Method and apparatus for predicting solar-generated electric power, computer device and storage medium

Abstract

SOLUTION: A method for predicting solar-generated electric power includes: correcting a first solar-generated electric power predicting model based on a correction factor into generation of a second solar-generated electric power predicting model; determining whether or not environment temperature information of a photovoltaic generation module meets preset request information; if not meeting the preset request information, settling a cell temperature average value of the photovoltaic generation module and a first relationship representative of a relationship between a cell temperature of the photovoltaic generation module waiting for prediction, an environment temperature and an irradiance; making a correction on the second photovoltaic power predicting model based on the cell temperature average value and the first relationship into generation of a third photovoltaic power predicting model; and perform prediction on the photovoltaic power value of the photovoltaic generation module with utilization of the third photovoltaic power predicting model.EFFECT: The prediction accuracy of solar-generated electric power can be greatly improved by correcting a solar-generated electric power predicting model a plurality of number of times.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to the field of computer technology, and more particularly to a method, an apparatus, a computer device, and a storage medium for predicting photovoltaic power generation. [Background technology]

[0002] The use of renewable energy is one of the important means to solve the problems of global energy shortages and environmental pollution, and photovoltaic power generation is the means with the most promising development prospects and convenient conditions in the current development of new energy. Photovoltaic power generation is affected by meteorological factors, and the process by which photovoltaic modules convert light energy into electrical energy is also affected by the device itself, resulting in strong randomness and variability in photovoltaic power generation. With the increase in grid-connected capacity, the randomness and variability of photovoltaic power generation poses a major threat to the safe and stable operation of the power grid. Therefore, accurate prediction of photovoltaic power generation is of great significance to improving the safety and stability of the power grid.

[0003] In conventional technology, neural network models are typically used to predict solar power generation. While neural network models have the advantages of high prediction accuracy and low data requirements, the low interpretability of the models makes it difficult to intuitively analyze the causes of power prediction results at the business level, making it difficult to achieve cross-connection with actual production and operation scenarios. Therefore, there is an urgent need for a solar power generation prediction method that can interpret model content and results while ensuring prediction accuracy, thereby realizing cross-connection between solar power generation prediction results and actual production and operation scenarios. Summary of the Invention

[0004] The technical problem that the present invention aims to solve is that the present invention provides a photovoltaic power generation power prediction method, device, computer device, and storage medium, and solves the following problems: the physical model prediction accuracy in related technologies is low, the causes of power prediction results cannot be intuitively analyzed from the business level, and it is difficult to achieve linkage with actual production and operation scenes.

[0005] In order to solve the above technical problems, the technical solutions adopted by the present invention are as follows: A photovoltaic power generation power forecasting method, including: The first solar power generation power forecasting model is modified based on the modification coefficient to generate a second solar power generation power forecasting model, which is constructed based on the cell temperature and irradiance of the solar power generation module. It is determined whether the environmental temperature information of the photovoltaic power generation module to be predicted matches the preset requirement information, and if it does not match the preset requirement information, a first relationship is determined with the average battery temperature of the photovoltaic power generation module to be predicted, and the first relationship represents the relationship between the battery temperature of the photovoltaic power generation module to be predicted, the environmental temperature, and the irradiance. The second photovoltaic power generation prediction model is corrected based on the average battery temperature and the first relationship to generate a third photovoltaic power generation prediction model. The third photovoltaic power generation prediction model is used to predict the photovoltaic power generation value of the photovoltaic power generation module awaiting prediction. According to the weather type of the photovoltaic power generation module to be predicted, a corresponding target weather correction coefficient is determined from a plurality of weather correction coefficients.

[0006] Preferably, the solar power generation value is adjusted based on the target weather correction factor to obtain a target solar power generation value.

[0007] Preferably, before determining the corresponding target weather correction factor from the plurality of weather correction factors, the method includes: The solar power generation module for prediction acquires a plurality of sets of data for different weather types in advance, where each set of data includes device capacity, irradiance value, battery temperature value and actual power generation amount for a corresponding weather type. The device capacity, irradiance value, battery temperature value and actual power generation amount for the corresponding weather type are input into a correction coefficient prediction model, and the weather correction coefficient for the corresponding weather type is output.

[0008] Preferably, the correction coefficients include a dust influence coefficient, a light attenuation loss coefficient, and a surface reflection loss coefficient. Correction is performed on the first solar power generation power forecasting model based on the correction coefficients to generate a second solar power generation power forecasting model, which includes:

number

[0009] Preferably, the second solar power generation power prediction model is corrected based on the battery temperature average value and the first relationship to generate a third solar power generation power prediction model, which includes: A first correction is made to the second photovoltaic power generation prediction model based on the average battery temperature. A second correction is performed on the first correction result based on the first relationship to generate a third photovoltaic power generation forecasting model.

[0010] Preferably, the battery temperature average value of the photovoltaic power generation module to be predicted is determined, and includes: All battery temperature values ​​within a preset time period of the solar power generation module waiting for prediction are obtained. From all the battery temperature values, battery temperature values ​​having irradiances smaller than a preset threshold value are excluded. The remaining battery temperature values ​​determine the average battery temperature value. A first correction is made to the second solar power generation power prediction model based on the battery temperature average value, and includes:

number

[0011] Preferably, the first relationship is a linear fitting function. The first relationship of the solar photovoltaic module to be predicted is determined, and includes: The prediction-waiting solar power generation module acquires a plurality of battery temperature values ​​within a preset time interval, a plurality of environmental temperature values ​​respectively corresponding to the plurality of battery temperature values, and a plurality of irradiance values ​​respectively corresponding to the plurality of battery temperature values. A linear fitting function is obtained based on the plurality of battery temperature values, the plurality of environmental temperature values, and the plurality of irradiances, which is used to indicate that the difference between the battery temperature values ​​and the environmental temperature values ​​exhibits a linear relationship with the irradiance.

number

number

[0012] Preferably, the solar power generation power prediction device includes a first model correction module, which is used to correct the first solar power generation power prediction model based on the correction coefficient, and generate a second solar power generation power prediction model, which is configured based on the battery temperature and irradiance of the solar power generation module. The information determining module is included, and is used to determine whether the environmental temperature information of the photovoltaic power generation module waiting for prediction matches the preset required information. A data determination module is included, which is used to determine a first relationship with the average battery temperature of the solar power generation module to be predicted if the predetermined required information does not match, and the first relationship represents the relationship between the battery temperature of the solar power generation module to be predicted, the ambient temperature and the irradiance. A second model correction module is included, which is used to correct the second solar power generation power prediction model based on the battery temperature average value and the first relationship, and generate a third solar power generation power prediction model. A photovoltaic power generation value prediction module is included, which is used to predict the photovoltaic power generation value of the photovoltaic power generation module to be predicted using the third photovoltaic power generation power prediction model.

[0013] Preferably, the computer device includes a memory and a processor, the memory and the processor being communicatively connected to each other, computer instructions stored in the memory, and the processor executing the computer instructions to perform any one of the photovoltaic power forecasting methods.

[0014] Preferably, a computer readable storage medium has computer instructions stored thereon, said computer instructions being used to cause a computer to perform any one of said photovoltaic power forecasting methods.

[0015] The photovoltaic power generation power forecasting method, device, computer device and storage medium provided by the present invention have the following beneficial effects. 1. The photovoltaic power forecasting method provided by the present invention selects a corresponding target weather correction coefficient according to the weather type, and adjusts the photovoltaic power value based on the target weather correction coefficient to obtain the target photovoltaic power value, thereby fully considering the effects of different weather factors on the photovoltaic module, and thereby greatly improving the accuracy of photovoltaic power forecasting. 2. The photovoltaic power generation power prediction method provided by the present invention inputs the device capacity, irradiance value, battery temperature value and actual power generation amount of the corresponding weather type into a correction coefficient prediction model to output the weather correction coefficient of the corresponding weather type. By determining the weather correction coefficient of different weather types using the correction coefficient prediction model and related data information of different weather types, the photovoltaic power generation status of the photovoltaic power generation module in different weather conditions can be effectively predicted. 3. The photovoltaic power prediction method provided by the present invention modifies the first photovoltaic power prediction model using a correction coefficient, and takes into account various factors affecting photovoltaic power, namely, the dust influence coefficient, the initial light attenuation loss coefficient, and the surface reflection loss coefficient, thereby improving the accuracy of photovoltaic power prediction. 4. The photovoltaic power generation prediction method provided by the present invention combines two methods: averaging the measured battery temperature of the photovoltaic power generation module and linear fitting the battery temperature and irradiance of the photovoltaic power generation module, which can jointly improve the accuracy of the photovoltaic power generation prediction results of the power plant. 5. The photovoltaic power forecasting method provided by the present invention eliminates from all battery temperature values ​​those whose irradiance is lower than a preset threshold, thereby fully considering the problem that the accuracy of the photovoltaic power forecast obtained under the above conditions is relatively low due to the occurrence of climate differences caused by different geographical locations and the dynamic changes in the module temperature of the photovoltaic power generation module caused by the environmental changes, and achieves the technical effect of improving the forecast accuracy. Furthermore, by averaging the measured battery temperatures of the photovoltaic power generation module, the accuracy of the photovoltaic power forecast results of the power plant can be further improved. 6. The photovoltaic power prediction method provided by the present invention can more accurately reflect the relationship between battery temperature value and irradiance by constructing a linear fitting function that represents the correlation between battery temperature value and irradiance, and can fully consider that irradiance and battery temperature are the main factors affecting photovoltaic power generation, thereby further improving the accuracy of photovoltaic power prediction. The present invention will now be further described in conjunction with the accompanying drawings and examples. [Brief explanation of the drawings]

[0016] [Figure 1] 1 is a flowchart of a photovoltaic power generation power prediction method according to an embodiment of the present invention. [Figure 2] 10 is a flowchart of another photovoltaic power generation power prediction method according to an embodiment of the present invention. [Figure 3] 10 is a flowchart of another photovoltaic power generation power prediction method according to an embodiment of the present invention. [Figure 4] 10 is a flowchart of another method for predicting photovoltaic power generation according to an embodiment of the present invention. [Figure 5] 10 is a flowchart of another method for predicting photovoltaic power generation according to an embodiment of the present invention. [Figure 6] FIG. 1 is a schematic diagram illustrating the use of a linear regression model to output weather correction factors in an embodiment of the present invention. [Figure 7] FIG. 2 is a schematic diagram of a model representation classified by weather in an embodiment of the present invention. [Figure 8] FIG. 1 is a schematic diagram illustrating the use of a linear regression model to output weather correction factors in an embodiment of the present invention. [Figure 9] 1 is a configuration block diagram of a photovoltaic power generation power prediction device according to an embodiment of the present invention. [Figure 10] FIG. 2 is a hardware configuration diagram of a computer device according to an embodiment of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0017] Example 1 As shown in FIG. 1, the solar power generation power forecasting method includes: The first solar power generation power forecasting model is modified based on the modification coefficient to generate a second solar power generation power forecasting model, which is constructed based on the cell temperature and irradiance of the solar power generation module. It is determined whether the environmental temperature information of the photovoltaic power generation module to be predicted matches the preset requirement information, and if it does not match the preset requirement information, a first relationship is determined with the average battery temperature of the photovoltaic power generation module to be predicted, and the first relationship represents the relationship between the battery temperature of the photovoltaic power generation module to be predicted, the environmental temperature, and the irradiance. The second photovoltaic power generation prediction model is corrected based on the average battery temperature and the first relationship to generate a third photovoltaic power generation prediction model. The third photovoltaic power generation prediction model is used to predict the photovoltaic power generation value of the photovoltaic power generation module awaiting prediction. According to the weather type of the photovoltaic power generation module to be predicted, a corresponding target weather correction coefficient is determined from a plurality of weather correction coefficients.

[0018] Preferably, the solar power generation value is adjusted based on the target weather correction factor to obtain a target solar power generation value.

[0019] Preferably, before determining the corresponding target weather correction factor from the plurality of weather correction factors, the method includes: A plurality of sets of data for different weather types of the solar power generation module to be predicted are obtained in advance, where each set of data includes device capacity, irradiance value, battery temperature value and actual power generation amount for a corresponding weather type. The device capacity, irradiance value, battery temperature value and actual power generation amount for the corresponding weather type are input into a correction coefficient prediction model, and the weather correction coefficient for the corresponding weather type is output.

[0020] Preferably, the correction factors include a dust influence factor, a light attenuation loss factor, and a surface reflection loss factor. The first solar power generation power forecasting model is corrected according to the correction coefficient to generate a second solar power generation power forecasting model, including:

number

[0021] Preferably, the second solar power generation power prediction model is corrected based on the battery temperature average value and the first relationship to generate a third solar power generation power prediction model, which includes: A first correction is made to the second photovoltaic power generation prediction model based on the average battery temperature. A second correction is performed on the first correction result based on the first relationship to generate a third photovoltaic power generation forecasting model.

[0022] Preferably, the battery temperature average value of the photovoltaic power generation module to be predicted is determined, and includes: All battery temperature values ​​within a preset time period of the solar power generation module waiting for prediction are obtained. From all the battery temperature values, battery temperature values ​​having irradiances smaller than a preset threshold value are excluded. The remaining battery temperature values ​​determine the average battery temperature value. A first correction is made to the second solar power generation power prediction model based on the battery temperature average value, and includes:

number

[0023] Preferably, the first relationship is a linear fitting function. The first relationship of the solar photovoltaic module to be predicted is determined, and includes: The prediction-waiting solar power generation module acquires a plurality of battery temperature values ​​within a preset time interval, a plurality of environmental temperature values ​​respectively corresponding to the plurality of battery temperature values, and a plurality of irradiance values ​​respectively corresponding to the plurality of battery temperature values. A linear fitting function is obtained based on the plurality of battery temperature values, the plurality of environmental temperature values, and the plurality of irradiances, which is used to indicate that the difference between the battery temperature values ​​and the environmental temperature values ​​exhibits a linear relationship with the irradiance.

number

number

[0024] Preferably, the solar power generation power prediction device includes a first model correction module, which is used to correct the first solar power generation power prediction model based on the correction coefficient, and generate a second solar power generation power prediction model, which is configured based on the battery temperature and irradiance of the solar power generation module. The information determining module is included, and is used to determine whether the environmental temperature information of the photovoltaic power generation module waiting for prediction matches the preset required information. A data determination module is included, which is used to determine a first relationship with the average battery temperature of the solar power generation module to be predicted if the predetermined required information does not match, and the first relationship represents the relationship between the battery temperature of the solar power generation module to be predicted, the ambient temperature and the irradiance. A second model correction module is included, which is used to correct the second solar power generation power prediction model based on the battery temperature average value and the first relationship, and generate a third solar power generation power prediction model. A photovoltaic power generation value prediction module is included, which is used to predict the photovoltaic power generation value of the photovoltaic power generation module to be predicted using the third photovoltaic power generation power prediction model.

[0025] Preferably, the computer device includes a memory and a processor, the memory and the processor being communicatively connected to each other, computer instructions stored in the memory, and the processor executing the computer instructions to perform any one of the photovoltaic power forecasting methods.

[0026] Preferably, a computer readable storage medium has computer instructions stored thereon, said computer instructions being used to cause a computer to perform any one of said photovoltaic power forecasting methods.

[0027] Example 2 The photovoltaic power generation power forecasting method according to the embodiment includes the following steps. In step S101, the first photovoltaic power generation forecasting model is corrected based on the correction coefficient to generate a second photovoltaic power generation forecasting model. Specifically, the first photovoltaic power generation power prediction model is constructed based on the battery temperature and irradiance of the photovoltaic power generation module, as shown in the following equation:

number

[0028] Specifically, the second solar power generation power prediction model is obtained by correcting the first solar power generation power prediction model using various correction coefficients, such as a dust influence coefficient, an initial light attenuation loss coefficient, and a surface reflection loss coefficient, as shown in the following equations.

number

[0029] The present invention improves the accuracy of predicting solar power by modifying the first solar power prediction model to incorporate the influence of dust, the influence of light attenuation due to the initial state of the solar power module, and the influence of module surface reflection loss into the solar power prediction process.

[0030] Step S102: Determine whether the environmental temperature information of the photovoltaic power generation module waiting for prediction matches the preset requirement information. Specifically, related technologies for PV power prediction do not take into account the fact that climate differences arise due to differences in the geographical location of PV power plants, and that the module temperature of PV power generation modules constantly changes dynamically due to environmental changes. This leads to a relatively low accuracy of the PV power value predicted under these conditions. For example, the lower the ambient temperature, the greater the deviation of the predicted PV power value. To solve this problem, the present invention first needs to determine whether the ambient temperature information of the PV power generation modules to be predicted matches the preset required information. For example, the average ambient temperature of the PV power generation modules to be predicted is lower than a preset temperature threshold, and the preset temperature threshold can be set to a corresponding temperature value depending on the location of the area, but is not specifically limited thereto.

[0031] Step S103: if the predetermined requirement information is not met, determine a first relationship between the average battery temperature of the photovoltaic power generation module to be predicted and the average battery temperature of the photovoltaic power generation module to be predicted; Specifically, when the average environmental temperature of the solar power generation module to be predicted is lower than a preset temperature threshold, it is necessary to determine the average battery temperature value of the solar power generation module to be predicted and a first relationship. The average battery temperature value can be understood as the average battery temperature value within a preset time interval. The first relationship is the relationship between the battery temperature of the solar power generation module to be predicted, the environmental temperature, and the irradiance, and the relationship can be understood as a chart and / or a linear fitting function used to display the linear relationship between the battery temperature, the environmental temperature, and the irradiance.

[0032] Step S104: Modify the second solar power forecasting model according to the battery temperature average value and the first relationship to generate a third solar power forecasting model. In the present invention, the two methods of averaging the measured battery temperature of the solar power generation module and linear fitting the battery temperature and irradiance of the solar power generation module can jointly improve the accuracy of the solar power generation power forecast results of the power plant.

[0033] Step S105: predicting the solar power value of the solar power generation module to be predicted using the third solar power prediction model. Specifically, by inputting the irradiance of the photovoltaic power generation module awaiting prediction into the third photovoltaic power generation prediction model, the photovoltaic power generation value of the photovoltaic power generation module awaiting prediction can be output, where the irradiance is an average irradiance, for example, an average irradiance per minute.

[0034] In this embodiment, a photovoltaic power generation forecasting method is provided, and the process includes the following steps: Step S201: Modify the first solar power generation power forecasting model based on the modification coefficient to generate a second solar power generation power forecasting model. The first solar power generation power forecasting model is constructed based on the battery temperature and irradiance of the solar power generation module. For details, refer to step S101 in the embodiment shown in FIG. 1 and will not be repeated here. In step S202, it is determined whether the environmental temperature information of the photovoltaic power generation module waiting for prediction matches the preset required information. For details, refer to step S102 in the embodiment shown in FIG. 1, and a detailed description thereof will be omitted here. Step S203: if the predetermined requirement information is not met, determine a first relationship between the average battery temperature value of the photovoltaic power generation module being predicted, where the first relationship represents the relationship between the battery temperature of the photovoltaic power generation module being predicted, the ambient temperature, and the irradiance. For details, refer to step S103 in the embodiment shown in Figure 1 and omit the description here. In step S204, the second solar power generation prediction model is modified based on the average battery temperature and the first relationship to generate a third solar power generation prediction model.

[0035] Specifically, step S204 includes steps S2041 to S2042 as shown in FIG. Step S2041: correcting the second photovoltaic power generation prediction model according to the battery temperature average value; Specifically, as shown in FIG. 3, before step S2041, the average battery temperature value of the photovoltaic power generation module to be predicted needs to be determined, which includes the following steps a1-a3. Step a1: obtain all battery temperature values ​​within a preset time period of the photovoltaic power generation module to be predicted, where the preset time period can be set according to actual circumstances and is not specifically limited herein. Step a2: Exclude battery temperature values ​​whose radiation temperature is less than a preset threshold from all the battery temperature values. Specifically, when the solar radiation of the photovoltaic module is 120W / m 2 Considering that under lower conditions, the sunlight received by the module surface is in a diffuse reflection state, the photon movement direction is random, the module cannot perform photovoltaic effect, and there is no power output, so the preset threshold is set to 120W / m 2 It can be installed in. Step a3: determine the average battery temperature value according to the remaining battery temperature values;

[0036] Specifically, in the process of determining the average battery temperature value, multiple identical time intervals can be set first, and then the average battery temperature value within each time interval can be calculated according to the remaining battery temperature values ​​and the number of time intervals. For example, if there are five time intervals, the corresponding battery temperature values ​​are y1, y2, y3, y4, and y5, and the average battery temperature value within each time interval is y1+y2+y3+y4+y5 / 5.

[0037] In the process S2041, the second solar power generation forecasting model can be corrected according to the following formula and the average battery temperature value:

number

[0038] The present invention performs a first correction on the second photovoltaic power forecasting model according to the average cell temperature value, which can fully take into account the climate differences in different regions and the dynamic differences in module temperature and time caused by environmental changes, and the impact on the photovoltaic power generated by the photovoltaic module, thereby improving the photovoltaic power forecast accuracy of the second photovoltaic power forecasting model after the first correction.

[0039] In step S2042, a second correction is performed on the first correction result based on the first relationship to generate a third photovoltaic power generation forecasting model. Specifically, before step S2042, the first relationship of the photovoltaic power generation modules to be predicted needs to be determined, which includes the following steps b1-b2: In process b1, a plurality of battery temperature values ​​within a predetermined time period of the solar power generation module waiting for prediction, a plurality of environmental temperature values ​​corresponding one-to-one to the plurality of battery temperature values, and a plurality of irradiance values ​​corresponding one-to-one to the plurality of battery temperature values ​​are obtained. Specifically, the preset time period can be set according to actual circumstances and is not limited thereto, for example, a quarter or a year. The battery temperature value can be the average battery temperature value per minute, the irradiance can be the average irradiance value per minute, and the ambient temperature value can be the average ambient temperature value per minute, with a one-to-one correspondence between the three, for example: ambient temperature average value 1 - battery temperature average value 1 - irradiance average value 1, ambient temperature average value 2 - battery temperature average value 2 - irradiance average value 2. More specifically, when obtaining a plurality of irradiance values ​​corresponding one-to-one to a plurality of battery temperature values ​​within a predetermined time period of the solar power generation module to be predicted, the irradiance values ​​need to be analyzed and screened, that is, data on non-operating conditions should be removed, such as power outage maintenance, rainy weather, and 120W / m 2 The low irradiance data is lower. During the process of removing non-operating data, the irradiance is processed using a box-and-whisker method, i.e., the power generation amount is divided into sections, and a box-and-whisker plot is created for the irradiance within each section. The low irradiance data corresponding to the outliers is then identified and filtered. Using the irradiance data filtered using the box-and-whisker plot, the Cook's distance for each irradiance data can be calculated. If the Cook's distance is greater than a preset multiple of the average distance, the irradiance data is deemed to be an outlier and needs to be filtered.

[0040] In process b2, a linear fitting function is obtained based on the plurality of battery temperature values, the plurality of environmental temperature values ​​and the plurality of irradiances, which indicates that the difference between the battery temperature values ​​and the environmental temperature values ​​has a linear relationship with the irradiance. Specifically, multiple sets of data are obtained, each set of data includes the battery temperature value, the battery temperature value and the corresponding irradiance, and the battery temperature value and the corresponding ambient temperature value. After filtering and screening, the screened data is used to perform fitting on a large number of test samples to obtain the relationship between the battery temperature value, the ambient temperature value, and the irradiance, that is, the solar power generation battery temperature T cell and the ambient temperature T aThe difference between the two has a linear relationship with the solar irradiance, and the actual operating temperature of the photovoltaic cell can be calculated by the following formula:

number

[0041] In the step S2042, a second correction is performed on the correction result according to the following formula and the first relationship to generate a third photovoltaic power generation forecasting model:

number

[0042] Considering that the environmental temperature, irradiance, and battery temperature are the main factors affecting the solar power generation, the present invention obtains the correlation between the environmental temperature, irradiance, and battery temperature by fitting a large number of test samples, and then performs a second correction on the first correction result according to the correlation to obtain a second solar power generation prediction model, which can further improve the prediction accuracy of the solar power generation.

[0043] In some optional embodiments, the method also includes steps c1-c2, as shown in FIG. In step c1, a corresponding target weather correction coefficient is determined from among a plurality of weather correction coefficients according to the weather type of the photovoltaic power generation module awaiting prediction. Specifically, the weather types are mainly divided into clear skies and cloudy skies, and include, for example, cloudy skies, cloudy skies, clear skies to cloudy skies, and cloudy skies to rain and snow. The weather modification coefficients also include a clear skies modification coefficient, a cloudy skies modification coefficient, and a rain and snow modification coefficient, and the corresponding weather modification coefficient is selected according to the weather type and used to adjust the photovoltaic power generation power value. In step c2, the photovoltaic power generation value is adjusted based on the target weather correction coefficient to obtain a target photovoltaic power generation value. Specifically, by selecting a corresponding target weather correction coefficient depending on the weather type and adjusting the photovoltaic power generation power value, the prediction accuracy of the target photovoltaic power generation power value to be output can be improved.

[0044] In some optional embodiments, as shown in FIG. 5, before determining the corresponding target weather correction factor from the plurality of weather correction factors, the method further includes steps d1 to d2. Step d1: obtain a plurality of sets of data of the solar power generation modules to be predicted in advance under different weather types, each set of data including device capacity, irradiance value, battery temperature value and actual power generation amount under a corresponding weather type. The solar power generation modules to be predicted are solar power generation modules that do not have access to centralized control data or whose centralized control data quality does not meet training requirements. Specifically, the device capacity represents the maximum power that the photovoltaic power module can continuously output. The irradiance value can be a daily average irradiance, and is determined according to a preset time period during which the set of data is acquired. If the preset time period is daily, the irradiance value is a daily average irradiance, and if the preset time period is monthly, the irradiance value is a monthly average irradiance. The battery temperature value can be the maximum battery temperature, the daily average battery temperature, or the monthly average battery temperature. The actual power generation is the actual power generation of the photovoltaic power module. In process d2, the device capacity, irradiance value, battery temperature value and actual power generation amount in the corresponding weather type are input into a correction coefficient prediction model, and the weather correction coefficient of the corresponding weather type is output. Specifically, the correction coefficient forecasting model uses a linear regression model to obtain the corresponding weather correction coefficient through linear regression. The correction coefficient forecasting model is expressed as follows:

number

[0045] The correction coefficient forecasting model outputs weather correction coefficients corresponding to weather types, allowing prediction results to be examined in conjunction with actual production scenarios, further improving the accuracy of the model by taking weather conditions into account. Considering the optimization of the linear model by training for each weather type, the model representation classified by weather type is shown in Figure 7 and Table 1. Weather is classified into three types of correction coefficients: (1) clear, (2) cloudy / cloudy / sunny then cloudy, and (3) rain / snow / cloudy then rain. Table 1 shows that the clear coefficient is high, the cloudy / cloudy coefficient is low, and the rain / snow coefficient is lowest. Figure 7 shows that power generation is highest for clear weather, low for cloudy / cloudy, and lowest for rain / snow. By optimizing the linear model using different weather coefficients, it is possible to link PV power prediction results with actual production operation scenarios, i.e., different weather types, thereby achieving the goal of improving the accuracy of the prediction results. (Model representation classified by weather) [Table 1]

[0046] In some optional embodiments, as shown in FIG. 8 , if the solar power generation module awaiting prediction is a solar power generation module that can use centralized control data but cannot drive data in an inverter, before determining a corresponding target weather correction coefficient from among a plurality of weather correction coefficients, the method further includes: In step e1, a plurality of sets of centralized control data for different weather types of the solar power generation modules to be predicted are obtained in advance, and each set of centralized control data includes output power and irradiance for a corresponding weather type. In process e2, multiple sets of centralized control data are preprocessed, and the average irradiance per minute is converted to the average irradiance per 10 minutes. The processed output power and irradiance, as well as the device capacity and battery temperature, are input into a linear regression model, and the corresponding sunny weather coefficient k1, cloudy weather coefficient k2, and rainy weather coefficient k3 are output. Specifically, when the solar power generation module waiting for prediction is a solar power generation module that can use centralized control data and uses an inverter to operate data, the device capacity data can be replaced with the operating capacity of the inverter data, and the other processes are the same as the above processes e1 and e2.

[0047] A photovoltaic power generation power prediction device is also provided in this embodiment, and is a device for realizing the above-described embodiment and preferred embodiments, so a description thereof will be omitted. As used below, the term "module" can realize a combination of software and / or hardware for a given function. Although the device described below is preferably realized in software, hardware or a combination of software and hardware can also be realized and is envisioned.

[0048] As shown in FIG. 9, this embodiment provides a photovoltaic power generation power prediction device. The system further includes a first model correction module for correcting the first solar power generation power prediction model based on the correction coefficient to generate a second solar power generation power prediction model, the first solar power generation power prediction model being configured based on the battery temperature and irradiance of the solar power generation module. The information determining module is included, and is used to determine whether the environmental temperature information of the photovoltaic power generation module waiting for prediction matches the preset required information. A data determination module is included, which is used to determine a first relationship with the average battery temperature of the solar power generation module to be predicted if the predetermined required information does not match, and the first relationship represents the relationship between the battery temperature of the solar power generation module to be predicted, the ambient temperature and the irradiance. A second model correction module is included, which is used to correct the second solar power generation power prediction model based on the battery temperature average value and the first relationship, and generate a third solar power generation power prediction model. A photovoltaic power generation value prediction module is included, which is used to predict the photovoltaic power generation value of the photovoltaic power generation module to be predicted using the third photovoltaic power generation power prediction model.

[0049] In some alternative embodiments, the device also includes: a target weather correction factor determining module, which is used to determine a corresponding target weather correction factor from among a plurality of weather correction factors according to the weather type of the solar power generation module to be predicted; A solar power generation value adjustment module is used to adjust the solar power generation value based on the target weather correction coefficient to obtain a target solar power generation value.

[0050] In some alternative embodiments, before determining the corresponding target weather correction factor from among the plurality of weather correction factors, the apparatus also includes: A data acquisition module is used to pre-acquire multiple sets of data of the solar power generation module to be predicted under different weather types, each set of data including device capacity, irradiance value, battery temperature value and actual power generation amount under a corresponding weather type. a weather correction coefficient output module, which is used to input the device capacity, irradiance value, battery temperature value and actual power generation amount in the corresponding weather type into a correction coefficient prediction model, and output the weather correction coefficient of the corresponding weather type;

[0051] In some alternative embodiments, the correction factors include a dust influence factor, a light attenuation loss factor, and a surface reflection loss factor. The first solar power generation power forecasting model is corrected according to the correction coefficient to generate a second solar power generation power forecasting model, including:

number

[0052] In some optional embodiments, the second model modification module includes: a primary correction unit, which is used to make a primary correction to the second photovoltaic power generation forecasting model based on the battery temperature average value; A secondary correction unit is used to perform a secondary correction on the primary correction result based on the first relationship to generate a third photovoltaic power generation forecasting model.

[0053] In some optional embodiments, the data determination module includes: A battery temperature value acquisition unit is used to acquire all battery temperature values ​​within a preset time stage of the solar power generation module waiting for prediction. a battery temperature value elimination unit, which is used to eliminate battery temperature values ​​whose irradiance is less than a preset threshold value from all the battery temperature values; a battery temperature average value determining unit, which is used to determine the battery temperature average value according to the remaining battery temperature values;

[0054] The primary correction unit includes:

number

[0055] In some optional embodiments, the first relationship is a linear fitting function. The data determination module includes: A data acquisition unit is used to acquire a plurality of battery temperature values ​​within a predetermined time interval of the solar power generation module to be predicted, a plurality of environmental temperature values ​​respectively corresponding to the plurality of battery temperature values, and a plurality of irradiance values ​​respectively corresponding to the plurality of battery temperature values. a linear fitting function determination unit, which obtains a linear fitting function based on the plurality of battery temperature values, the plurality of environmental temperature values, and the plurality of irradiances, used to indicate that the difference between the battery temperature value and the environmental temperature value has a linear relationship with the irradiance; The linear fitting function includes:

number

[0056] The secondary compensator unit includes:

number

[0057] Further functional descriptions of each of the above-mentioned modules and units are the same as those of the corresponding embodiments described above, and are omitted here.

[0058] There is also provided a computer device having the photovoltaic power generation power prediction device shown in FIG. 9 described above. FIG. 10 is a schematic diagram of a computer system provided by any embodiment of the present invention, including one or more processors 10, memory 20, and interfaces for connecting components, including high-speed and low-speed interfaces. The components are communicatively connected to each other using different buses and may be mounted on a common motherboard or in other manners as needed. The processor may process instructions executed within the computer system, including instructions stored in or on the memory, to display graphical information for the GUI on an external input / output device, such as a display device, coupled to the interface. In some embodiments, multiple processors and / or multiple buses may be used, along with multiple memories, as needed. Similarly, multiple computer systems may be connected, each providing a portion of the necessary operations, such as a server array, a set of blade servers, or a multiprocessor system. FIG. 10 illustrates a single processor 10 as an example.

[0059] The processor 10 may be a central processor, a network processor, or a combination thereof. The processor 10 may further include a hardware chip. The hardware chip may be an application specific integrated circuit, a programmable logic device, or a combination thereof. The programmable logic device may be a composite programmable logic device, a field programmable logic gate array, a general purpose array logic, or any combination thereof.

[0060] The aforementioned memory 20 stores instructions executable by the at least one processor 10 to cause the at least one processor 10 to execute the methods illustrated in the above-described embodiments.

[0061] The memory 20 may include a program storage area and a data storage area. The program storage area may store an operating system and / or application programs required for at least one function. The data storage area may store data generated in response to use of the computer device. The memory 20 may also include high-speed random access memory, or non-transitory memory such as at least one magnetic disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, the memory 20 may optionally include memory configured remotely from the processor 10, and these remote memories may be connected to the computer device via a network. Examples of such networks include, but are not limited to, the Internet, a corporate intranet, a local area network, a mobile communication network, and combinations thereof.

[0062] Memory 20 may include volatile memory, such as random access memory. Memory may also include non-volatile memory, such as flash memory, hard disk drives, solid state drives, etc. Memory 20 may also include a combination of the above types of memory.

[0063] The computing device also includes a communications interface 30 for communicating with other devices or communications networks.

[0064] Embodiments of the present invention may also provide a computer-readable storage medium that can implement the methods according to the above-described embodiments of the present invention as hardware, firmware, or computer code that can be recorded on the storage medium, or stored on a remote storage medium or a non-transitory machine-readable storage medium downloaded via a network, or stored on a local storage medium. Therefore, the methods described herein can be processed by a general-purpose computer, a dedicated processor, or programmable or dedicated hardware using software stored on a storage medium. The storage medium may be a magnetic disk, optical disk, read-only storage memory, random-access storage memory, flash memory, hard disk, solid-state drive, or the like. Furthermore, the storage medium may further include a combination of the above-mentioned types of memory. It is understood that a computer, processor, microprocessor controller, or programmable hardware includes a storage component that can store or receive software or computer code, and the software or computer code, when accessed and executed by the computer, processor, or hardware, implements the methods described in the above embodiments.

[0065] Although the embodiments of the present invention have been described with reference to the drawings, those skilled in the art may make various modifications and variations within the scope defined by the appended claims without departing from the spirit and scope of the present invention.

Claims

1. 1. A method for forecasting solar power generation, comprising: Correcting the first solar power generation power prediction model based on the correction coefficient to generate a second solar power generation power prediction model, wherein the first solar power generation power prediction model is constructed based on the battery temperature and irradiance of the solar power generation module; Determine whether the environmental temperature information of the photovoltaic power generation module to be predicted matches the preset requirement information, and if it does not match the preset requirement information, determine a first relationship with the average battery temperature of the photovoltaic power generation module to be predicted, where the first relationship represents a relationship between the battery temperature of the photovoltaic power generation module to be predicted, the environmental temperature, and the irradiance; correcting the second photovoltaic power generation prediction model based on the average battery temperature and the first relationship to generate a third photovoltaic power generation prediction model; Using the third photovoltaic power generation prediction model, predict the photovoltaic power generation value of the photovoltaic power generation module to be predicted; A photovoltaic power generation power forecasting method, characterized in that a corresponding target weather correction coefficient is determined from a plurality of weather correction coefficients according to the weather type of the photovoltaic power generation module awaiting prediction.

2. The method for predicting photovoltaic power generation according to claim 1 , further comprising adjusting the photovoltaic power generation value based on the target weather correction coefficient to obtain a target photovoltaic power generation value.

3. Before determining the corresponding target weather correction factor from the plurality of weather correction factors, including: A plurality of sets of data of the solar power generation module to be predicted under different weather conditions are obtained in advance, where each set of data includes device capacity, irradiance value, battery temperature value and actual power generation amount under a corresponding weather condition; 3. The photovoltaic power generation power prediction method according to claim 2, wherein the device capacity, irradiance value, battery temperature value and actual power generation amount for the corresponding weather type are input into a correction coefficient prediction model, and a weather correction coefficient for the corresponding weather type is output.

4. The correction factors include a dust influence factor, a light attenuation loss factor, and a surface reflection loss factor; The first solar power generation power forecasting model is adjusted according to the adjustment coefficient to generate a second solar power generation power forecasting model, including: [Equation 1]

5. Correct the second solar power generation power prediction model based on the battery temperature average value and the first relationship to generate a third solar power generation power prediction model, the third solar power generation power prediction model including: performing a first correction on the second photovoltaic power generation prediction model based on the average battery temperature; The photovoltaic power generation power prediction method according to claim 1 , further comprising: performing a second correction on the first correction result based on the first relationship to generate a third photovoltaic power generation power prediction model.

6. Determining the average battery temperature of the solar power generation module to be predicted, including: Acquire all battery temperature values ​​within a predetermined time period of the solar power generation module waiting for prediction; excluding battery temperature values ​​whose irradiance is less than a preset threshold value from all the battery temperature values; determining the average battery temperature value according to the remaining battery temperature values; Perform a first correction on the second solar power generation forecasting model based on the battery temperature average value, including: [Equation 2]

7. The first relationship is a linear fitting function, and the first relationship of the solar photovoltaic module to be predicted is determined, and includes: Obtaining a plurality of battery temperature values ​​within a predetermined time period of the solar power generation module to be predicted, a plurality of environmental temperature values ​​respectively corresponding to the plurality of battery temperature values, and a plurality of irradiance values ​​respectively corresponding to the plurality of battery temperature values; Obtaining a linear fitting function based on the plurality of battery temperature values, the plurality of environmental temperature values, and the plurality of irradiances, which is used to indicate that the difference between the battery temperature values ​​and the environmental temperature values ​​has a linear relationship with the irradiance; [Equation 3] A second correction is performed on the first correction result according to the first relationship to generate a third solar power generation power forecasting model, the third correction includes: [Equation 4]

8. A photovoltaic power generation prediction apparatus, the apparatus including a first model correction module, which is used to correct a first photovoltaic power generation prediction model based on a correction coefficient and generate a second photovoltaic power generation prediction model, the first photovoltaic power generation prediction model being configured based on a battery temperature and irradiance of a photovoltaic power generation module; An information determination module is included, which is used to determine whether the environmental temperature information of the photovoltaic power generation module to be predicted matches the preset required information; a data determination module, which is used to determine a first relationship between the average battery temperature of the solar power generation module to be predicted and the battery temperature of the solar power generation module to be predicted if the average battery temperature does not match the preset required information, and the first relationship represents a relationship between the battery temperature of the solar power generation module to be predicted, the ambient temperature and the irradiance; a second model correction module, which is used to correct the second solar power generation power prediction model based on the average battery temperature and the first relationship to generate a third solar power generation power prediction model; a photovoltaic power generation power prediction module, the photovoltaic power generation power prediction device being used to predict the photovoltaic power generation power value of the photovoltaic power generation module awaiting prediction using the third photovoltaic power generation power prediction model.

9. A computing device comprising a memory and a processor, The memory and the processor are communicatively connected to each other, computer instructions are stored in the memory, and the processor executes the computer instructions to perform the photovoltaic power generation power prediction method according to any one of claims 1 to 7. A computer device.

10. A computer-readable storage medium, comprising: A computer-readable storage medium having computer instructions stored therein, the computer instructions being used to cause a computer to execute the photovoltaic power generation power prediction method according to any one of claims 1 to 7. A computer-readable storage medium.

Images