Photovoltaic generating capacity prediction method and system

Abstract

The invention discloses a photovoltaic power generation capacity prediction method and system, and relates to the field of photovoltaic power generation. The method comprises the steps of obtaining power curve data of a single user, performing data preprocessing on the power curve data, inputting the preprocessed power curve data into a photovoltaic generating capacity prediction model, and predicting the generating capacity of a whole transformer area; the training method of the photovoltaic generating capacity prediction model comprises the steps of collecting a photovoltaic data set of photovoltaic users in the same transformer area, and performing data preprocessing on the data set; the method comprises the following steps: quantifying the influence of weather factors on the photovoltaic power generation efficiency by utilizing photovoltaic power curve data of the morning every day, and constructing various regression models in combination with all-day photovoltaic power generation data; comparing to obtain an optimal model; and replacing the all-day photovoltaic data set of the photovoltaic users in the same district with the all-day photovoltaic data set of the district, retraining the optimal model, and finally obtaining a photovoltaic generating capacity prediction model. According to the method, the photovoltaic generating capacity of the whole transformer area can be predicted through the power curve data of the single user.

Description

Technical Field

[0001] The invention relates to the field of photovoltaic power generation and proposes a photovoltaic power generation prediction method and system. Background Art

[0002] With the widespread application of distributed photovoltaic power generation technology, accurate prediction of photovoltaic power generation has become a key link in power grid operation and dispatch. Traditional photovoltaic power generation prediction methods mostly rely on meteorological data, but these methods often have problems such as poor real-time performance and high data acquisition costs.

[0003] The patent application with Chinese patent publication number CN117713188A discloses a data-physics-driven short-term photovoltaic output prediction method and system. This method constructs a physically driven photovoltaic output prediction model by fitting weather information and its corresponding historical photovoltaic output data, thereby extracting short-term changes between weather factors and photovoltaic output, and then training similar daily photovoltaic output data based on the long-term and short-term time series network optimized by the whale optimization algorithm to obtain a data-driven photovoltaic output prediction model, thereby extracting the long-term trend between weather factors and photovoltaic output, and then generalizing the photovoltaic output prediction results of the two models using the Stacking integrated learning framework, which can ensure good prediction results in both typical weather and transitional weather. However, this method requires a large amount of data and cannot predict the power generation throughout the day through the morning photovoltaic power curve data. Summary of the invention

[0004] In order to solve the problems existing in the prior art, the present invention proposes a photovoltaic power generation prediction method and system, which can predict the photovoltaic power generation of the entire area through the power curve data of a single user.

[0005] In order to solve the above technical problems, the technical solution adopted by the present invention is: In a first aspect, the present invention provides a photovoltaic power generation prediction method, comprising: The power curve data of a single user is obtained, the power curve data is preprocessed, and the preprocessed power curve data is input into a photovoltaic power generation prediction model to predict the power generation of the entire station area; the training method of the photovoltaic power generation prediction model includes: Collect photovoltaic data sets from photovoltaic users in the same area and perform data preprocessing on the data sets; Divide the processed data set into training set and test set; Using the photovoltaic power curve data every morning, we quantified the impact of weather factors on photovoltaic power generation efficiency, and combined it with the photovoltaic power generation data throughout the day to build a variety of regression models; The training set is used to train various regression models respectively, and the root mean square error and mean absolute percentage error are calculated using the test set to evaluate the performance of various regression models, and the optimal model is obtained by comparison; The photovoltaic data set of photovoltaic users in the same substation for a whole day is replaced with the photovoltaic data set of the substation for a whole day, and the optimal model is retrained to finally obtain a photovoltaic power generation prediction model.

[0006] As a further improvement of the present invention, the photovoltaic data set of photovoltaic users in the same substation area is collected and the data set is preprocessed, including: The 96-point power curve data and daily power generation data of photovoltaic users are used as data sets; the data sets are preprocessed, and time adjustment and day length adjustment are performed on the data sets; time adjustment and day length adjustment are performed on the data of all dates; after time adjustment, the corresponding photovoltaic power after time adjustment is obtained every day; after day length adjustment, the adjusted power generation is obtained every day.

[0007] As a further improvement of the present invention, the time adjustment includes: setting the sampling time point T of the photovoltaic power curve data to i , i=1,...,96, according to the sunrise time S of the jth day j Adjustment is made, and the sunrise time S1 of the first day is selected as the benchmark. The sampling time of the i-th sampling point on the j-th day is:

[0008] Adjusted sampling time T j i The photovoltaic power value P j i , according to the nearest sampling time point as T k and T k+1 , T k <T j i ≤T k+1 , the corresponding photovoltaic power value P k and P k+1 , calculated by linear interpolation method, specifically: .

[0009] As a further improvement of the present invention, the day length adjustment includes: the power generation Q on the jth day j According to the length of daylight hours L j , select the daytime duration L1 of the first day as the benchmark, perform standardization processing, and obtain the standardized power generation W j , the power generation Q j Multiply the reference day's day length L1 and the current day's day length L j Ratio of: .

[0010] As a further improvement of the present invention, the construction of multiple regression models includes: The first is a linear regression model: a linear regression model is used to establish a linear equation to describe the relationship between the morning photovoltaic power curve and the photovoltaic power generation throughout the day:

[0011] Among them, y represents the photovoltaic power generation throughout the day, x1, x2, ..., x n represents the photovoltaic power generation at different time points in the morning, β0 is the intercept, β1, β2, ..., β n is the regression coefficient, ϵ is the error term; The second is the ridge regression model: the ridge regression model is introduced, and the L2 regularization term is added to the estimation of the regression coefficient:

[0012] Among them, λ is the regularization parameter; The third is the LASSO regression model: The LASSO regression model adds the L1 regularization term:

[0013] Among them, λ is the regularization parameter.

[0014] As a further improvement of the present invention, the root mean square error is a measure of the standard deviation between the model prediction value and the actual value, specifically:

[0015] In the formula, y i represents the actual photovoltaic power generation on the i-th day, ŷ i represents the predicted photovoltaic power generation on the i-th day, and n represents the total number of days.

[0016] As a further improvement of the present invention, the mean absolute percentage error measures the relative value of the prediction error, specifically:

[0017] In the formula, RMSE is the root mean square error, y i Indicates i Actual photovoltaic power generation per day, ŷ i represents the predicted photovoltaic power generation on the i-th day, n Indicates the total number of forecast days; MAPE is the mean absolute percentage error, In a second aspect, the present invention provides a photovoltaic power generation prediction system, comprising: An acquisition module, used to acquire power curve data of a single user; A preprocessing module, used for preprocessing the power curve data; The prediction module is used to input the pre-processed power curve data into the photovoltaic power generation prediction model to predict the power generation of the entire station area; The training method of the photovoltaic power generation prediction model includes: Collect photovoltaic data sets from photovoltaic users in the same area and perform data preprocessing on the data sets; Divide the processed data set into training set and test set; Using the photovoltaic power curve data every morning, we quantified the impact of weather factors on photovoltaic power generation efficiency, and combined it with the photovoltaic power generation data throughout the day to build a variety of regression models; The training set is used to train various regression models respectively, and the root mean square error and mean absolute percentage error are calculated using the test set to evaluate the performance of various regression models, and the optimal model is obtained by comparison; The photovoltaic data set of photovoltaic users in the same substation for a whole day is replaced with the photovoltaic data set of the substation for a whole day, and the optimal model is retrained to finally obtain a photovoltaic power generation prediction model.

[0018] In a third aspect, the present invention provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the photovoltaic power generation prediction method is implemented when the processor executes the computer program.

[0019] In a fourth aspect, the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the photovoltaic power generation prediction method is implemented.

[0020] In a fifth aspect, the present invention provides a computer program product, wherein the computer program product includes computer instructions, and the computer instructions instruct a computer to execute the photovoltaic power generation prediction method.

[0021] The beneficial effects of the present invention compared with the prior art are as follows: The present invention realizes the effective prediction of photovoltaic power generation throughout the day by establishing a linear regression model, a ridge regression model and a LASSO regression model, and at the same time proposes a method for predicting the power generation of the entire area using the power curve data of a single user, which has important application value in power grid dispatching. By establishing a regression model and applying the morning power data of photovoltaic users, the influence of weather factors on photovoltaic power generation efficiency is quantified, and the photovoltaic power generation throughout the day is effectively predicted. The ridge regression and LASSO regression models are more robust than traditional linear regression models in processing data. Predicting the power generation of the entire area through the power curve data of a single user simplifies the calculation process and improves the efficiency and accuracy of the prediction. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the embodiments of the present invention or the drawings of related technical solutions in the prior art are introduced below. It should be understood that the drawings introduced below are only for the convenience of clearly describing some embodiments of the technical solutions of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative work.

[0023] Figure 1 A flow chart of a photovoltaic power generation prediction method provided by the present invention; Figure 2 A prediction workflow diagram of an embodiment of the present invention; Figure 3 The actual power generation of a photovoltaic user, the model-fitted power generation, and the model-predicted power generation curve; Figure 4 The power curve of a photovoltaic user and the total photovoltaic power curve of the area where the user is located; Figure 5 The actual total photovoltaic power generation in the area, the model-fitted total power generation, and the model-predicted total power generation curve; Figure 6 A photovoltaic power generation prediction device provided by the present invention; Figure 7 A schematic diagram of an electronic device provided by the present invention. DETAILED DESCRIPTION

[0024] The embodiments of the present invention are described in detail below, and examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and are not to be construed as limitations of the present invention. For the step numbers in the following embodiments, they are only provided for the convenience of explanation, and the order between the steps is not limited in any way, and the execution order of each step in the embodiment can be adaptively adjusted according to the understanding of those skilled in the art.

[0025] In the description of the present invention, unless otherwise clearly defined, terms such as setting, installing, connecting, etc. should be understood in a broad sense, and technicians in the relevant technical field can reasonably determine the specific meanings of the above terms in the present invention based on the specific content of the technical solution.

[0026] Explanation of relevant terms A linear regression model is a mathematical model that describes the linear relationship between two variables. It is used to describe the linear relationship between a dependent variable and one or more independent variables.

[0027] Ridge regression (Tikhonov regularization) is a biased estimation regression method specifically used for collinear data analysis. It is essentially an improved least squares estimation method.

[0028] LASSO regression (Least Absolute Shrinkage and Selection Operator) is a regression analysis method that compresses variable coefficients by introducing L1 regularization.

[0029] The present invention proposes a method for predicting the photovoltaic power generation of the whole day based on the morning photovoltaic power curve data, and can predict the photovoltaic power generation of the entire area through the power curve data of a single user. The purpose of the present invention is to solve the problem of inaccurate photovoltaic power generation prediction in the prior art, especially the shortcomings of the traditional method in terms of real-time performance and data acquisition cost.

[0030] The first object of the present invention is to provide a photovoltaic power generation prediction method, comprising: S100, obtaining power curve data of a single user; S200, performing data preprocessing on the power curve data; S300, inputting the preprocessed power curve data into a photovoltaic power generation prediction model to predict the power generation of the entire area; the training method of the photovoltaic power generation prediction model includes: S1, collects photovoltaic data sets of photovoltaic users in the same area and performs data preprocessing on the data sets; S2, divide the processed data set into training set and test set; S3, using the photovoltaic power curve data every morning, quantify the impact of weather factors on photovoltaic power generation efficiency, and build multiple regression models based on the photovoltaic power generation data throughout the day; S4, use the training set to train multiple regression models respectively, use the test set to calculate the root mean square error and mean absolute percentage error to evaluate the performance of multiple regression models, and compare to obtain the optimal model; S5, replacing the photovoltaic data set of photovoltaic users in the same substation with the photovoltaic data set of the substation for the whole day, retraining the optimal model, and finally obtaining a photovoltaic power generation prediction model.

[0031] The principle of the present invention is to use the photovoltaic power curve data every morning to quantify the impact of weather factors on photovoltaic power generation efficiency, and to build and optimize the prediction model in combination with the photovoltaic power generation data throughout the day. Specifically: Power curve data is obtained from a single user's photovoltaic system, which records the power generated by the photovoltaic system at different time points. The data is preprocessed, including cleaning outliers, filling missing values, and smoothing to ensure the accuracy and consistency of the data. The photovoltaic power curve data every morning is used to analyze the impact of weather factors (such as light intensity, temperature, cloud thickness, etc.) on photovoltaic power generation efficiency. This can be achieved by feature extraction and pattern recognition of the power curve.

[0032] Combined with the photovoltaic power generation data throughout the day, a variety of regression models are constructed, such as linear regression, nonlinear regression, machine learning algorithms (such as support vector machines, random forests, etc.). These models are designed to learn the complex relationship between weather factors and photovoltaic power generation. The processed data set is divided into a training set and a test set. The training set is used to train the model, and the test set is used to evaluate the performance of the model.

[0033] The training set was used to train multiple regression models, and the test set was used to calculate indicators such as the root mean square error (RMSE) and mean absolute percentage error (MAPE) to evaluate the prediction accuracy of the model. By comparing the performance of different models, the optimal model was selected as the final photovoltaic power generation prediction model. The photovoltaic data set of photovoltaic users in the same substation for the whole day was replaced with the photovoltaic data set of the substation for the whole day to more comprehensively reflect the photovoltaic power generation situation in the substation. The optimal model was retrained to further optimize its prediction performance, and finally a photovoltaic power generation prediction model suitable for the entire substation was obtained.

[0034] Specifically, the present invention utilizes photovoltaic power curve data every morning to quantify the impact of weather factors on photovoltaic power generation efficiency, and combines it with the photovoltaic power generation data for the whole day to establish an effective prediction model; the power generation of the entire area is predicted through the power curve data of a single user, thereby improving the accuracy and calculation simplicity of the power generation prediction for the entire area.

[0035] Furthermore, the present invention quantifies the impact of weather factors on photovoltaic power generation efficiency through photovoltaic power curve data every morning, which helps to improve the accuracy and robustness of the prediction model. By collecting the power curve data of a single user and using the trained prediction model to predict the power generation of the entire area, the prediction process is simplified and the accuracy of the prediction is improved. By constructing multiple regression models and performing performance evaluation, the present invention can select the optimal prediction model to meet the needs of photovoltaic power generation prediction in different scenarios. The method achieves accurate prediction of photovoltaic power generation through the steps of quantifying the impact of weather factors, predicting the power generation of the entire area using single user data, and constructing and selecting the optimal prediction model.

[0036] The method of the present invention is described in detail below with reference to specific drawings and specific embodiments: like Figure 2 As shown, the present invention provides a method for predicting the photovoltaic power generation throughout the day based on the morning photovoltaic power curve data, and predicting the power generation of the entire area through the power curve data of a single user. The training method of the photovoltaic power generation prediction model is implemented by the following steps: S1. Data preprocessing: Collect PV data sets and perform data preprocessing on the data sets, including time adjustment and day length adjustment.

[0037] Time adjustment: adjust the sampling time point T of the photovoltaic power curve data i (i=1,...,96), according to the sunrise time S of the jth day j Adjustments are made and the sunrise time S1 of the first day is selected as the benchmark to ensure that the solar altitude at the i-th sampling point on the j-th day is relatively consistent with the solar altitude at the i-th sampling point on the first day to eliminate the error caused by the difference in solar altitude.

[0038] Adjusted sampling time T j i The photovoltaic power value P j i , the nearest sampling time point can be T k and T k+1 (T k <T j i ≤T k+1), the corresponding photovoltaic power value P k and P k+1 , calculated by linear interpolation method:

[0039] Day length adjustment: Power generation Q on day j j According to the length of daylight hours L j , select the daytime duration L1 of the first day as the benchmark, perform standardization processing, and obtain the standardized power generation W j , the power generation Q j Multiply the reference day's day length L1 and the current day's day length L j The ratio of 2.5 to 2.5 is used to eliminate the impact of day length differences on power generation, making it easier to accurately train the model and predict results.

[0040]

[0041] S2 Dataset division: The processed data set is divided into a training set and a test set. The training set is used for model training, and the test set is used for model verification and effect evaluation.

[0042] S3. Model establishment. The present invention uses three regression models to predict the photovoltaic power generation of users throughout the day: Linear regression model: The most basic linear regression model is used to establish a linear equation to describe the relationship between the morning photovoltaic power curve and the photovoltaic power generation throughout the day. The model is as follows:

[0043] Among them, y represents the photovoltaic power generation throughout the day, x1, x2, ..., x n represents the photovoltaic power generation at different time points in the morning, β0 is the intercept, β1, β2, ..., β n is the regression coefficient and ϵ is the error term.

[0044] Ridge regression model: To solve the multicollinearity problem that may occur in linear regression, the ridge regression model is introduced. This model adds an L2 regularization term to the estimation of the regression coefficient, making the model more robust. Its mathematical form is:

[0045] Among them, λ is the regularization parameter, which controls the complexity of the model.

[0046] LASSO regression model: Similar to the ridge regression model, the LASSO regression model controls the size of the regression coefficient by adding the L1 regularization term and realizes feature selection. Its mathematical form is:

[0047] Among them, λ is the regularization parameter.

[0048] S4. Model training and testing: The three regression models were trained, and the root mean square error (RMSE) and mean absolute percentage error (MAPE) were calculated using the test set to evaluate the model performance, and the optimal model was obtained by comparison.

[0049] RMSE: measures the standard deviation between the model's predicted value and the actual value, the formula is:

[0050] MAPE: measures the relative value of the prediction error, the formula is:

[0051] In the formula, RMSE is the root mean square error, y i Indicates i Actual photovoltaic power generation per day, ŷ i represents the predicted photovoltaic power generation on the i-th day, n Indicates the total number of forecast days; MAPE is the mean absolute percentage error, S5. Prediction of power generation in the entire area based on a single user Since the lighting environment of photovoltaic users in the same area is similar, the power curves are similar. Based on this, the present invention proposes a method for predicting the power generation of the entire area through the power curve data of a single user. The method repeats S3 and S4, replaces the user's full-day power generation with the total power generation of the area for the whole day, retrains the model, compares the error results, and finally obtains the optimal model. By extending the prediction results of a single user to the entire area, the simplicity and efficiency of the power generation prediction of the entire area are improved.

[0052] Example After preprocessing, the data is divided into a training set and a test set. Linear regression, ridge regression and LASSO regression models are used for training, and the model performance is verified on the test set to select the optimal model. The results show that the method of the present invention has high accuracy in the prediction of power generation throughout the day. The method of predicting the power generation of the entire station area based on the power curve of a single user shows good practicality and accuracy.

[0053] The following is a specific embodiment of the present invention. People familiar with the art can easily understand other advantages and effects of the present invention from the contents disclosed in this specification. A method for predicting power generation consumption based on a regression model ( Figure 2 ), including the following steps: Step 1: Select the 96-point power curve data and daily power generation data of photovoltaic users in a photovoltaic area in Shunping County, Baoding City, Hebei Province from February 15 to March 15, 2024 as the data set.

[0054] Furthermore, the dataset is preprocessed and the time and day length are adjusted.

[0055] Taking February 15, 2024 as the benchmark, its sunrise time is 07:11:12, and the day length is 10 hours, 44 minutes and 46 seconds. Taking February 16, 2024 as an example, its sunrise time is 07:09:59, and the day length is 10 hours, 47 minutes and 07 seconds, which is 1 minute and 13 seconds earlier than the sunrise time on February 15, 2024, and the day length is 2 minutes and 21 seconds longer. The photovoltaic 96-point power is recorded at 15 minutes, so the solar altitude at 8:00 on February 15, 2024 should be close to the solar altitude at 7:58:47 on February 16, 2024. The photovoltaic power at 7:58:47 on February 16, 2024 can be calculated by linear interpolation from the photovoltaic power at 7:45 and 8:00 on the same day.

[0056] The power generation on February 16, 2024 is calculated by multiplying the day length adjustment by the adjustment ratio, which is the ratio of the day length on February 15, 2024, which is used as the base day, to the day length on February 16, 2024. Similarly, the data for all dates are adjusted for time and day length.

[0057] After time adjustment, five time points are selected: 8:00, 8:15, 8:30, 8:45, and 9:00 on February 15, 2024, to obtain the corresponding photovoltaic power after time adjustment every day. After day length adjustment, the adjusted power generation every day is obtained.

[0058] Step 2: After preprocessing, the data from February 15 to March 6, 2024 is used as the training set, and the data from March 7 to March 15 is used as the test set.

[0059] Step 3: Based on the data set in step 2, the morning user photovoltaic power curve data is used as the independent variable of the regression model, and the user's full-day power generation is used as the dependent variable. Through different independent variable selection combinations, the data is input into the constructed linear regression, ridge regression and LASSO regression models respectively to obtain the prediction of power generation in the test set time period.

[0060] Step 4: Compare the prediction errors of the three regression models and output the results of the optimal model.

[0061] Step 5: According to Figure 3 In the , the correlation between the power generation of photovoltaic users and the power generation of the substation can replace the user's full-day power generation forecast with the total power generation of the substation for the whole day. Figure 4In the above, the correlation between the power generation of photovoltaic users and the power generation of photovoltaic power in the substation is used to predict the power generation of photovoltaic users. Repeat steps 3 and 4, replace the dependent variable of the user's full-day power generation with the total power generation of the substation throughout the day, predict the power generation of the substation, and finally output the results of the optimal model, see Figure 5 .

[0062] The method of preprocessing the photovoltaic power curve and photovoltaic power generation in this embodiment so that the data can be used for model training is to adjust the time by preprocessing the photovoltaic power curve so that different dates can be at similar solar altitudes, eliminating the influence of sunrise early and late, and reflect the influence of weather factors through photovoltaic power. The photovoltaic power generation is preprocessed to adjust the day length so that different dates can avoid the influence of different day lengths on power generation, thereby affecting model training and prediction.

[0063] Then, a photovoltaic power generation prediction method based on regression model is provided, and linear regression, ridge regression and LASSO regression models are established. The morning photovoltaic power curve data is used to improve the accuracy of photovoltaic power generation prediction throughout the day. Based on the power curve data of a single user, the correlation of photovoltaic power generation of users in the same area is used to predict the power generation of the entire area.

[0064] In summary, the present invention provides a method for predicting photovoltaic power generation throughout the day based on morning photovoltaic power curve data and regression model, which has high prediction accuracy and can predict the power generation of the entire area through single user data. It can be seen that the present invention avoids the lack of real-time weather forecasting and improves the accuracy of photovoltaic power generation prediction by establishing a regression model and taking morning photovoltaic power as an independent variable to predict photovoltaic power generation throughout the day.

[0065] like Figure 6 As shown, the third object of the present invention is to provide a photovoltaic power generation prediction system, comprising: An acquisition module 100 is used to acquire power curve data of a single user; A preprocessing module 200 is used to perform data preprocessing on the power curve data; The prediction module 300 is used to input the pre-processed power curve data into the photovoltaic power generation prediction model to predict the power generation of the entire area; The training method of the photovoltaic power generation prediction model includes: Collect photovoltaic data sets from photovoltaic users in the same area and perform data preprocessing on the data sets; Divide the processed data set into training set and test set; Using the photovoltaic power curve data every morning, we quantified the impact of weather factors on photovoltaic power generation efficiency, and combined it with the photovoltaic power generation data throughout the day to build a variety of regression models; The training set is used to train various regression models respectively, and the root mean square error and mean absolute percentage error are calculated using the test set to evaluate the performance of various regression models, and the optimal model is obtained by comparison; The photovoltaic data set of photovoltaic users in the same substation for a whole day is replaced with the photovoltaic data set of the substation for a whole day, and the optimal model is retrained to finally obtain a photovoltaic power generation prediction model.

[0066] The photovoltaic power generation prediction system of the present invention is based on the photovoltaic power generation prediction method.

[0067] like Figure 7 As shown, the third object of the embodiment of the present invention is to provide an electronic device, including a memory 701, a processor 702, and a computer program stored in the memory 701 and executable on the processor, wherein the processor implements the photovoltaic power generation prediction method when executing the computer program. It also includes a communication interface 703 and a bus 704.

[0068] A fourth objective of an embodiment of the present invention is to provide a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the above-mentioned photovoltaic power generation prediction method is implemented.

[0069] A fifth objective of an embodiment of the present invention is to provide a computer program product, wherein the computer program product comprises computer instructions, and the computer instructions instruct a computer to execute the above-mentioned photovoltaic power generation prediction method.

[0070] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0071] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0072] The present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, readable storage media, optical storage, etc.) containing computer-usable program codes.

[0073] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0074] Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0075] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the relevant field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A photovoltaic power generation prediction method, characterized in that: include: Obtain the power curve data of a single user, perform data preprocessing on the power curve data, input the preprocessed power curve data into the photovoltaic power generation prediction model, and predict the power generation of the entire station area; The training method of the photovoltaic power generation prediction model includes: Collect photovoltaic data sets from photovoltaic users in the same area and perform data preprocessing on the data sets; Divide the processed data set into training set and test set; Using the photovoltaic power curve data every morning, we quantified the impact of weather factors on photovoltaic power generation efficiency, and combined it with the photovoltaic power generation data throughout the day to build a variety of regression models; The training set is used to train various regression models respectively, and the root mean square error and mean absolute percentage error are calculated using the test set to evaluate the performance of various regression models, and the optimal model is obtained by comparison; The photovoltaic data set of photovoltaic users in the same substation for a whole day is replaced with the photovoltaic data set of the substation for a whole day, and the optimal model is retrained to finally obtain a photovoltaic power generation prediction model.

2. The photovoltaic power generation prediction method according to claim 1, characterized in that: The collecting of photovoltaic data sets of photovoltaic users in the same photovoltaic area and preprocessing of the data sets include: The 96-point power curve data and daily power generation data of photovoltaic users are used as data sets; the data sets are preprocessed, and time adjustment and day length adjustment are performed on the data sets; time adjustment and day length adjustment are performed on the data of all dates; after time adjustment, the corresponding photovoltaic power after time adjustment is obtained every day; after day length adjustment, the adjusted power generation is obtained every day.

3. The photovoltaic power generation prediction method according to claim 2, characterized in that: The time adjustment includes: adjusting the sampling time point T of the photovoltaic power curve data i , i=1,...,96, according to the sunrise time S of the jth day j Adjustment is made, and the sunrise time S1 of the first day is selected as the benchmark. The sampling time of the i-th sampling point on the j-th day is: Adjusted sampling time T j i The photovoltaic power value P j i , according to the nearest sampling time point as T k and T k+1 , T k <T j i ≤T k+1 , the corresponding photovoltaic power value P k and P k+1 , calculated by linear interpolation method, specifically: 。 4. The photovoltaic power generation prediction method according to claim 2, characterized in that: The day length adjustment includes: the power generation Q on the jth day j According to the length of daylight hours L j , select the daytime duration L1 of the first day as the benchmark, perform standardization processing, and obtain the standardized power generation W j , the power generation Q j Multiply the reference day's day length L1 and the day's day length L j Ratio of: 。 5. The photovoltaic power generation prediction method according to claim 1, characterized in that: The multiple regression models are constructed, including: The first is a linear regression model: a linear regression model is used to establish a linear equation to describe the relationship between the morning photovoltaic power curve and the photovoltaic power generation throughout the day: Among them, y represents the photovoltaic power generation throughout the day, x1, x2, ..., x n represents the photovoltaic power generation at different time points in the morning, β0 is the intercept, β1, β2, ..., β n is the regression coefficient, ϵ is the error term; The second is the ridge regression model: the ridge regression model is introduced, and the L2 regularization term is added to the estimation of the regression coefficient: Among them, λ is the regularization parameter; The third is the LASSO regression model: The LASSO regression model adds the L1 regularization term: Among them, λ is the regularization parameter.

6. The photovoltaic power generation prediction method according to claim 1, characterized in that: The root mean square error is a measure of the standard deviation between the model prediction value and the actual value, specifically: In the formula, RMSE is the root mean square error, y i Indicates i Actual photovoltaic power generation per day, ŷ i represents the predicted photovoltaic power generation on the i-th day, n Indicates the total number of forecast days; The mean absolute percentage error measures the relative value of the prediction error, specifically: In the formula, MAPE is the mean absolute percentage error.

7. A photovoltaic power generation prediction system, characterized in that: include: An acquisition module, used to acquire power curve data of a single user; A preprocessing module, used for preprocessing the power curve data; The prediction module is used to input the pre-processed power curve data into the photovoltaic power generation prediction model to predict the power generation of the entire station area; The training method of the photovoltaic power generation prediction model includes: Collect photovoltaic data sets from photovoltaic users in the same area and perform data preprocessing on the data sets; Divide the processed data set into training set and test set; Using the photovoltaic power curve data every morning, we quantified the impact of weather factors on photovoltaic power generation efficiency, and combined it with the photovoltaic power generation data throughout the day to build a variety of regression models; The training set is used to train various regression models respectively, and the root mean square error and mean absolute percentage error are calculated using the test set to evaluate the performance of various regression models, and the optimal model is obtained by comparison; The photovoltaic data set of photovoltaic users in the same substation for a whole day is replaced with the photovoltaic data set of the substation for a whole day, and the optimal model is retrained to finally obtain a photovoltaic power generation prediction model.

8. An electronic device, characterized in that: The method comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor implements the photovoltaic power generation prediction method according to any one of claims 1 to 7 when executing the computer program.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the photovoltaic power generation prediction method according to any one of claims 1 to 7 is implemented.

10. A computer program product, comprising computer instructions, characterized in that: The computer instructions instruct the computer to execute the photovoltaic power generation prediction method according to any one of claims 1 to 7.

Images