A distribution network source-load joint profiling method and system based on photovoltaic historical data

Abstract

The present invention discloses a method and system for joint source-load profiling of a distribution network based on photovoltaic historical data, and relates to the technical field of distribution networks. The method comprises: using historical data of photovoltaic power generation groups as fuzzy soft sets, and using the normalized fuzzy soft sets to predict the power generation forecast for a future set time period; determining an initial model that meets the training end condition as a load prediction model; inputting actual operation data into the load prediction model for processing to obtain the predicted load for the future set time period; using the predicted power generation power and the predicted load to establish a sensitivity matrix in the distribution network, and analyzing the local optimal matching combination of power sources and loads in the distribution network based on the sensitivity matrix; establishing the sensitivity matrix in the distribution network based on the predicted power generation power and load is helpful to analyze the local optimal matching combination of power sources and loads under the operation of the active distribution network, realize the optimized distribution of the flow, and improve the stability and economy of the power system.

Description

Technical Field

[0001] The present invention relates to the technical field of distribution network technology, and more specifically, to a method and system for joint source-load profiling of a distribution network based on photovoltaic historical data. Background Art

[0002] Currently, photovoltaic power generation has become the most promising renewable energy generation technology due to its low cost and zero pollution. However, photovoltaic power generation is affected by various factors, especially the volatility of weather factors, which leads to high volatility and intermittent output. As the primary energy source in the future renewable energy sector, accurately predicting photovoltaic power generation is crucial for achieving grid integration and improving the efficiency of photovoltaic power consumption. The distribution and behavior of sources and loads significantly influence the operation of distribution networks. Analysis of source-load distribution and behavior plays a vital role in grid load forecasting, power management, demand response, operational status analysis, and safety warnings. However, existing technologies cannot effectively integrate photovoltaic power generation (source) with load (load), resulting in instability in distribution networks and significant waste of power resources. Summary of the Invention

[0003] The purpose of the present invention is to provide a method and system for joint source-load profiling of a distribution network based on photovoltaic historical data to solve the problems existing in the above-mentioned background technology.

[0004] The above technical objectives of the present invention are achieved through the following technical solutions:

[0005] In a first aspect, the present application provides a method for generating a joint source-load portrait of a distribution network based on photovoltaic historical data, comprising the following specific steps:

[0006] The historical data of photovoltaic power generation groups are used as fuzzy soft sets, the fuzzy soft sets are normalized, and the predicted power generation power in the future set time period is predicted using the normalized fuzzy soft sets.

[0007] The preset initial model is trained based on historical data until the initial model reaches a preset training end condition, and the initial model that reaches the training end condition is determined as the load forecasting model;

[0008] Obtain actual operating data and input it into the load forecasting model for processing to obtain the predicted load for a set time period in the future;

[0009] The sensitivity matrix in the distribution network is established using the predicted power generation and predicted load, and the local optimal matching combination of power sources and loads in the distribution network is analyzed based on the sensitivity matrix.

[0010] On the basis of the above technical solution, the present invention can also be improved as follows.

[0011] Furthermore, the above initial model is a three-layer BP perceptron neural network, including an input layer, an output layer and a hidden layer, specifically:

[0012] ;

[0013] Where, represents the output of the output layer in the initial model, represents the number of nodes in the hidden layer, Represents the conversion function of the output layer in the initial model, which is an equivalence function. K represents the amount of historical data input to the initial model, Represents the historical data input to the initial model k , represents the hidden layer weight, represents the output layer weight, represents the hidden layer offset, Indicates the output layer offset, the superscript h represents the label of the hidden layer, tanh represents the hyperbolic tangent function.

[0014] Furthermore, the above training ends when the error function of the initial model does not exceed the threshold. The error function is specifically:

[0015] ;

[0016] Where, represents the error function value, T represents the sample size of historical data used to train the initial model, Indicates the historical data i training sample response variables Y The value of Indicates that through i Explanatory variables for training samples Obtained The conditional quantile of , is the indicator function.

[0017] Furthermore, the model parameters of the above initial model are expressed as:

[0018] ,in:

[0019] ;

[0020] Where, represents the penalty parameter, represents the number of hidden layer nodes, Krepresents the amount of historical data input to the initial model, is the quadratic norm, represents the hidden layer weight vector, T represents the sample size of historical data used to train the initial model, Represents the error function value of the initial model.

[0021] Furthermore, the optimal values ​​of the penalty parameters and the number of hidden layer nodes are determined by the AIC criterion, specifically:

[0022] ;

[0023] Where, To choose different and hour The minimum error function under quantile, T represents the sample size of historical data used to train the initial model, K represents the amount of historical data input to the initial model, Indicates the number of hidden layer nodes.

[0024] Furthermore, the above-mentioned fuzzy soft set prediction after normalization is used to obtain the predicted power generation power for the future set time period, specifically:

[0025] The parameter values ​​used to predict the power generation prediction are determined using the normalized fuzzy soft sets, and the power generation prediction is calculated based on the parameter values ​​and the surface temperature of the photovoltaic modules.

[0026] Furthermore, the above parameter values ​​include open circuit voltage, output current, maximum power point current and maximum power point voltage, and the parameter values ​​are determined by fuzzy soft sets, specifically:

[0027] ; , , ;

[0028] Where, represents the open circuit voltage, Indicates the output current, represents the maximum power point voltage, represents the maximum power point current, A and k represents the diode characteristic factor and the Boltzmann constant, and represent the surface temperature, series internal resistance and parallel internal resistance of the photovoltaic module respectively. q is the unit charge, represents the diode saturation current, represents the current generated by light radiation, represents the total composite current, represents the total composite voltage, The fill factor or curve factor of the photovoltaic cell is and The area of ​​the rectangle formed by and The ratio of the areas of the curves formed;

[0029] The surface temperature of the photovoltaic module is specifically: , where Indicates the surface temperature of the photovoltaic module, Indicates the ambient temperature, represents the coefficient, , Indicates light intensity.

[0030] In a second aspect, the present application provides a distribution network source-load joint portrait system based on photovoltaic historical data, which is applied to a distribution network source-load joint portrait method based on photovoltaic historical data according to any one of the first aspects, including:

[0031] The first module is used to treat the historical data of the photovoltaic power generation group as a fuzzy soft set, perform normalization processing on the fuzzy soft set, and use the normalized fuzzy soft set to predict the power generation forecast for a set time period in the future;

[0032] The second module is used to train the preset initial model based on historical data until the initial model reaches a preset training end condition, and determine the initial model that reaches the training end condition as the load forecasting model;

[0033] The third module is used to obtain actual operation data and input the actual operation data into the load forecasting model for processing to obtain the predicted load for a set time period in the future;

[0034] The fourth module is used to establish a sensitivity matrix in the distribution network using the predicted power generation and predicted load, and analyze the local optimal matching combination of power sources and loads in the distribution network based on the sensitivity matrix.

[0035] In a third aspect, the present application provides an electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements any one of the methods in the first aspect when executing the computer program.

[0036] In a fourth aspect, the present application provides a non-transitory computer-readable storage medium, which stores computer instructions, and the computer instructions enable a computer to execute any one of the methods in the first aspect.

[0037] Compared with the prior art, the present invention has at least the following beneficial effects:

[0038] In this application, by processing the historical data of photovoltaic power generation, outliers and missing values ​​in the data can be removed, the accuracy and reliability of the data can be improved, and a reliable data basis can be provided for subsequent power generation prediction; among them, by predicting the load of the distribution network through the load forecasting model, an accurate estimation of future load can be achieved, thereby facilitating the distribution network to better grasp the load changes during the scheduling and planning process, and improve the stability and economy of the power system; establishing a sensitivity matrix in the distribution network based on the predicted power generation and load is helpful to analyze the local optimal matching combination of power source and load under the operation of the active distribution network, realize the optimal distribution of the flow, and improve the stability and economy of the power system. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] The drawings described herein are used to provide a further understanding of the embodiments of the present invention, constitute a part of this application, and do not constitute a limitation of the embodiments of the present invention. In the drawings:

[0040] Figure 1 A flowchart of a method for profiling in an embodiment of the present invention;

[0041] Figure 2 This is a connection diagram of the portrait system in an embodiment of the present invention;

[0042] Figure 3 Schematic diagram of the connection of electronic equipment in an embodiment of the present invention. DETAILED DESCRIPTION

[0043] To make the objectives, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions of the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Generally, the components of the embodiments of the present invention described and shown in the drawings herein can be arranged and designed in various different configurations.

[0044] Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the invention as claimed, but rather merely represents selected embodiments of the present invention. All other embodiments derived by persons of ordinary skill in the art based on the embodiments of the present invention without creative effort shall fall within the scope of protection of the present invention.

[0045] It should be noted that similar reference numerals and letters denote similar items in the following drawings, and therefore, once an item is defined in one drawing, it does not require further definition or explanation in subsequent drawings.

[0046] In the description of the embodiments of the present invention, "a plurality of" means at least two.

[0047] Example 1: In order to effectively combine photovoltaic power generation (source) with load (load) to avoid the problems of instability of the distribution network and waste of more power resources during distribution, this embodiment provides a distribution network source-load joint portrait method based on photovoltaic historical data, such as Figure 1 As shown, the following specific steps are included:

[0048] S1, taking the historical data of the photovoltaic power generation group as a fuzzy soft set, normalizing the fuzzy soft set, and using the normalized fuzzy soft set to predict the power generation forecast for the future set time period.

[0049] Among them, a notable feature of photovoltaic power generation is its intermittent and volatile nature. Since photovoltaic resources are affected by natural factors (parameters) such as weather and season, the historical data of photovoltaic power generation has the characteristics of time series, randomness, diversity and large quantity. If these historical data are not processed, it will be difficult to provide certain data support for photovoltaic power generation.

[0050] Specifically, historical data includes solar radiation, ambient temperature measurement range, battery charging current, battery discharging current and other data; these historical data are used as fuzzy soft sets to process the fuzzy soft sets so as to normalize the fuzzy soft sets, and then the photovoltaic power generation power is predicted based on the normalized fuzzy soft sets.

[0051] Specifically, a fuzzy soft set can be viewed as a family of parameterized fuzzy sets for a given domain. It is a fuzzified extension of a soft set and an object expression model based on parameters and fuzzy information. It is no longer limited to the simple description of 0 / 1 object parameters by exact soft sets, but expresses the degree of parameter membership in a more flexible fuzzy form, which can be applied to the uncertain conditions of photovoltaic power generation in power systems. The fuzzy information contained in the fuzzy soft set is in different forms, and the decision models constructed are also independent of each other. It is obvious that the fusion of mixed fuzzy information can support the solution of power generation prediction problems. In order to meet the needs of real-time and efficient decision-making, the following definition of a one-dimensional mixed fuzzy soft set is given in this embodiment:

[0052] For the domain Fuzzy sets on ,in , if there are two elements right A Membership and When expressed in different types of fuzzy numbers, they can be called fuzzy sets. A is a mixed fuzzy set; otherwise, A It is called a non-mixed fuzzy set; such as: ; In the formula, the three elements and The membership degree of is expressed as general fuzzy number, interval value and trapezoidal fuzzy number respectively. U Fuzzy soft sets on If the fuzzy number type describing the object parameter membership is not unique, it can be called a fuzzy soft set. is a mixed fuzzy soft set.

[0053] Further, we can make For a given domain U All non-mixing fuzzy subsets on , E is a parameter set, and For the domain U Mixed fuzzy soft sets on ,like S Approximate function of is a mapping whose values ​​are non-mixed fuzzy sets, then It is called one-dimensional mixed fuzzy soft set. ,but Specifically, one-dimensional mixed fuzzy soft sets are considered a specific type of mixed fuzzy soft sets, characterized by different fuzzy membership representations between different parameters, while maintaining consistency within the same parameter. This characteristic is realistically and practically reflected in power generation forecasting systems, where the representations of attribute indicators vary across different entities while maintaining consistency within the indicators themselves.

[0054] After establishing a one-dimensional mixed fuzzy soft set model, this embodiment further proposes a normalization method based on distance measurement. To maximize the preservation of the degree information of the original fuzzy numbers, the normalization process for the one-dimensional mixed fuzzy soft set requires not only maintaining the magnitude relationship between the original parameter values ​​but also conveying the relative distance relationship between them. Since the essence of the parameter values ​​is the fuzzy membership, to avoid mutual interference, direct comparison or mixed operations are not performed on different parameters, thereby ensuring their independence. Based on these considerations, the basic strategy for the mixed fuzzy soft set normalization adopted in this embodiment is as follows: first, the relative distances between various fuzzy numbers and 0 / 1 are calculated. Then, these distances are used to locate the fuzzy numbers within a range, and finally, they are converted into general fuzzy numbers, as shown below:

[0055] Assumptions is a given domain The one-dimensional mixed fuzzy soft set on yes The expression matrix in tabular form, where is an object Parameters Obviously Is a mixed fuzzy matrix. If Normalized to general fuzzy soft sets The corresponding expression matrix is , then according to the normalization strategy, the matrix and The corresponding relationship between them can be described as: Where, Fuzzy number The distance from 0, Fuzzy number The distance between 1.

[0056] Furthermore, due to the variety of fuzzy numbers and different distance measurement methods, this embodiment uses the interval number (IVFS) as an example to demonstrate and The production process:

[0057] Number of intervals Defined as the set of real numbers Subset of: .therefore, Can also be written as ; Two interval numbers , the distance between them is:

[0058] .

[0059] because , so Let be the interval number. According to the above formula, Distance from 1 , Distance from 0 They are:

[0060] ;

[0061] .

[0062] Optionally, after the above fuzzy number conversion method, the mixed fuzzy matrix The general fuzzy matrix is ​​obtained , we get the tabular form of the general fuzzy soft set, and the normalization of the mixed fuzzy soft set is completed. Subsequently, the photovoltaic power generation power can be predicted based on the normalized fuzzy soft set.

[0063] Furthermore, the principle of photovoltaic power generation is through the photovoltaic effect. Photovoltaic cells have a structure similar to the PN junction of a diode. When light shines on the cell, a voltage is generated across the PN junction. The output power of a single photovoltaic panel is relatively small, so photovoltaic power generation units usually connect a large number of photovoltaic cells in series and parallel to form a photovoltaic array for grid-connected power generation. The modeling method for the output characteristics of photovoltaic modules is usually to convert the photovoltaic cells into an equivalent circuit model; specifically, for given external conditions such as stability and light intensity, when the load changes from 0 to When the open circuit voltage is changes to 0; at the same time, the output current will Under given external conditions, the output voltage and current of the photovoltaic cell are distributed on a curve, and the output power The size of also changes with the voltage value. The optimal working point of the photovoltaic cell, its corresponding voltage and current are the maximum power point voltage Maximum power point current ;Depend on and The rectangular area formed is also called the maximum output power of the photovoltaic cell ,Right now:

[0064] , where is the fill factor or curve factor of the photovoltaic cell, which is and The area of ​​the rectangle and and The area of ​​the curve ratio.

[0065] Among them, since the dynamic characteristics of photovoltaic cells are affected by a variety of external conditions, such as light intensity, ambient temperature, particle radiation, etc., these factors often affect the performance of photovoltaic cells at the same time, such as temperature and light intensity. In the application process, the above-mentioned normalized fuzzy soft set is used to determine In actual application, the parameter value should be considered in terms of ambient temperature. and light intensity The surface temperature of photovoltaic modules The effect of light intensity The parameter values ​​under different light intensities and battery temperatures are obtained by compensating for photovoltaic cell problems.

[0066] Optionally, the above parameter values ​​include open circuit voltage, output current, maximum power point current, and maximum power point voltage, and the parameter values ​​are determined by fuzzy soft sets, specifically:

[0067] ; , , ;

[0068] Where, represents the open circuit voltage, Indicates the output current, represents the maximum power point voltage, represents the maximum power point current, A and k represents the diode characteristic factor and the Boltzmann constant, and represent the surface temperature, series internal resistance and parallel internal resistance of the photovoltaic module respectively. q is the unit charge, represents the diode saturation current, represents the current generated by light radiation, represents the total composite current, represents the total composite voltage, The fill factor or curve factor of the photovoltaic cell is and The area of ​​the rectangle formed by and The ratio of the areas of the curves formed;

[0069] The surface temperature of the photovoltaic module is specifically: , where Indicates the surface temperature of the photovoltaic module, Indicates the ambient temperature, represents the coefficient, , Indicates light intensity.

[0070] Specifically, it is assumed that the attribute standards generated by the performance requirements of photovoltaic cells are mixed fuzzy numbers: ((0.6, 0.7, 0.8), (0.6, 0.8), (0.6, 0.4), (0.5, 0.6, 0.7, 0.8)).

[0071] Assuming that the dynamic characteristic requirement of photovoltaic cells is a comprehensive utility of 0.75, the application normalization strategy is transformed into a general fuzzy soft set as shown in Table 1 and generates their respective comprehensive performances.

[0072] Table 1

[0073] hesitant fuzzy number Number of intervals Trapezoidal fuzzy number Comprehensive performance Open circuit voltage 0.63 0.59 0.45 0.56 Output current 0.77 0.70 0.79 0.75 Maximum power point voltage 0.70 0.85 0.62 0.72 Maximum power point current 0.73 0.74 0.89 0.77

[0074] At this time, the parameters , If the value of is greater than 0.75, the parameter can be used as the parameter for predicting power generation. The other two parameters can be determined in the same way.

[0075] S2: Training a preset initial model based on historical data until the initial model reaches a preset training end condition, and determining the initial model that reaches the training end condition as the load forecasting model.

[0076] The above initial model is a three-layer BP perceptron neural network, including an input layer, an output layer, and a hidden layer. The model is expressed as:

[0077] ;in:

[0078] is a hidden layer node, satisfying: .

[0079] Specifically, the hidden layer transfer function of the load forecasting model constructed in this embodiment is Select the hyperbolic tangent tanh function and the output layer conversion function The equivalence function is selected so that the nonlinear relationship from the input layer to the output layer is as follows:

[0080] ;

[0081] Where, represents the output of the output layer in the initial model, represents the number of nodes in the hidden layer, Represents the conversion function of the output layer in the initial model, which is an equivalence function. K represents the amount of historical data input to the initial model, Represents the historical data input to the initial model k , represents the hidden layer weight, represents the output layer weight, represents the hidden layer offset, Indicates the output layer offset, the superscript h represents the label of the hidden layer, tanh represents the hyperbolic tangent function.

[0082] Specifically, the model parameters in the initial model can be unified using the parameter vector express:

[0083] ;

[0084] Similar to the parameter estimation formula of the quantile regression model, the parameter vector in the load forecasting model is The estimation of , also aims to minimize the error function.

[0085] Optionally, the training termination condition is that the error function of the initial model does not exceed a threshold. Specifically, the error function is:

[0086] ;

[0087] Where, represents the error function value, T represents the sample size of historical data used to train the initial model, Indicates the historical data i training sample response variables Y The value of Indicates that through i Explanatory variables for training samples Obtained The conditional quantile of , is the indicator function.

[0088] Specifically, to prevent the initial model from falling into overfitting, a penalty term is added to the objective function. At this time, the parameter estimation of the initial model is transformed into the following optimization problem:

[0089] The model parameters of the above initial model are expressed as:

[0090] ,in:

[0091] ;

[0092] Where, represents the penalty parameter, represents the number of hidden layer nodes, K represents the amount of historical data input to the initial model, is the quadratic norm, represents the hidden layer weight vector, T represents the sample size of historical data used to train the initial model, Represents the value of the error function of the initial model.

[0093] Furthermore, the optimal values ​​of the penalty parameters and the number of hidden layer nodes are determined by the AIC criterion, specifically:

[0094] ;

[0095] Where, To choose different and hour The minimum error function under quantile, T represents the sample size of historical data used to train the initial model, K represents the amount of historical data input to the initial model, Indicates the number of hidden layer nodes.

[0096] S3, obtaining actual operation data, and inputting the actual operation data into the load forecasting model for processing to obtain the forecast load for the future set time period.

[0097] Specifically, the penalty parameter is selected according to the AIC criterion and the number of hidden layer nodes The optimal value of , calculate the optimal parameter that minimizes the error function ; Due to the indicator function in the error function and activation function Since it is not differentiable at the origin, the traditional nonlinear optimization algorithm based on gradient is not suitable for direct optimization. It is proposed to first use the Huber norm smoothing approximation function which is differentiable everywhere to replace and , and then solve it using nonlinear optimization method.

[0098] S4, using the predicted power generation and predicted load to establish a sensitivity matrix in the distribution network, and based on the sensitivity matrix, analyzing the local optimal matching combination of power source and load in the distribution network.

[0099] Among them, dynamic sensitivity was selected as the operation analysis indicator. Based on the inherent admittance matrix of the power grid, multiple influencing factors were formed and constructed in a regional density manner. Its overall complexity and cost are better than the reconstruction of the distribution network alone; the construction of compactness mainly starts from sensitivity analysis. Sensitivity calculation can describe the sensitivity between grid nodes based on the relationship between various dependent variables and is the most intuitive indicator of compactness. Specifically, sensitivity calculation is generally used to study the differential relationship between various corresponding physical variables under the operation state of the power grid. It is used to obtain the sensitivity between dependent variables and independent variables, thereby analyzing specific variables of power grid operation. It has a wide range of applications in power grids, such as power grid reliability analysis, network loss analysis, transmission capacity, etc., which can improve system safety, increase system stability margin and economy, etc.

[0100] Furthermore, the power flow sensitivity matrix constructed in the embodiment involves three types of variables: independent parameters , including the invariant parameters such as line admittance, Representation; state variable , including load node voltage, phase angle, power generation node phase angle, etc. Represents; control variables , active power and voltage of power generation nodes, voltage and phase angle of balance nodes, etc., are used express.

[0101] The main analysis and calculations currently performed using the sensitivity matrix include: ① busbar sensitivity, which can determine the location of reactive compensation and judge the weak state of the busbar; ② branch sensitivity, which can determine the degree of influence of a branch on the system voltage stability; ③ photovoltaic power generation sensitivity, which determines the photovoltaic power generation that is most important for voltage stability near the critical point. The methods used for flow sensitivity calculation mainly include static sensitivity and trajectory sensitivity, which analyze the system from the perspectives of static section and dynamic response respectively. The general expression of steady-state flow, based on the above parameters, is: .

[0102] Among them, it can be assumed that the system stable point is , if the stable point after the system disturbance is , according to Taylor expansion:

[0103] ; Substituting the above two equations into the solution, we have:

[0104] ;

[0105] The above formula is caused by changes in Variation of the sensitivity matrix.

[0106] Specifically, for example, when voltage is used as the primary reference variable for system safety, the voltage fluctuation sensitivity relationship between the power node and the load node is used as the sensitivity matrix. The weak points in the voltage changes at different nodes in the system can be used as constraints on the safety margin during system operation to guide the system's optimized operation. For the differential relationship matrix between the generator and load nodes with node voltage changes, the reactive power balance equation injected by the node is:

[0107] ;

[0108] Where, is the output of the power grid to the load, is the imaginary part, is the total number of nodes, For nodes j The voltage amplitude at .

[0109] Furthermore, the power generation and load nodes are rewritten separately into a matrix form:

[0110] ;

[0111] Where, 、 、 、 is the invariant parameter corresponding to the load and power generation nodes, is the voltage amplitude at the load, is the voltage amplitude at the power generation point, is the change in load and power generation nodes. Assuming the adjustment back unchanged, that is , which is:

[0112] ,

[0113] ,

[0114] .

[0115] in, This is the sensitivity matrix of the voltage between the power generation and load nodes. Through this calculation, we can analyze the power generation node that has the greatest impact on the load node change. If there is a voltage rise at the power node, it will affect the load voltage qualification rate. Therefore, based on this sensitivity matrix, we can further determine the system safety boundary and the ranking of the impact of reactive power changes.

[0116] Specifically, establishing a sensitivity matrix in the distribution network based on the predicted power generation and load helps to analyze the local optimal matching combination of power source and load under the operation of the active distribution network, realize the optimal distribution of power flow, and improve the stability and economy of the power system.

[0117] Example 2: The present application provides a distribution network source-load joint portrait system based on photovoltaic historical data, which is applied to a distribution network source-load joint portrait method based on photovoltaic historical data in any one of Example 1, such as Figure 2 Shown, including:

[0118] The first module is used to take the historical data of the photovoltaic power generation group as a fuzzy soft set, normalize the fuzzy soft set, and use the normalized fuzzy soft set to predict the power generation forecast for a set time period in the future.

[0119] The second module is used to train the preset initial model based on historical data until the initial model reaches a preset training end condition, and determine the initial model that reaches the training end condition as the load forecasting model.

[0120] The third module is used to obtain actual operation data and input the actual operation data into the load forecasting model for processing to obtain the predicted load for the future set time period.

[0121] The fourth module is used to establish a sensitivity matrix in the distribution network using the predicted power generation and predicted load, and analyze the local optimal matching combination of power sources and loads in the distribution network based on the sensitivity matrix.

[0122] Example 3: This embodiment of the present application provides an electronic device, such as Figure 3 As shown, it includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, any one of the methods in Example 1 is implemented.

[0123] Example 4: An embodiment of the present application provides a non-transitory computer-readable storage medium, which stores computer instructions, and the computer instructions enable a computer to execute any one of the methods in Example 1.

[0124] The above specific implementation methods further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above are only specific implementation methods of the present invention and are not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for joint source-load profiling of distribution network based on photovoltaic historical data, characterized in that: The specific steps include: Taking historical data of the photovoltaic power generation group as a fuzzy soft set, normalizing the fuzzy soft set, and using the normalized fuzzy soft set to predict the power generation forecast for a set time period in the future; Training a preset initial model based on the historical data until the initial model reaches a preset training end condition, and determining the initial model that reaches the training end condition as the load forecasting model; Acquiring actual operating data and inputting the actual operating data into the load forecasting model for processing to obtain a forecast load for a set time period in the future; Establishing a sensitivity matrix in the distribution network using the predicted power generation and the predicted load, and analyzing the local optimal matching combination of power sources and loads in the distribution network based on the sensitivity matrix; The predicted power generation in a future set time period is obtained by using the normalized fuzzy soft set prediction, specifically: The parameter values ​​for predicting the power generation prediction are determined using the normalized fuzzy soft set, and the power generation prediction is calculated based on the parameter values ​​and the surface temperature of the photovoltaic module; the parameter values ​​include open circuit voltage, output current, maximum power point current, and maximum power point voltage. The parameter values ​​are determined using the fuzzy soft set, specifically: in: Where, represents the open circuit voltage, Indicates the output current, represents the maximum power point voltage, represents the maximum power point current, A and k represents the diode characteristic factor and the Boltzmann constant, and represent the surface temperature, series internal resistance and parallel internal resistance of the photovoltaic module respectively. q is the unit charge, represents the diode saturation current, represents the current generated by light radiation, represents the total composite current, represents the total composite voltage, The fill factor or curve factor of the photovoltaic cell is and The area of ​​the rectangle formed by and The ratio of the areas of the curves formed; The surface temperature of the photovoltaic module is specifically: , where Indicates the surface temperature of the photovoltaic module, Indicates the ambient temperature, represents the coefficient, , Indicates light intensity.

2. A method for joint source-load profiling of a distribution network based on photovoltaic historical data according to claim 1, characterized in that: The initial model is a three-layer BP perceptron neural network, including an input layer, an output layer and a hidden layer, specifically: Where, represents the output of the output layer in the initial model, represents the number of nodes in the hidden layer, Represents the conversion function of the output layer in the initial model, which is an equivalence function. K represents the amount of historical data input to the initial model, Represents the historical data input to the initial model k , represents the hidden layer weight, represents the output layer weight, represents the hidden layer offset, Indicates the output layer offset, the superscript h represents the label of the hidden layer, tanh represents the hyperbolic tangent function, is the conditional quantile.

3. A method for joint source-load profiling of a distribution network based on photovoltaic historical data according to claim 2, characterized in that: The training end condition is that the error function of the initial model does not exceed a threshold value, and the error function is specifically: Where, represents the error function value of the initial model, T represents the sample size of historical data used to train the initial model, Indicates the historical data i training sample response variables Y The value of Indicates that through i Explanatory variables for training samples Obtained The conditional quantile of , is the indicator function.

4. A method for joint source-load profiling of a distribution network based on photovoltaic historical data according to claim 3, characterized in that: The model parameters of the initial model are expressed as: in: Where, represents the penalty parameter, represents the number of hidden layer nodes, K represents the amount of historical data input to the initial model, is the quadratic norm, represents the hidden layer weight vector, T represents the sample size of historical data used to train the initial model, Represents the error function value of the initial model.

5. A method for joint source-load profiling of a distribution network based on photovoltaic historical data according to claim 4, characterized in that: The optimal values ​​of the penalty parameter and the number of hidden layer nodes are determined by the AIC criterion, specifically: Where, To choose different and hour The minimum error function under quantile, T represents the sample size of historical data used to train the initial model, K represents the amount of historical data input to the initial model, Indicates the number of hidden layer nodes.

6. A distribution network source-load joint portrait system based on photovoltaic historical data, applied to a distribution network source-load joint portrait method based on photovoltaic historical data according to any one of claims 1 to 5, characterized in that: include: The first module is used to use the historical data of the photovoltaic power generation group as a fuzzy soft set, normalize the fuzzy soft set, and use the normalized fuzzy soft set to predict the power generation forecast for a set time period in the future; A second module is configured to train a preset initial model based on the historical data until the initial model reaches a preset training end condition, and determine the initial model that reaches the training end condition as the load forecasting model; The third module is used to obtain actual operation data and input the actual operation data into the load forecasting model for processing to obtain the predicted load for a set time period in the future; The fourth module is used to establish a sensitivity matrix in the distribution network using the predicted power generation and the predicted load, and analyze the local optimal matching combination of power sources and loads in the distribution network based on the sensitivity matrix.

7. 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 method according to any one of claims 1 to 5 is implemented when the processor executes the computer program.

8. A non-transitory computer-readable storage medium, characterized in that The non-transitory computer-readable storage medium stores computer instructions, which enable a computer to execute the method according to any one of claims 1 to 5.

Images