Long-sequence load prediction method and device of power system, electronic equipment and medium

Abstract

The invention relates to a long-sequence load prediction method and device for a power system, electronic equipment and a medium, and the method comprises the steps: carrying out the multi-stage operation of an input matrix through a Kronecker product layer, so as to construct a high-dimensional matrix; arranging the high-dimensional matrix by using the rotation matrix mask, and deleting high-order duplicates by using the high-dimensional matrix meeting a preset condition; applying a preset kernel function to the input matrixes of different orders; processing the output matrix according to a preset multi-layer perceptron; evaluating a preset power system long sequence load prediction model by using the mean square error and the mean absolute error; and analyzing the relationship and influence between the internal variables of the preset power system long-sequence load prediction model, and determining an optimization scheme of the preset power system long-sequence load prediction model. According to the embodiment of the invention, accurate long-sequence load prediction can be realized by combining the large model with the Kronecker product, the rotation matrix mask and the like, an effective system optimization strategy is provided, and the feature extraction capability, the model robustness, the model interpretability and the long-sequence prediction are remarkably improved.

Description

Technical Field

[0001] The present application relates to the technical field of time series prediction, and in particular to a method, device, electronic equipment and medium for long-sequence load prediction of an electric power system. Background Art

[0002] Long sequence prediction is a new field extended from short sequence prediction. Earlier, recurrent neural networks (RNNs) and their variant networks LSTM and GRU were used for long sequence prediction. Researchers hope to use RNN-approximated networks to extract long-term load features and generate predictions. Recently, the Transformer framework based on the self-attention mechanism has shown strong capabilities in sequence data. The self-attention mechanism has the advantage of encoding long-term time series at one time, which can avoid the problem of losing long-term information. At the same time, the encoder in the Transformer can encode features in long time series and decode them in the decoder for prediction.

[0003] Current research on load forecasting focuses on point forecasting. However, with the increasing number of distributed renewable energy generators and new responsive loads in smart grids, accurate load forecasting becomes increasingly challenging. The variability and uncertainty of power demand have become a challenge for the utility industry. The nonlinear characteristics of existing large models are mainly expressed through activation functions. Although these activation functions can capture the complex nonlinear relationships in the data, they also make the internal mechanism of the model black-boxed, making it difficult to intuitively interpret and understand. Users often cannot clearly know how each input variable affects the prediction results, resulting in a lack of transparency and credibility in practical applications. The nonlinear characteristics of existing large models are mainly expressed through activation functions. Although these activation functions can capture the complex nonlinear relationships in the data, they also make the internal mechanism of the model black-boxed, making it difficult to intuitively interpret and understand. Users often cannot clearly know how each input variable affects the prediction results, resulting in a lack of transparency and credibility in practical applications.

[0004] In addition, existing large models find it difficult to fully utilize and describe relationships across variables. Although they can handle multi-variable inputs, it is difficult to intuitively display the mutual influence and relationship between variables. The parameters of the model are usually distributed in a high-dimensional space and cannot clearly reflect the interaction between variables. This is particularly insufficient for application scenarios that require understanding variable relationships and performing system optimization. Summary of the invention

[0005] The present application provides a long-sequence load forecasting method, device, electronic device and medium for an electric power system to solve the problems in the related art that the existing large models have huge amounts of data in time series forecasting, the nonlinear characteristics are expressed through complex activation functions, resulting in black boxing, and it is difficult to intuitively express cross-variable relationships, which limits the effectiveness and interpretability of the model in practical applications and affects the user's understanding and trust in the model prediction results.

[0006] The first aspect of the present application provides a method for long-sequence load forecasting of an electric power system, comprising the following steps: performing multi-level operations on an input matrix using a Kronecker product layer to construct a high-dimensional matrix; organizing the high-dimensional matrix using a rotation matrix mask to generate a high-dimensional matrix that meets preset conditions, and using the high-dimensional matrix that meets the preset conditions to delete high-order duplicates to generate deleted data; based on the deleted data, applying a preset kernel function to input matrices of different orders to convert the input matrices into output matrices of corresponding orders; processing the output matrix according to a preset multi-layer perceptron to generate a final prediction result; based on the final prediction result, evaluating a preset electric power system long-sequence load forecasting model using mean square error and mean absolute error to generate an evaluation result; based on the evaluation result, analyzing the relationship and influence between internal variables of the preset electric power system long-sequence load forecasting model to generate an analysis result, and determining an optimization scheme for the preset electric power system long-sequence load forecasting model according to the analysis result.

[0007] Optionally, in one embodiment of the present application, before using the Kronecker product layer to perform multi-level operations on the input matrix to construct a high-dimensional matrix, it also includes: performing data cleaning, missing value filling and outlier processing on the load sequence data set of the power system to remove erroneous information in the load sequence data set to obtain the removed data; based on the removed data, dividing the load sequence data set into a training set and a test set, and normalizing the training set and the test set to obtain the processed data.

[0008] Optionally, in one embodiment of the present application, the calculation formula of the high-dimensional matrix is:

[0009] X N (i,j,…,n)=X e (i)×X e (j)×…×X e (n)'

[0010] Among them, X e (n) is the input matrix X after dimension transformation, X N The matrix is ​​composed of N X e The result obtained by the matrix Kronecker product.

[0011] Optionally, in one embodiment of the present application, the calculation formula of the elements in the rotation matrix mask is:

[0012]

[0013] Among them, i, j, n are the position codes of the corresponding dimensions of the mask matrix, and m represents the mask matrix.

[0014] Optionally, in one embodiment of the present application, the calculation formula of the mean square error is:

[0015]

[0016] Among them, y i For the prediction target, is the predicted value;

[0017] The calculation formula of the mean absolute error is:

[0018]

[0019] Among them, y i For the prediction target, is the predicted value.

[0020] The second aspect of the present application provides a long-sequence load forecasting device for an electric power system, comprising: a construction module, which is used to perform multi-level operations on an input matrix using a Kronecker product layer to construct a high-dimensional matrix; a sorting module, which is used to sort the high-dimensional matrix using a rotation matrix mask to generate a high-dimensional matrix that meets preset conditions, and use the high-dimensional matrix that meets the preset conditions to delete high-order duplicates to generate deleted data; an input module, which is used to apply a preset kernel function to input matrices of different orders based on the deleted data to convert the input matrix into an output matrix of a corresponding order; a processing module, which is used to process the output matrix according to a preset multi-layer perceptron to generate a final prediction result; an evaluation module, which is used to evaluate a preset electric power system long-sequence load forecasting model based on the final prediction result using a mean square error and a mean absolute error to generate an evaluation result; a prediction module, which is used to analyze the relationship and influence between internal variables of the preset electric power system long-sequence load forecasting model based on the evaluation result to generate an analysis result, and determine an optimization scheme for the preset electric power system long-sequence load forecasting model based on the analysis result.

[0021] Optionally, in one embodiment of the present application, it also includes: a removal module, which is used to perform data cleaning, missing value filling and outlier processing on the load sequence data set of the power system before using the Kronecker product layer to perform multi-level operations on the input matrix to construct a high-dimensional matrix, so as to remove erroneous information in the load sequence data set to obtain the removed data; a generation module, which is used to divide the load sequence data set into a training set and a test set based on the removed data, and normalize the training set and the test set to obtain the processed data.

[0022] Optionally, in one embodiment of the present application, the calculation formula of the high-dimensional matrix is:

[0023] X N (i,j,…,n)=X e (i)×X e (j)×…×X e (n)'

[0024] Among them, X e (n) is the input matrix X after dimension transformation, X N The matrix is ​​composed of N X e The result obtained by the matrix Kronecker product.

[0025] Optionally, in one embodiment of the present application, the calculation formula of the elements in the rotation matrix mask is:

[0026]

[0027] Among them, i, j, n are the position codes of the corresponding dimensions of the mask matrix, and m represents the mask matrix.

[0028] Optionally, in one embodiment of the present application, the calculation formula of the mean square error is:

[0029]

[0030] Among them, y i For the prediction target, is the predicted value;

[0031] The calculation formula of the mean absolute error is:

[0032]

[0033] Among them, y i For the prediction target, is the predicted value.

[0034] The third aspect of 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 executes the program to implement the long sequence load forecasting method for the power system as described in the above embodiment.

[0035] A fourth aspect of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, which, when executed by a processor, implements the above-mentioned method for long-sequence load forecasting of an electric power system.

[0036] The fifth aspect of the present application provides a computer program product, which stores a computer program that, when executed by a processor, implements the above-mentioned method for long-sequence load forecasting of an electric power system.

[0037] The embodiments of the present application can achieve accurate long-sequence load forecasting, intuitively express variable relationships, and provide effective system optimization strategies by combining large models with Kronecker products, rotation matrix masks and other technologies, and significantly improve feature extraction capabilities, model robustness, model interpretability and long-sequence prediction, providing a more effective and reliable method for power system load forecasting. Thus, the problems in the related technology that the existing large models in time series prediction are huge, the nonlinear characteristics are expressed through complex activation functions, resulting in black boxing, and the difficulty in intuitively expressing cross-variable relationships are solved, which limits the effectiveness and interpretability of the model in practical applications and affects the user's understanding and trust in the model prediction results.

[0038] Additional aspects and advantages of the present application will be given in part in the description below, and in part will become apparent from the description below, or will be learned through the practice of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] The above and / or additional aspects and advantages of the present application will become apparent and easily understood from the following description of the embodiments in conjunction with the accompanying drawings, in which:

[0040] Figure 1 A flowchart of a long sequence load forecasting method for a power system provided according to an embodiment of the present application;

[0041] Figure 2 A schematic diagram of a prediction model structure of a long sequence load prediction method for a power system according to an embodiment of the present application;

[0042] Figure 3 A schematic diagram of an overall algorithm of a prediction model of a method for long-sequence load prediction of a power system according to an embodiment of the present application;

[0043] Figure 4It is a schematic diagram of the overall process of a method for long-sequence load forecasting of a power system according to an embodiment of the present application;

[0044] Figure 5 A schematic diagram of the structure of a long sequence load forecasting device for a power system provided according to an embodiment of the present application;

[0045] Figure 6 It is a schematic diagram of the structure of an electronic device provided according to an embodiment of the present application. DETAILED DESCRIPTION

[0046] The embodiments of the present application 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 intended to be used to explain the present application, and should not be construed as limiting the present application.

[0047] The following describes the long-sequence load forecasting method, device, electronic device and medium of the power system of the embodiment of the present application with reference to the accompanying drawings. In view of the related technologies mentioned in the above background technology, the existing large models have a large amount of black box caused by the expression of nonlinear characteristics through complex activation functions in time series prediction, and it is difficult to intuitively express the relationship between variables, which limits the effectiveness and interpretability of the model in practical applications, and affects the user's understanding and trust of the model prediction results. The present application provides a long-sequence load forecasting method for a power system, in which the large model can be combined with Kronecker product, rotation matrix mask and other technologies to achieve accurate long-sequence load forecasting, intuitively express variable relationships, perform detailed variable impact analysis, and provide effective system optimization strategies, thereby improving the overall performance of the power system and ensuring the safety and efficient operation of the power system. As a result, the problem of the existing large models in the related technology that there is a large amount of black box caused by the expression of nonlinear characteristics through complex activation functions in time series prediction and it is difficult to intuitively express the relationship between variables is solved, which limits the effectiveness and interpretability of the model in practical applications, and affects the user's understanding and trust of the model prediction results.

[0048] Before introducing the long sequence load forecasting method for the power system proposed in the embodiment of the present application, the background technology involved in the method is briefly introduced.

[0049] When making deterministic point predictions for power system loads, the core process is to allow the model to learn the linear and nonlinear characteristics in the load sequence and fit the future conditions that need to be predicted. However, due to the limited load sequence data, the prediction model needs to be able to extract truly effective linear and nonlinear features from the limited energy data set, which places high demands on the performance of the prediction model. The existing methods are still lacking in the ability to extract load sequence features and are in urgent need of improvement.

[0050] In the research on short-term load sequence forecasting, most of the studies are short-sequence forecasting, that is, only predicting one or a few steps into the future, and fewer are predicting more than a hundred steps into the future, that is, long-sequence forecasting. However, long-sequence forecasting plays an important role in the medium-term planning and safe operation of loads in power systems. Therefore, a long-sequence load forecasting model with excellent performance is urgently needed.

[0051] Specifically, Figure 1 A flow chart of a method for long-sequence load forecasting of a power system provided in an embodiment of the present application.

[0052] like Figure 1 As shown, the long sequence load forecasting method of the power system includes the following steps:

[0053] In step S101, a Kronecker product layer is used to perform multi-level operations on the input matrix to construct a high-dimensional matrix.

[0054] It can be understood that the Kronecker product layer in the embodiment of the present application can perform multi-order Kronecker product operations on the input matrix to form an N-order input matrix, thereby better capturing and expressing the nonlinear relationship of the system.

[0055] In the actual execution process, the input data in the embodiment of the present application is converted into an input matrix through the input layer and then passed through the Kronecker product layer. The Kronecker product layer can represent the high-order nonlinear relationship of the system. By passing the input matrix through the N-order Kronecker product, an N-order input matrix can be generated. For any input matrix X, the mapping relationship between the numbers before and after expand is as follows: X(i, j) = X e (m), where: m = (i-1) * N + j. X after the N-order Kronecker product is denoted as X in this application. N , X N The shape of is (length,…,length) and the dimension is N.

[0056] The high-dimensional representation in the embodiments of the present application can effectively capture the complex nonlinear relationships within the system and intuitively display the interactions between variables, which helps users to have a deeper understanding of the prediction mechanism of the model and thus enhance their trust in the prediction results.

[0057] In one embodiment of the present application, the calculation formula of the high-dimensional matrix is:

[0058] X N (i,j,…,n)=X e (i)×X e (j)×…×X e (n)'

[0059] Among them, X e (n) is the input matrix X after dimension transformation, X N The matrix is ​​composed of N X e The result is obtained by the matrix Kronecker product, so the result has a large number of repeated high-order terms.

[0060] Here X N The matrix is ​​composed of N X e The result is obtained by matrix Kronecker product, so its result has a large number of repeated high-order terms.

[0061] Optionally, in one embodiment of the present application, before using the Kronecker product layer to perform multi-level operations on the input matrix to construct a high-dimensional matrix, it also includes: performing data cleaning, missing value filling and outlier processing on the load sequence data set of the power system to remove erroneous information in the load sequence data set to obtain removed data; based on the removed data, dividing the load sequence data set into a training set and a test set, and normalizing the training set and the test set to obtain processed data.

[0062] It can be understood that the embodiments of the present application can use linear interpolation methods to fill missing values ​​in the collected load sequence data set; use global threshold detection, box plot and other methods to process outliers in the collected load sequence data set; and use Min-Max standardization to normalize the collected load sequence data set.

[0063] In the actual implementation process, first of all, in terms of data preprocessing, the embodiment of the present application can clean the data, process outliers and missing values ​​to remove erroneous information in the load sequence data set, ensure the consistency, accuracy and data quality of the data, and divide the data set according to the following method: select the first eighty percent of the data set as the training set, and the last twenty percent as the test set, and normalize the training set and the test set to generate processed data, so that the input data has a more stable gradient and reduces the magnitude difference between features, thereby improving the accuracy of the model, ensuring that the dynamic characteristics of the power system load are fully captured and the prediction performance of the evaluation model on unseen data sets is evaluated, providing a reliable foundation for model training and verification.

[0064] In step S102, the high-dimensional matrix is ​​reorganized using the rotation matrix mask to generate a high-dimensional matrix that meets preset conditions, and the high-order duplicate items are deleted using the high-dimensional matrix that meets the preset conditions to generate deleted data.

[0065] It is understandable that the high-dimensional matrix that meets the preset conditions in the embodiment of the present application can be an N-order high-dimensional matrix. The embodiment of the present application can solve the problem of repeated elements in the high-order case of the model by using a rotation matrix mask, and create an N-order rotation matrix mask, which is formally similar to the N-order input matrix X N Have exactly the same form.

[0066] Specifically, the rotation matrix mask in the embodiment of the present application can be used to transform X after the N-order Kronecker product. N Arrange it to generate an N-order high-dimensional matrix so that it ignores high-order duplicates to generate deleted data.

[0067] The embodiment of the present application introduces a rotation matrix mask technology, which effectively eliminates the redundancy caused by the Kronecker product, reduces the complexity of the model, and improves the efficiency of the model.

[0068] It should be noted that the preset conditions can be set by those skilled in the art according to actual conditions and are not specifically limited here.

[0069] In one embodiment of the present application, the calculation formula of the elements in the rotation matrix mask is:

[0070]

[0071] Among them, i, j, n are the position codes of the corresponding dimensions of the mask matrix, m represents the mask matrix, and for an element at a certain position, the value of the element at this position is 1 if and only if i≤j≤…≤n, and it is 0 in other cases.

[0072] In step S103, based on the deleted data, a preset kernel function is applied to input matrices of different orders to convert the input matrices into output matrices of corresponding orders.

[0073] It can be understood that the preset kernel function in the embodiment of the present application can be N kernel functions.

[0074] During the actual execution process, the embodiment of the present application can define N kernel functions at the kernel function layer, and apply the kernel functions to input matrices of corresponding orders, which can effectively extract the key features in the input matrix and convert it into an output matrix of corresponding order through nonlinear mapping.

[0075] In step S104, the output matrix is ​​processed according to a preset multi-layer perceptron to generate a final prediction result.

[0076] It can be understood that the output matrices in the embodiment of the present application are further processed in the MLP after splicing. Unlike the traditional MLP, the MLP used in the present application does not contain a nonlinear activation function.

[0077] In the actual execution process, the embodiment of the present application splices the output matrix through the multi-layer perceptron layer, and outputs it through a purely linear multi-layer perceptron to obtain the final classification result, that is, the output matrix is ​​directly fused through the linear layer using the linear superposition method, and after linear superposition of the corresponding order, the prediction result is finally obtained and output. The embodiment of the present application adopts a linear multi-layer perceptron to avoid the black box effect brought by the traditional activation function, improve the interpretability of the model, and facilitate users to understand the working principle of the model.

[0078] The following is a mathematical expression of the model structure: y(k) = b + k1 × y (1) (k)+…+k n ×y (n) (k), where for any order of output, y (m) (k) is obtained through the above operations in this study: (m) (k) = h m ×X N It is worth noting that the × here is redefined through the einsum operation, and its einsum description is: a…,b…→ab.

[0079] Accurate prediction of long-sequence loads in power systems: The model is required to be able to handle different lengths of long-sequence load input data, verify the performance of the proposed model at different prediction steps, and fully capture the dynamic characteristics of the power system. By effectively utilizing these variables, the model should achieve high-precision prediction results and ensure reliable prediction results under various operating conditions.

[0080] It should be noted that, due to the omission of the nonlinear activation function, the prediction process in this application has a clear mathematical expression, which can intuitively show the impact of each input variable on the prediction result, making the model more transparent and reliable, and convenient for users to understand and apply.

[0081] In step S105, based on the final prediction result, the preset power system long sequence load forecasting model is evaluated using the mean square error and the mean absolute error to generate an evaluation result.

[0082] It can be understood that in order to evaluate the point prediction performance of this model, the embodiment of the present application introduces two error indicators, MAE and MSE, to evaluate the prediction accuracy of the model, where MAE is the mean of the absolute error and is a general error measurement indicator. MSE is the mean of the square of the error between the predicted value and the actual observed value, and it is more sensitive to sample points with high errors.

[0083] In one embodiment of the present application, the calculation formula of the mean square error is:

[0084]

[0085] Among them, y i For the prediction target, is the predicted value;

[0086] The formula for calculating the mean absolute error is:

[0087]

[0088] Among them, y i For the prediction target, is the predicted value.

[0089] In step S106, based on the evaluation results, the relationship and influence between the internal variables of the preset power system long sequence load forecasting model are analyzed to generate analysis results, and the optimization scheme of the preset power system long sequence load forecasting model is determined according to the analysis results.

[0090] As a possible implementation method, the embodiment of the present application has an interpretability analysis function, which can reveal the relationship between the internal variables of the preset power system long-sequence load forecasting model and its impact on the forecast results. Based on these analysis results, users can formulate corresponding optimization plans to further improve the performance of the model.

[0091] It should be noted that the interpretability of the model in the embodiment of the present application is that it can not only make accurate predictions, but also intuitively express the relationship between the variables. The model parameters can reflect the specific contribution and mutual influence of each variable on the prediction results, help users understand the internal mechanism of the model, and can perform detailed impact analysis on the variables, quantifying the impact of different variables on the predicted future power system load. Through techniques such as feature importance analysis and partial dependence diagrams, the role and function of each variable in the prediction are clarified.

[0092] Furthermore, the optimization strategy of the power system load system is proposed and verified: using the variable relationship and impact analysis provided by the model, specific optimization plans are formulated, such as adjusting the supply and demand relationship of the power system. Through simulation and practical application, the effectiveness of these optimization strategies is verified, the efficiency and stability of the power system are improved, and the stable operation of the system equipment under the optimal load is ensured.

[0093] The embodiments of the present application can surpass the best existing models in multiple data sets and significantly reduce the computational requirements required for fitting.

[0094] Specifically, it can be combined Figures 2 to 4 As shown, the working principle of the long sequence load forecasting method for the power system in the embodiment of the present application is described in detail with a specific embodiment.

[0095] like Figure 2 As shown, the core prediction model in the embodiment of the present application includes: Multilayer Perceptron (MLP), Kronecker product layer, kernel function, rotation matrix mask, which aims to achieve accurate prediction of power system load, intuitively express variable relationships, conduct detailed variable impact analysis, and provide effective system optimization strategies to further help the power system achieve supply and demand balance, and help power companies achieve more reasonable power distribution through more accurate load forecasting.

[0096] Further, Figure 3 This is a schematic diagram of the overall algorithm of the prediction model. Figure 4 The power system load data set is first preprocessed by the preprocessing module and then input into the prediction model. After the model processes the data, it predicts and outputs the load parameters of the power system at future times.

[0097] According to the long-sequence load forecasting method for the power system proposed in the embodiment of the present application, the large model can be combined with Kronecker product, rotation matrix mask and other technologies to achieve accurate prediction of the power system load, intuitively express the variable relationship, conduct detailed variable impact analysis, and provide effective system optimization strategies, further helping the power system to achieve supply and demand balance, and helping power companies to achieve more reasonable power distribution through more accurate load forecasting. As a result, the problems in the related technology that the existing large models in time series prediction are huge, the nonlinear characteristics are expressed through complex activation functions, resulting in black boxing, and the difficulty in intuitively expressing the relationship between variables are solved, which limits the effectiveness and interpretability of the model in practical applications and affects the user's understanding and trust in the model prediction results.

[0098] Next, a long sequence load forecasting device for a power system proposed according to an embodiment of the present application will be described with reference to the accompanying drawings.

[0099] Figure 5 It is a structural schematic diagram of a long sequence load forecasting device for an electric power system according to an embodiment of the present application.

[0100] like Figure 5 As shown, the long sequence load forecasting device 10 of the power system includes:

[0101] Specifically, the construction module 100 is used to perform multi-level operations on the input matrix using the Kronecker product layer to construct a high-dimensional matrix.

[0102] The sorting module 200 is used to sort the high-dimensional matrix using the rotation matrix mask to generate a high-dimensional matrix that meets the preset conditions, and delete high-order duplicate items using the high-dimensional matrix that meets the preset conditions to generate deleted data.

[0103] The input module 300 is used to apply a preset kernel function to input matrices of different orders based on the deleted data to convert the input matrices into output matrices of corresponding orders.

[0104] The processing module 400 is used to process the output matrix according to a preset multi-layer perceptron to generate a final prediction result.

[0105] The evaluation module 500 is used to evaluate the preset power system long sequence load forecasting model based on the final forecasting result by using the mean square error and the mean absolute error to generate an evaluation result.

[0106] The prediction module 600 is used to analyze the relationship and influence between the internal variables of the preset power system long sequence load prediction model based on the evaluation results to generate analysis results, and determine the optimization scheme of the preset power system long sequence load prediction model according to the analysis results.

[0107] Optionally, in one embodiment of the present application, the long sequence load forecasting device 10 for the power system further includes: a processing module and a generating module.

[0108] Among them, the removal module is used to perform data cleaning, missing value filling and outlier processing on the load sequence data set of the power system before using the Kronecker product layer to perform multi-level operations on the input matrix to construct a high-dimensional matrix, so as to remove the erroneous information in the load sequence data set and obtain the removed data.

[0109] The generation module is used to divide the load sequence data set into a training set and a test set based on the removed data, and normalize the training set and the test set to obtain processed data.

[0110] Optionally, in one embodiment of the present application, the calculation formula of the high-dimensional matrix is:

[0111] X N (i,j,…,n)=X e (i)×X e (j)×…×X e (n)'

[0112] Among them, X e (n) is the input matrix X after dimension transformation, X N The matrix is ​​composed of N X e The result obtained by the matrix Kronecker product.

[0113] Optionally, in one embodiment of the present application, the calculation formula for the elements in the rotation matrix mask is:

[0114]

[0115] Among them, i, j, n are the position codes of the corresponding dimensions of the mask matrix, and m represents the mask matrix.

[0116] Optionally, in one embodiment of the present application, the calculation formula of the mean square error is:

[0117]

[0118] Among them, y i For the prediction target, is the predicted value;

[0119] The formula for calculating the mean absolute error is:

[0120]

[0121] Among them, y i For the prediction target, is the predicted value.

[0122] It should be noted that the above explanation of the embodiment of the long-sequence load forecasting method for the power system is also applicable to the long-sequence load forecasting device for the power system of this embodiment, and will not be repeated here.

[0123] According to the long-sequence load forecasting device for the power system proposed in the embodiment of the present application, accurate long-sequence load forecasting can be achieved by combining a large model with Kronecker product, rotation matrix mask and other technologies, intuitively expressing variable relationships, and providing effective system optimization strategies, and there are significant improvements in feature extraction capabilities, model robustness, model interpretability and long-sequence prediction, providing a more effective and reliable method for power system load forecasting. Thus, the problems in the related technology that the existing large models in time series prediction are huge, the nonlinear characteristics are expressed through complex activation functions, resulting in black boxing, and the difficulty in intuitively expressing cross-variable relationships are solved, which limits the effectiveness and interpretability of the model in practical applications and affects the user's understanding and trust in the model prediction results.

[0124] Figure 6 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. The electronic device may include:

[0125] A memory 601 , a processor 602 , and a computer program stored in the memory 601 and executable on the processor 602 .

[0126] When the processor 602 executes the program, the long sequence load forecasting method for the power system provided in the above embodiment is implemented.

[0127] Furthermore, the electronic device further comprises:

[0128] The communication interface 603 is used for communication between the memory 601 and the processor 602 .

[0129] The memory 601 is used to store computer programs that can be executed on the processor 602 .

[0130] The memory 601 may include a high-speed RAM memory, and may also include a non-volatile memory (non-volatile memory), such as at least one disk memory.

[0131] If the memory 601, the processor 602 and the communication interface 603 are implemented independently, the communication interface 603, the memory 601 and the processor 602 can be connected to each other through a bus and communicate with each other. The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component (PCI) bus or an Extended Industry Standard Architecture (EISA) bus. The bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 6 Only one thick line is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0132] Optionally, in a specific implementation, if the memory 601, the processor 602 and the communication interface 603 are integrated on a chip, the memory 601, the processor 602 and the communication interface 603 can communicate with each other through an internal interface.

[0133] The processor 602 may be a central processing unit (CPU), or an application specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of the present application.

[0134] This embodiment also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the above-mentioned method for long-sequence load forecasting of a power system.

[0135] An embodiment of the present application also provides a computer program product, on which a computer program is stored, and when the program is executed by a processor, the long sequence load forecasting method for the power system as described above is implemented.

[0136] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or N embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, without contradiction.

[0137] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include at least one of the features. In the description of this application, "N" means at least two, such as two, three, etc., unless otherwise clearly and specifically defined.

[0138] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, fragment or portion of code comprising one or N executable instructions for implementing the steps of a custom logical function or process, and the scope of the preferred embodiments of the present application includes alternative implementations in which functions may not be performed in the order shown or discussed, including performing functions in a substantially simultaneous manner or in reverse order depending on the functions involved, which should be understood by technicians in the technical field to which the embodiments of the present application belong.

[0139] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in combination with these instruction execution systems, devices or apparatuses. For the purpose of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in combination with these instruction execution systems, devices or apparatuses. More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or N wirings (electronic devices), a portable computer disk box (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), a fiber optic device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be paper or other suitable medium on which the program is printed, since the program may be obtained electronically by optically scanning the paper or other medium and then editing, interpreting or processing in other suitable ways as necessary and then storing it in a computer memory.

[0140] It should be understood that the various parts of the present application can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiment, the N steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0141] A person skilled in the art may understand that all or part of the steps in the method for implementing the above-mentioned embodiment may be completed by instructing related hardware through a program, and the program may be stored in a computer-readable storage medium, which, when executed, includes one or a combination of the steps of the method embodiment.

[0142] In addition, each functional unit in each embodiment of the present application may be integrated into a processing module, or each unit may exist physically separately, or two or more units may be integrated into one module. The above-mentioned integrated module may be implemented in the form of hardware or in the form of a software functional module. If the integrated module is implemented in the form of a software functional module and sold or used as an independent product, it may also be stored in a computer-readable storage medium.

[0143] The storage medium mentioned above may be a read-only memory, a magnetic disk or an optical disk, etc. Although the embodiments of the present application have been shown and described above, it can be understood that the above embodiments are exemplary and cannot be understood as limiting the present application. A person of ordinary skill in the art may change, modify, replace and modify the above embodiments within the scope of the present application.

Claims

1. A method for long-sequence load forecasting of a power system, characterized in that: The following steps are involved: Use Kronecker product layers to perform multi-level operations on the input matrix to construct a high-dimensional matrix; Arranging the high-dimensional matrix using a rotation matrix mask to generate a high-dimensional matrix that meets preset conditions, and deleting high-order duplicate items using the high-dimensional matrix that meets the preset conditions to generate deleted data; Based on the deleted data, applying a preset kernel function to input matrices of different orders to convert the input matrices into output matrices of corresponding orders; Processing the output matrix according to a preset multi-layer perceptron to generate a final prediction result; Based on the final prediction result, a preset power system long sequence load forecasting model is evaluated using mean square error and mean absolute error to generate an evaluation result; Based on the evaluation results, the relationship and influence between the internal variables of the preset power system long sequence load forecasting model are analyzed to generate analysis results, and the optimization scheme of the preset power system long sequence load forecasting model is determined according to the analysis results.

2. The method according to claim 1, characterized in that Before using the Kronecker product layer to perform multi-level operations on the input matrix to construct a high-dimensional matrix, it also includes: Performing data cleaning, missing value filling and outlier processing on the load sequence data set of the power system respectively, so as to remove erroneous information in the load sequence data set and obtain the removed data; Based on the removed data, the load sequence data set is divided into a training set and a test set, and the training set and the test set are normalized to obtain the processed data.

3. The method according to claim 1, characterized in that The calculation formula of the high-dimensional matrix is: X N (i,j,…,n)=X e (i)×X e (j)×…×X e (n)' Among them, X e (n) is the input matrix X after dimension transformation, X N The matrix is ​​composed of N X e The result obtained by the matrix Kronecker product.

4. The method according to claim 1, characterized in that The calculation formula of the elements in the rotation matrix mask is: Among them, i, j, n are the position codes of the corresponding dimensions of the mask matrix, and m represents the mask matrix.

5. The method according to claim 1, characterized in that The calculation formula of the mean square error is: Among them, y i For the prediction target, is the predicted value; The calculation formula of the mean absolute error is: Among them, y i For the prediction target, is the predicted value.

6. A long sequence load forecasting device for a power system, characterized in that: include: A building block for performing multi-level operations on the input matrix using Kronecker product layers to construct a high-dimensional matrix; A sorting module, used to sort the high-dimensional matrix using a rotation matrix mask to generate a high-dimensional matrix that meets preset conditions, and use the high-dimensional matrix that meets the preset conditions to delete high-order duplicate items to generate deleted data; An input module, used for applying a preset kernel function to input matrices of different orders based on the deleted data to convert the input matrices into output matrices of corresponding orders; A processing module, used for processing the output matrix according to a preset multi-layer perceptron to generate a final prediction result; An evaluation module, used for evaluating a preset power system long sequence load forecasting model based on the final forecasting result by using a mean square error and a mean absolute error to generate an evaluation result; A prediction module is used to analyze the relationship and influence between the internal variables of the preset power system long sequence load prediction model based on the evaluation results to generate an analysis result, and determine the optimization scheme of the preset power system long sequence load prediction model according to the analysis result.

7. The device according to claim 6, characterized in that Also includes: A removal module is used for performing data cleaning, missing value filling and outlier processing on the load sequence data set of the power system respectively before performing multi-level operations on the input matrix using the Kronecker product layer to construct a high-dimensional matrix, so as to remove erroneous information in the load sequence data set and obtain the removed data; A generating module is used to divide the load sequence data set into a training set and a test set based on the removed data, and perform normalization processing on the training set and the test set to obtain the processed data.

8. An electronic device, characterized in that: include: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the long sequence load forecasting method for the power system as described in any one of claims 1 to 5.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: The program is executed by a processor to implement the long sequence load forecasting method for the power system as described in any one of claims 1 to 5.

10. A computer program product, comprising a computer program, characterized in that The computer program is executed to implement the long sequence load forecasting method for the power system as described in any one of claims 1-5.

Images