of analysis data 60 has not been generated (step S14: No), the model generation device 2 returns to step S11, selects new power generation performance data 42, and repeats the process of generating the analysis data 60 (steps S11 to S14).

[0156] On the other hand, when the required number of analysis data 60 has been generated (step S14: YES), the model generation device 2 ends the generation process of the analysis data 60 (step S1).

[0157] As shown in FIG. 6, after step S1 is completed, the model generation device 2 generates a model 70 using the plurality of analysis data 60 generated in step S1 (step S2). Specifically, the model generation unit 24 creates the model 70 by performing regression analysis or machine learning on the plurality of analysis data 60 created in step S1 based on a predetermined algorithm by the above-described method.

[0158] When sufficient accuracy is obtained for the model 70, the model 70 is stored in the storage unit 25. Note that the determination as to whether sufficient accuracy has been obtained for the model 70 may be made as follows. For example, an arithmetic unit having the same function as the estimation unit 32 of the power generation amount estimation device 3 is provided in the model generation device 2, and some sample data is input to the model 70 in the arithmetic unit to calculate the power generation amount. Then, when the average value of the errors or error rates of the power generation amounts calculated by the arithmetic unit in the samples is equal to or less than a predetermined threshold value, it is determined that sufficient detection accuracy has been obtained for the model 70, and when the average value of the errors or error rates of the power generation amounts calculated by the arithmetic unit in the samples is greater than the predetermined threshold value, the regression analysis or machine learning of the analysis data 60 may be continued.

[0159] Next, the model 70 generated by the model generation device 2 in step S2 is registered in the power generation amount estimation device 3 (step S3). For example, in response to an operation of the model generation device 2 or the power generation amount estimation device 3 by the user, the model generation device 2 transmits the model 70 stored in the storage unit 25 to the power generation amount estimation device 3, and the power generation amount estimation device 3 stores the received model 70 in the storage unit 36. Also, when the functions of the model generation device 2 and the power generation amount estimation device 3 operate in the same server or personal computer, or when the functions are realized as the same program in the same server or personal computer, the above-described transmission and reception of the model 70 are executed as the transfer of data within the same server or personal computer, or the transfer of data between programs or within the same program.

[0160] Next, the power generation amount estimation device 3 executes a power generation amount estimation process (step S4). For example, when the user operates the power generation amount estimation device 3 to instruct the execution of the estimation of the power generation amount of the solar power generation device 4 under specified conditions, the power generation amount estimation device 3 starts the power generation amount estimation process in response to the instruction.

[0161] FIG. 8 is a flowchart showing the flow of the power generation amount estimation process (step S4) according to the embodiment.

[0162] In step S4, first, the power generation amount estimation device 3 acquires input data 80 including weather data 50A related to the conditions specified by the user (step S41). For example, when the weather data 50A is input to the power generation amount estimation device 3 together with an instruction to execute the estimation of the power generation amount, the input data acquisition unit 31 stores the weather data 50A in the storage unit 36 as the input data 80. At this time, if the output limit information 44 has been input to the power generation amount estimation device 3, the input data acquisition unit 31 stores the output limit information 44 in the storage unit 36 as the input data 80 together with the weather data 50A.

[0163] Next, the power generation amount estimation device 3 calculates an estimated value of the power generation amount of the photovoltaic power generation device 4 based on the model 70 and the input data 80 acquired in step S41 (step S42). Specifically, as described above, the power generation amount calculation unit 33 in the estimation unit 32 inputs the weather data 50A (such as the solar radiation amount 51A and the temperature 52A, etc.) included in the input data 80 acquired in step S41 to the model 70, thereby calculating an estimated value 90 of the power generation amount of the photovoltaic power generation device 4.

[0164] Next, the power generation amount estimation device 3 determines whether or not the power generation amount of the photovoltaic power generation device 4 is restricted under the conditions on which the estimated value 90 of the power generation amount calculated in step S42 is based (step S43).

[0165] For example, the determination unit 34 determines whether or not the power generation amount of the photovoltaic power generation device 4 is restricted based on the comparison result between the PCS rated output value 45 of the photovoltaic power generation device 4 or the value obtained by converting the rated output value into the power generation amount per unit time and the estimated value 90 of the power generation amount calculated in step S42 by the method described above. Also, for example, the determination unit 34 determines whether or not the output suppression command value 46 is included in the input data 80 (or the set value of the execution necessity information) by the method described above, thereby determining whether or not the power generation amount of the photovoltaic power generation device 4 is restricted.

[0166] When the power generation amount estimation device 3 determines that the power generation amount of the photovoltaic power generation device 4 is not restricted under the conditions on which the estimated value 90 of the power generation amount calculated in step S42 is based (step S43: NO), the power generation amount estimation device 3 outputs the estimated value 90 of the power generation amount calculated in step S42 as the power generation amount estimation result 92 (step S46).

[0167] On the other hand, when the power generation amount estimation device 3 determines that the power generation amount of the photovoltaic power generation device 4 is restricted under the conditions on which the estimated value 90 of the power generation amount calculated in step S42 is based (step S43: YES), the power generation amount estimation device 3 corrects the estimated value 90 of the power generation amount calculated in step S42 based on the output restriction information 44 (step S44). For example, when it is determined in step S43 that the power generation amount of the photovoltaic power generation device 4 is restricted because the estimated value 90 of the power generation amount is larger than the PCS rated output value 45 included in the input data 80 or the value obtained by converting the rated output value into the power generation amount per unit time (or the PCS rated output value 45 stored in the storage unit 36 or the value obtained by converting the rated output value into the power generation amount per unit time), as described above, the output unit 35 corrects the estimated value 90 of the power generation amount so as to be a value corresponding to the PCS rated output value 45 or the value obtained by converting the PCS rated output value 45 into the power generation amount per unit time, and stores the corrected value (= the PCS rated output value 45 or the value obtained by converting the PCS rated output value 45 into the power generation amount per unit time) in the storage unit 36 as the power generation amount correction value 91.

[0168] Also, for example, when it is determined in step S43 that the power generation amount of the photovoltaic power generation device 4 is restricted because the output suppression command value 46 is included in the input data 80 acquired in step S41, the output unit 35 corrects the estimated value 90 of the power generation amount so as to be a value corresponding to the power generation amount after suppression based on the output suppression command value 46, and stores the corrected value (= the power generation amount after suppression based on the output suppression command value 46) in the storage unit 36 as the power generation amount correction value 91.

[0169] Next, the power generation amount estimation device 3 outputs the power generation amount correction value 91 calculated in step S44 as the power generation amount estimation result 92 (step S45).

[0170] Through the above processing procedure, in the power generation amount estimation system 1, the generation of the model 70 and the estimation of the power generation amount of the photovoltaic power generation device 4 using the model 70 are performed.

[0171] In addition, in the above flowchart, the power generation amount estimation process (step S4) does not necessarily need to be continuously performed following the process related to model generation (steps S1 to S3). That is, the process related to model generation (steps S1 to S3) and the power generation amount estimation process (step S4) may be executed at different timings. For example, before executing the power generation amount estimation process (step S4), it is sufficient that the model 70 is generated and registered in the power generation amount estimation device 3.

[0172] The effects of the power generation amount estimation system 1 according to the embodiment will be described. First, the model generated by the model generation device 2 will be described.

[0173] FIG. 9 is a diagram for explaining the model 70 generated by the model generation device 2 according to the embodiment.

[0174] In FIG. 9, the horizontal axis represents the solar irradiance [W / m 2 , and the vertical axis represents the power generation output [kW] of the photovoltaic power generation device 4. In FIG. 9, the plots of circles (O) represent data pairs (solar irradiance and power generation output) when the photovoltaic power generation device 4 is not overloaded, the plots of crosses (X) represent data pairs (solar irradiance and power generation output) in which the power generation output is limited due to the overload of the photovoltaic power generation device 4 (PCS rated output value = 700 kW), and the plots of triangles represent data pairs (solar irradiance and power generation output) selected as the analysis data 60 in consideration of the PCS rated output value by the method described above.

[0175] Also, in FIG. 9, the solid line graph indicated by reference numeral 501 is a model (quadratic regression model: Y = -2E-05X) generated by learning the plots of the above circles (O) (data pairs when the photovoltaic power generation device 4 is not overloaded) 2It represents (+0.92X - 1.6719) (Comparative Example). The dashed line graph indicated by reference numeral 502 is a model (quadratic regression model: Y = -0.0004X 2 + 1.1964X - 31.693) generated by learning the plot of the cross marks (X) (data pairs where the power generation output was limited due to overloading of the solar power generation device 4) (Comparative Example). The dashed line graph indicated by reference numeral 503 is a model (quadratic regression model: Y = -0.0002X 2 + 0.9947X - 7.8986) (Model 70 according to the embodiment of the present application).

[0176] As shown in FIG. 9, the power generation output of the actual solar power generation device 4 is limited by the PCS rated output value (700 kW) when the solar irradiance is 800 W / m 2 or more. Therefore, as shown in FIG. 9, the estimated value of the power generation output calculated using the graph 501, which is a model obtained by performing regression analysis using data pairs (solar irradiance and power generation output) when the solar power generation device 4 is not overloaded, or the graph 502, which is a model obtained by performing regression analysis using data pairs (solar irradiance and power generation output) when the solar power generation device 4 is overloaded, has a large error from the actual value of the actual power generation amount (power generation output) near a solar irradiance of 800 W / m 2 or more.

[0177] For example, in the case of the graph 502, which is a model obtained by performing regression analysis using data pairs (solar irradiance and power generation output) when the solar power generation device 4 is overloaded, the root mean squared error (RMSE) between the estimated value of the power generation output based on the graph 502 and the actual value of the actual power generation output is 4.2%. In particular, the RMSE at a solar irradiance of 600 W / m 2 or more is 3.8%.

[0178] Therefore, the model generation device 2 according to the embodiment excludes data pairs in which the power generation output is restricted by the PCS rated output value from the analysis data and performs regression analysis (or machine learning) by the method described above. For example, as shown in FIG. 9, data pairs with a power generation output of 690 kW or less are adopted as analysis data for performing regression analysis. As a result, a graph 503 is obtained as the model.

[0179] According to the graph 503, the RMSE between the calculated estimated value of the power generation output and the measured value of the actual power generation output is 3.8%. In particular, the RMSE at a solar irradiance of 600 W / m 2 or more is 3.1%. Thus, it is understood that according to the model 70 generated by the model generation device 2 according to the embodiment, the error can be reduced more than when learning is performed using data pairs when the solar power generation device 4 is overloaded.

[0180] As described above, the model generation device 2 according to the embodiment determines whether or not the power generation amount is restricted based on at least one of the output restriction information and the power generation amount included in the power generation performance data of the solar power generation device 4, and performs regression analysis or machine learning on the power generation performance data determined not to be restricted in power generation amount, thereby generating a model for estimating the power generation amount based on physical quantities related to weather. That is, by selecting data pairs to be adopted as the analysis data 60 based on the PCS rated output value 45, a more accurate model can be generated. Also, as described above, by selecting data pairs to be adopted as the analysis data 60 based on the output suppression command value 46 of the output suppression command signal, a more accurate model can be generated.

[0181] Also, as described above, the model generation device 2 determines whether or not the power generation amount included in the power generation performance data is restricted based on the comparison result between the PCS rated output value 45 or the value obtained by converting the PCS rated output value 45 into the power generation amount per unit time and the actual power generation amount value 43. According to this, it becomes easy to detect the actual power generation amount value 43 in which the power generation amount is restricted by the rated output of the PCS.

[0182] Also, as described above, when the actual power generation value 43 is smaller than the PCS rated output value 45 or the value obtained by multiplying the PCS rated output value 45 by a predetermined coefficient K1 (0 < K1 ≤ 1) in terms of the power generation amount per unit time, the model generation device 2 may determine that the power generation amount is not restricted. According to this, by adjusting the coefficient K1, it is possible to surely exclude the actual power generation value 43 that is likely to be restricted, which contributes to the generation of a more accurate model.

[0183] Also, as described above, when the output suppression command value 46 is not included in the power generation actual data 42, the model generation device 2 determines that the power generation amount is not restricted. According to this, it becomes easy to detect the actual power generation value 43 restricted by the output suppression command signal.

[0184] Also, as described above, when the actual power generation value 43 is smaller than the value obtained by multiplying the power generation amount after suppression based on the output suppression command value 46 by a predetermined coefficient K2 (0 < K2 ≤ 1), the model generation device 2 determines that the power generation amount is not restricted. According to this, by adjusting the value of the coefficient K2, it is possible to surely exclude the actual power generation value 43 that is likely to be restricted by the output suppression command signal, which contributes to the generation of a more accurate model.

[0185] Also, as described above, when the solar radiation amount included in the weather data 50 corresponding to the power generation actual data 42 is smaller than a predetermined reference value, the model generation device 2 may determine that the power generation amount is not restricted. According to this, it becomes easy to detect the actual power generation value 43 restricted by the PCS rated output value 45 or the output suppression command value 46 due to a high solar radiation amount.

[0186] Next, the estimation result of the power generation amount using the model 70 by the power generation amount estimation device 3 will be described. FIGS. 10A, 10B, 11A, and 11B show an example of the estimation result of the power generation output by the power generation amount estimation device 3.

[0187] FIG. 10A is a diagram for explaining an estimation result of the power generation output when the output suppression of the photovoltaic power generation device 4 is in effect by the power generation amount estimation device 3 according to the embodiment.

[0188] FIG. 10B is a diagram for explaining an estimation result of the power generation output when it is assumed that the output suppression of the photovoltaic power generation device 4 is not in effect by the power generation amount estimation device 3 according to the embodiment.

[0189] In FIGS. 10A and 10B, the horizontal axis represents time (day), and the vertical axis represents the power generation output [kW] of the photovoltaic power generation device 4.

[0190] In FIG. 10A, the model generation device 2 extracts data pairs (solar irradiance and power generation output) regarding the power generation performance when the output suppression of the photovoltaic power generation device 4 by the output suppression signal is not in effect as analysis data 60, generates a model 70 by performing a regression analysis on these analysis data 60, and the power generation amount estimation device 3 corrects the estimated value of the power generation output calculated by the model 70 based on the output suppression command value, and the estimation result of the power generation output in this case is shown.

[0191] Specifically, the graph indicated by the solid line of reference numeral 601 represents the temporal change of the actual value (measured value) of the power generation output of the photovoltaic power generation device 4. The graph indicated by the dotted line of reference numeral 602 represents the temporal change of the output suppression command value of the power generation output of the photovoltaic power generation device 4. The graph indicated by the alternate long and short dash line of reference numeral 603 represents the temporal change of the estimated value of the power generation output (with output suppression) calculated by the power generation amount estimation device 3 according to the embodiment, considering the output suppression command value.

[0192] In FIG. 10B, the model generation device 2 generates a model 70 by the same method as in the case of FIG. 10A, and the power generation amount estimation device 3 shows the result of estimating the power generation output using the model 70 assuming no output suppression.

[0193] Specifically, in FIG. 10B, the graph indicated by the solid line of reference numeral 601A represents the temporal change in the actual value (measured value) of the power generation output of the photovoltaic power generation device 4. The graph indicated by the dotted line of reference numeral 602A represents the temporal change in the output suppression command value of the power generation output of the photovoltaic power generation device 4. The graph indicated by the dashed-dotted line of reference numeral 603A represents the temporal change in the estimated value (without output suppression) of the power generation output when it is assumed that there is no output suppression, calculated by the power generation amount estimation device 3 according to the embodiment.

[0194] As understood from FIG. 10A, according to the power generation amount estimation system 1 according to the embodiment, the estimated value 90 of the power generation amount is calculated based on the model 70 generated by performing regression analysis (or machine learning) on the data pair (solar radiation amount and power generation amount) regarding the actual power generation when the output suppression is not in effect, and the estimated value 90 of the power generation amount is corrected by the output suppression command value 46, so that it is possible to more accurately calculate the estimated result 92 of the power generation output or the power generation amount per unit time when the output suppression of the power generation amount of the photovoltaic power generation device 4 is performed.

[0195] Also, as understood from FIG. 10B, according to the power generation amount estimation system 1 according to the embodiment, it is also possible to calculate in the same manner the estimated value 90 of the power generation amount when it is assumed that the output suppression of the power generation amount of the photovoltaic power generation device 4 is not performed.

[0196] FIG. 11A is a diagram for explaining another estimated result of the power generation output of the photovoltaic power generation device 4 when the output suppression of the power generation amount estimation device 3 according to the embodiment is in effect.

[0197] FIG. 11B is a diagram for explaining another estimated result of the power generation output of the photovoltaic power generation device 4 when it is assumed that the output suppression of the power generation amount estimation device 3 according to the embodiment is not in effect.

[0198] In FIGS. 11A and 11B, the horizontal axis represents time (days), and the vertical axis represents the power generation output [kW] of the photovoltaic power generation device 4.

[0199] In Fig. 11A, when the output suppression of the solar power generation device 4 by the output suppression signal is not in effect, the model generation device 2 extracts, as analysis data 60, data pairs (solar irradiance and power generation output) regarding the power generation performance in which the solar irradiance is smaller than the reference value from among the data pairs, and generates a model 70 by performing a regression analysis on these analysis data 60. The power generation output estimation result is shown when the power generation output estimator 3 corrects the estimated value of the power generation output calculated using the model 70 based on the output suppression command value.

[0200] Specifically, the graph indicated by the solid line of reference numeral 701 represents the temporal change in the actual value (measured value) of the power generation output of the solar power generation device 4. The graph indicated by the dotted line of reference numeral 602 represents the temporal change in the output suppression command value of the power generation output of the solar power generation device 4. The graph indicated by the dashed-dotted line of reference numeral 603 represents the temporal change in the estimated value of the power generation output (with output suppression) calculated by the power generation output estimator 3 according to the embodiment, taking into account the output suppression command value.

[0201] In Fig. 11B, the model generation device 2 generates a model 70 by the same method as in Fig. 11A, and the result of estimating the power generation output using the model 70 assuming no output suppression is shown by the power generation output estimator 3.

[0202] Specifically, in Fig. 11B, the graph indicated by the solid line of reference numeral 701A represents the temporal change in the actual value (measured value) of the power generation output of the solar power generation device 4. The graph indicated by the dotted line of reference numeral 702A represents the temporal change in the output suppression command value of the power generation output of the solar power generation device 4. The graph indicated by the dashed-dotted line of reference numeral 703A represents the temporal change in the estimated value of the power generation output (without output suppression) calculated by the power generation output estimator 3 according to the embodiment, assuming no output suppression.

[0203] As can be understood from FIG. 11A, according to the power generation amount estimation system 1 according to the embodiment, an estimated value 90 of the power generation amount is calculated based on a model 70 generated by performing regression analysis (or machine learning) on a data pair (solar radiation amount and power generation amount) regarding a power generation record in which the solar radiation amount is smaller than a reference value, and by correcting the estimated value 90 of the power generation amount with an output suppression command value 46, it becomes possible to more accurately calculate an estimation result 92 of the power generation amount when the output of the solar power generation device 4 is suppressed. Further, as can be understood from FIG. 11B, according to the power generation amount estimation system 1 according to the embodiment, it is also possible to calculate an estimated value 90 of the power generation amount in the case where it is assumed that the output of the solar power generation device 4 is not suppressed.

[0204] As described above, the power generation amount estimation device 3 according to the embodiment determines whether or not the power generation amount of the solar power generation device 4 is restricted under the condition of estimating the power generation amount of the solar power generation device 4. When it is determined that the power generation amount is not restricted, based on the above model, an estimated value of the power generation amount corresponding to the physical quantity related to the weather is calculated and output as an estimation result of the power generation amount of the solar power generation device 4. When it is determined that the power generation amount is restricted, the estimated value of the power generation amount calculated based on the model is corrected to be smaller, and the corrected value is output as an estimation result of the power generation amount.

[0205] According to this, as described above, under the condition that the power generation amount of the solar power generation device 4 is restricted by the PCS rated output value 45 or the output suppression command value 46, it becomes possible to estimate the power generation amount in consideration of the PCS rated output value 45 or the output suppression command value 46.

[0206] Specifically, as described above, when the power generation amount estimation device 3 determines that the power generation amount is restricted because the estimated value of the power generation amount calculated based on the model 70 is greater than the PCS rated output value 45 or the value obtained by converting the rated output value into the power generation amount per unit time, the estimated value of the power generation amount is corrected so as not to exceed the PCS rated output value 45 or the value obtained by converting the rated output value into the power generation amount per unit time, and the corrected value is output as the estimation result of the power generation amount of the photovoltaic power generation device. According to this, it becomes possible to accurately estimate the power generation amount under the condition that the power generation output of the photovoltaic power generation device 4 is restricted by the PCS rated output value 45.

[0207] Also, as described above, when the power generation amount estimation device 3 determines that the power generation amount of the photovoltaic power generation device 4 is restricted because the output suppression command value 46 is included in the input data 80, the estimated value of the power generation amount is corrected so as not to exceed the power generation amount after suppression based on the output suppression command value 46, and the corrected value is output as the estimation result of the power generation amount of the photovoltaic power generation device. According to this, it becomes possible to accurately estimate the power generation amount under the condition that the power generation output of the photovoltaic power generation device 4 is restricted by the output suppression command value 46 (see FIGS. 10A and 11A).

[0208] Note that according to the power generation amount estimation device 3, not only can it predict the power generation amount at a certain future time point, but it can also be used for diagnosing the presence or absence of a failure in the solar power generation device 4. For example, the model generation device 2 generates a model 70 based on the analysis data 60 (power generation performance data 42 and meteorological data 50) for the past one year, and applies the model 70 to the power generation amount estimation device 3. A diagnostic unit is further provided in the power generation amount estimation device 3. First, by inputting the most recent meteorological data 50A into the estimation unit 32, an estimated value of the power generation amount estimated from the most recent meteorological data 50A is calculated. Next, the diagnostic unit compares the estimated value of the power generation amount estimated from the most recent meteorological data 50A with the actual value 43 of the power generation amount corresponding to the most recent meteorological data 50A. When the estimated value of the power generation amount and the actual value 43 of the power generation amount match (for example, when the error is within ±10% etc.), the diagnostic unit determines that the solar power generation device 4 is normal, and when the estimated value of the power generation amount and the actual value 43 of the power generation amount do not match (for example, when the error is more than ±10% etc.), the diagnostic unit determines that the solar power generation device 4 is abnormal. By estimating the power generation amount using the power generation amount estimation device 3 in this way, it becomes possible to more accurately diagnose the presence or absence of a failure in the solar power generation device 4. Further, the diagnostic unit may diagnose the presence or absence of a failure in the solar power generation device 4 by comparing the integrated value of the estimated value of the power generation amount over a certain period with the integrated value of the actual value 43 of the power generation amount over the same certain period.

[0209] ≪Expansion of Embodiment≫ As described above, the invention made by the present inventor has been specifically described based on the embodiments. However, it goes without saying that the present invention is not limited thereto and can be variously modified without departing from the gist thereof.

[0210] For example, although the case where the model generation device 2 and the power generation amount estimation device 3 are respectively realized by separate information processing devices has been exemplified, the present invention is not limited thereto, and the model generation device 2 and the power generation amount estimation device 3 may be realized by one information processing device, or may be integrated as one system composed of a plurality of information processing devices. Further, the model generation device 2 and the power generation amount estimation device 3 do not necessarily have to be provided within the same location (site), and each may be provided at a different location. Further, the model generation device 2 may be realized by a plurality of information processing devices connected to each other via a network or the like. Similarly, the power generation amount estimation device 3 may also be realized by a plurality of information processing devices connected to each other via a network or the like.

[0211] Further, the above-described flowchart shows an example for explaining the operation, and is not limited thereto. That is, the steps shown in each figure of the flowchart are specific examples and are not limited to this flow. For example, the order of some processes may be changed, other processes may be inserted between each process, or some processes may be performed in parallel.

Explanation of Reference Numerals

[0212] 1... Power generation amount estimation system, 2... Model generation device, 3... Power generation amount estimation device, 4... Photovoltaic power generation device, 5... Server (weather DBS), 6... Network, 7... Power management device, 21... Data acquisition unit, 22... Determination unit, 23... Analysis data generation unit, 24... Model generation unit, 25... Storage unit, 31... Input data acquisition unit, 32... Estimation unit, 33... Power generation amount calculation unit, 34... Determination unit, 35... Output unit, 36... Storage unit, 40... Solar panel, 41... Power conversion device, 42... Power generation performance data, 43... Actual value of power generation amount, 44... Output limit information, 45... PCS rated output value, 46... Output suppression command value, 50, 50A... Weather data, 51, 51A... Solar radiation amount, 52, 52A... Temperature, 60... Analysis data, 70... Model, 80... Input data, 90... Estimated value of power generation amount, 91... Correction value of power generation amount, 92... Estimation result of power generation amount.

Claims

1. A data acquisition unit that acquires power generation performance data including an actual value of power generation amount indicating the magnitude of power output from a photovoltaic power generation device and output limit information regarding the limit of the power generation amount, and a physical quantity related to weather corresponding to the power generation performance data; A determination unit that determines whether or not the actual value of the power generation amount is subject to a limit based on at least one of the output limit information and the actual value of the power generation amount included in the power generation performance data acquired by the data acquisition unit; An analysis data generation unit that generates analysis data in which the actual value of the power generation amount determined by the determination unit not to be subject to a limit is associated with the physical quantity related to the weather; A storage unit that stores the analysis data generated by the analysis data generation unit; A model generation unit that generates a model for causing a computer to function so as to calculate the power generation amount based on the input physical quantity related to the weather by performing regression analysis or machine learning on the analysis data stored in the storage unit based on a predetermined algorithm. A model generation device.

2. In the model generation device according to Claim 1, The photovoltaic power generation device includes a solar panel and a power conversion device that converts the power generated by the solar panel into AC power and outputs it. The output limit information includes a rated output value indicating the magnitude of the rated output of the power conversion device. The determination unit determines whether or not the actual value of the power generation amount is subject to a limit based on a comparison result between the rated output value and the actual value of the power generation amount or a comparison result between a value obtained by converting the rated output value into the power generation amount per unit time and the actual value of the power generation amount. A model generation device.

3. In the model generation device according to Claim 2, The determination unit determines that the actual value of the power generation amount is not subject to a limit when the actual value of the power generation amount is smaller than a value obtained by multiplying the rated output value or a value obtained by converting the rated output value into the power generation amount per unit time by a predetermined coefficient K1 (0 < K1 ≤ 1). A model generation device.

4. In the model generation device according to any one of Claims 1 to 3, The determination unit determines that the actual value of the power generation amount is not subject to a limit when an output suppression command value specifying the suppression amount of the power generation amount output from the photovoltaic power generation device is not included in the power generation performance data as the output limit information. A model generation device.

5. In the model generation device according to Claim 4, When the actual value of the power generation amount is smaller than the value obtained by multiplying the power generation amount after suppression based on the output suppression command value by a predetermined coefficient K2 (0 < K2 ≤ 1), the determination unit determines that the actual value of the power generation amount is not restricted. Model generation device.

6. In the model generation device according to claim 4, the output limit information further includes execution necessity information indicating the necessity of executing output suppression, and when the execution necessity information indicates that it is not necessary to execute output suppression, the determination unit determines that the actual value of the power generation amount is not restricted regardless of the output suppression command value. Model generation device.

7. In the model generation device according to claim 1, the physical quantity related to the weather includes the solar radiation amount, and when the solar radiation amount is smaller than a predetermined reference value, the determination unit determines that the actual value of the power generation amount is not restricted. Model generation device.

8. An input data acquisition unit that acquires input data including a physical quantity related to the weather, a storage unit that stores a model for causing a computer to function so as to calculate the power generation amount based on the input physical quantity related to the weather, the model being generated by performing regression analysis or machine learning on a plurality of analysis data generated by associating the physical quantity related to the weather with the magnitude of the power output from the photovoltaic power generation device based on a predetermined algorithm, and an estimation unit that estimates the power generation amount corresponding to the physical quantity related to the weather included in the input data acquired by the input data acquisition unit based on the model stored in the storage unit, wherein the estimation unit includes a power generation amount calculation unit that calculates an estimated value of the power generation amount by inputting the physical quantity related to the weather included in the input data into the model, a determination unit that determines whether or not the power generation amount is restricted under the condition of estimating the power generation amount of the photovoltaic power generation device, and an output unit that outputs the estimated value of the power generation amount calculated based on the model as the estimation result of the power generation amount when the determination unit determines that the power generation amount is not restricted, and outputs a value obtained by correcting the estimated value of the power generation amount calculated based on the model to be smaller as the estimation result of the power generation amount when the determination unit determines that the power generation amount is restricted. Power generation amount estimation device.

9. In the power generation amount estimation device according to claim 8, The photovoltaic power generation device includes a solar panel and a power conversion device that converts the power generated by the solar panel into AC power and outputs it. The storage unit stores a rated output value indicating the magnitude of the rated output of the power conversion device. When the estimated power generation amount is smaller than the rated output value or the value obtained by converting the rated output value into the power generation amount per unit time, the determination unit determines that the power generation amount is not restricted under the condition of estimating the power generation amount of the photovoltaic power generation device. When the estimated power generation amount is larger than the rated output value or the value obtained by converting the rated output value into the power generation amount per unit time, the determination unit determines that the power generation amount is restricted under the condition of estimating the power generation amount of the photovoltaic power generation device. When it is determined that the power generation amount is restricted because the estimated power generation amount is larger than the rated output value or the value obtained by converting the rated output value into the power generation amount per unit time, the output unit corrects the estimated power generation amount so as not to exceed the rated output value or the value obtained by converting the rated output value into the power generation amount per unit time, and outputs the corrected value as the estimated result of the power generation amount of the photovoltaic power generation device. Power generation amount estimation device.

10. In the power generation amount estimation device according to claim 8 or 9, When the output suppression command value for designating the suppression amount of the power generation amount output from the photovoltaic power generation device is not included in the input data, the determination unit determines that the power generation amount is not restricted under the condition of estimating the power generation amount of the photovoltaic power generation device. When the output suppression command value is included in the input data, the determination unit determines that the power generation amount is restricted under the condition of estimating the power generation amount of the photovoltaic power generation device. When it is determined that the power generation amount is restricted because the output suppression command value is included in the input data, the output unit corrects the estimated power generation amount so as not to exceed the suppressed power generation amount based on the output suppression command value, and outputs the corrected value as the estimated result of the power generation amount of the photovoltaic power generation device. Power generation amount estimation device.

11. A first step of acquiring power generation performance data including an actual value of the power generation amount indicating the magnitude of the power output from the photovoltaic power generation device and output restriction information regarding the restriction of the power generation amount, and a physical quantity related to the weather corresponding to the power generation performance data. A second step of determining whether the actual power generation value is restricted based on at least one of the output restriction information and the actual value of the power generation amount included in the power generation performance data obtained in the first step; A third step of generating analysis data in which the actual value of the power generation amount determined not to be restricted in the second step is associated with the physical quantity related to the weather; A fourth step of storing the analysis data generated in the third step; A fifth step of generating a model for causing a computer to function so as to calculate the power generation amount based on the input physical quantity related to the weather by performing regression analysis or machine learning on the analysis data stored in the fourth step based on a predetermined algorithm. An information processing method.

12. A program for causing the computer to execute the first to fifth steps in the information processing method according to claim 11. A program.

13. A first step of obtaining input data including a physical quantity related to the weather; Based on a model generated by performing regression analysis or machine learning on a plurality of pieces of analysis data generated by associating the physical quantity related to the weather with the magnitude of the power output from the photovoltaic power generation device, the power generation amount corresponding to the physical quantity related to the weather included in the input data obtained in the first step is estimated. A second step, The second step includes: A third step of calculating an estimated value of the power generation amount by inputting the physical quantity related to the weather included in the input data into the model; A fourth step of determining whether the power generation amount is restricted under the condition of estimating the power generation amount of the photovoltaic power generation device; A fifth step of outputting, as the estimated result of the power generation amount, the estimated value of the power generation amount calculated based on the model when it is determined in the fourth step that the power generation amount is not restricted; A sixth step of outputting, as the estimated result of the power generation amount, a value obtained by correcting the estimated value of the power generation amount calculated based on the model to be smaller when it is determined in the fourth step that the power generation amount is restricted. An information processing method.

14. A program for causing a computer to execute the first step to the sixth step in the information processing method according to claim 13 .

Images