Substation metering system state prediction method, device and electronic equipment

Abstract

The embodiment of the application provides a kind of substation metering system state prediction method, device and electronic equipment, it is related to sampling algorithm technical field.The method comprises: obtaining the historical data of the substation including environmental state quantity and corresponding metering quantity;The historical data is preprocessed;The preprocessed data is processed by closed loop clustering algorithm to obtain data classification under different scenarios, and the preprocessed data after data classification is divided into training set and validation set;The training set is used to train the combined model, and the validation set is used to verify the trained combined model;The combined model includes random forest algorithm, convolutional neural network algorithm and echo state network algorithm;The combined model after verification is used to output corresponding prediction result based on input environmental state quantity and corresponding metering quantity.The embodiment provided in the application effectively improves the intelligent level of substation operation and maintenance.

Description

Technical Field

[0001] This application relates to the field of sampling algorithm technology, specifically to a substation metering system state prediction method, a substation metering system state prediction device, an electronic device, a storage medium, and a computer program product. Background Technology

[0002] With the development of intelligent power systems, substations, as an important component of the power system, require real-time monitoring and assessment of their operating status for the safe and stable operation of the power grid. Traditional substation monitoring methods mainly rely on manual periodic inspections and simple monitoring equipment, which suffer from problems such as long detection cycles, data lag, and insufficient assessment accuracy, making it difficult to meet the needs of modern power grids for efficient and accurate monitoring.

[0003] During substation operation, various environmental factors and equipment conditions significantly impact system stability, such as temperature, humidity, load conditions, and equipment aging. Real-time acquisition and analysis of these environmental parameters and metering data are crucial for preventing equipment failures and optimizing operation and maintenance decisions. However, the complex substation environment, diverse data sources, and massive data volumes make efficient data processing and real-time status assessment difficult using traditional monitoring methods. Summary of the Invention

[0004] The purpose of this application is to provide a method, device, and electronic equipment for predicting the status of a substation metering system. By combining advanced deep learning algorithms, a predictive model and real-time evaluation system for equipment status are established to achieve real-time monitoring, prediction, and evaluation of the status of substation equipment, thereby at least solving some of the problems in the background art.

[0005] To achieve the above objectives, this application provides a substation metering system state prediction method. The method includes: acquiring historical data of the substation, including environmental state quantities and their corresponding metering quantities; preprocessing the historical data to obtain preprocessed data; processing the preprocessed data using a closed-loop clustering algorithm to obtain data classifications under different scenarios, and dividing the preprocessed data into a training set and a validation set; inputting the data in the training set into a random forest algorithm to obtain corresponding outputs, and fusing the input and output data of the random forest algorithm to obtain a processed training set; training a combined model using the processed training set, and validating the trained combined model using the validation set; the combined model includes a convolutional neural network part and an echo state network part; the validated combined model is used to output corresponding prediction results based on the input environmental state quantities and their corresponding metering quantities.

[0006] Optionally, acquiring historical data of the substation, including environmental state quantities and their corresponding metering quantities, includes: collecting historical data of the substation, wherein the historical data includes environmental state quantities and their corresponding metering quantities; distinguishing the environmental state quantities and their corresponding metering quantities by different scenarios, and combining the environmental state quantities and metering quantities with corresponding relationships into a state matrix form.

[0007] Optionally, the historical data is preprocessed to obtain preprocessed data, including: constructing a device state matrix based on the historical data; determining the numerical range of the elements in the device state matrix and identifying values ​​outside the preset numerical range as outliers; deleting the outliers and then using Lagrange interpolation to generate new values ​​to fill the positions of the deleted outliers; standardizing the elements in the device state matrix through linear transformation; and using principal component analysis to reduce the dimensionality of the standardized device state matrix to obtain the preprocessed data.

[0008] Optionally, the preprocessed data is processed using a closed-loop clustering algorithm to obtain data classifications for different scenarios, including: inputting the preprocessed data into a closed-loop clustering algorithm model to divide all scenarios into several categories; based on the metrics of scenarios within the same category after clustering as prediction objects, training convolutional neural network models with the goal of minimizing the root mean square error to obtain prediction models that correspond one-to-one with the clustered categories; processing the data for each scenario in all scenarios as follows: using the prediction model to make predictions to obtain multiple sets of prediction results equal to the number of prediction models; if one of the multiple sets of prediction results has the smallest root mean square error, then moving the scenario corresponding to that set of prediction results to the corresponding category; repeating the aforementioned steps of obtaining prediction models and data processing for each scenario until the termination condition is met, the termination condition including the clustering result no longer changing or reaching the maximum number of iterations; using the clustering result at the termination condition as the data classification for the different scenarios.

[0009] Optionally, the method further includes: generating cluster labels based on the clustering results when the termination condition is met and adding them to the preprocessed data; and performing imbalance processing on the preprocessed data after data classification under different scenarios.

[0010] Optionally, the random forest algorithm in the combined model is configured as follows: the data in the training set is input into the random forest algorithm to obtain the corresponding output, and the input data and output data of the random forest algorithm are fused to obtain fused data; the convolutional neural network algorithm in the combined model is configured as follows: the feature extraction result is obtained based on the input fused data; the echo state network algorithm in the combined model is configured as follows: a state matrix is ​​generated based on the feature extraction result, and the state matrix is ​​used to obtain the final output result through the output weight matrix.

[0011] Optionally, the input and output data of the random forest algorithm are fused to obtain fused data, including: concatenating the input and output data of the random forest algorithm to obtain fused data; the echo state network algorithm is trained through the following steps: initializing the weight matrix and hyperparameters in the echo state network algorithm; the data in the training set is processed by the convolutional neural network part to generate corresponding feature extraction results, the feature extraction results are input to the echo state network part to train the connection weights from the hidden layer to the output layer; the hidden layer obtains the final output result through the output weight matrix of the output layer.

[0012] Optionally, the validated combined model is used to output a prediction result based on the input environmental state quantity and its corresponding measurement quantity, including: acquiring real-time data of the environmental state quantity and its corresponding measurement quantity; constructing the real-time data into a state matrix form; inputting the real-time data in the state matrix form into the validated combined model, and using the output of the validated combined model as the corresponding prediction result.

[0013] Optionally, after outputting the corresponding prediction results based on the input environmental state quantities and their corresponding metering quantities, the method further includes: fitting the prediction results output by the verified combined model into a dynamic prediction curve; calculating the residual between the dynamic prediction curve and the actual curve obtained by fitting real-time data within a set time interval, and obtaining a residual dynamic change curve based on the residual; and determining the equipment status of the substation according to the fluctuation of the residual dynamic change curve.

[0014] This application also provides a substation metering system condition prediction device, which includes:

[0015] The system includes a data acquisition module for acquiring historical data of the substation, including environmental state quantities and their corresponding metering quantities; a preprocessing module for preprocessing the historical data to obtain preprocessed data; a data clustering module for processing the preprocessed data using a closed-loop clustering algorithm to obtain data classifications under different scenarios, and dividing the preprocessed data into a training set and a validation set; and a combined model module for training a combined model using the training set and validating the trained combined model using the validation set. The combined model includes a random forest algorithm, a convolutional neural network algorithm, and an echo state network algorithm. The validated combined model is used to output corresponding prediction results based on the input environmental state quantities and their corresponding metering quantities.

[0016] Optionally, acquiring historical data of the substation, including environmental state quantities and their corresponding metering quantities, includes: collecting historical data of the substation, wherein the historical data includes environmental state quantities and their corresponding metering quantities; distinguishing the environmental state quantities and their corresponding metering quantities by different scenarios, and combining the environmental state quantities and metering quantities with corresponding relationships into a state matrix form.

[0017] Optionally, the historical data is preprocessed to obtain preprocessed data, including: constructing a device state matrix based on the historical data; determining the numerical range of the elements in the device state matrix and identifying values ​​outside the preset numerical range as outliers; deleting the outliers and then using Lagrange interpolation to generate new values ​​to fill the positions of the deleted outliers; standardizing the elements in the device state matrix through linear transformation; and using principal component analysis to reduce the dimensionality of the standardized device state matrix to obtain the preprocessed data.

[0018] Optionally, the preprocessed data is processed using a closed-loop clustering algorithm to obtain data classifications for different scenarios, including: inputting the preprocessed data into a closed-loop clustering algorithm model to divide all scenarios into several categories; based on the metrics of scenarios within the same category after clustering as prediction objects, training convolutional neural network models with the goal of minimizing the root mean square error to obtain prediction models that correspond one-to-one with the clustered categories; processing the data for each scenario in all scenarios as follows: using the prediction model to make predictions to obtain multiple sets of prediction results equal to the number of prediction models; if one of the multiple sets of prediction results has the smallest root mean square error, then moving the scenario corresponding to that set of prediction results to the corresponding category; repeating the aforementioned steps of obtaining prediction models and data processing for each scenario until the termination condition is met, the termination condition including the clustering result no longer changing or reaching the maximum number of iterations; using the clustering result at the termination condition as the data classification for the different scenarios.

[0019] Optionally, the device further includes an imbalance processing module, which is used to: generate cluster labels based on the clustering results when the termination condition is met and add them to the preprocessed data; and perform imbalance processing on the preprocessed data after data classification under different scenarios.

[0020] Optionally, the random forest algorithm in the combined model is configured as follows: the data in the training set is input into the random forest algorithm to obtain the corresponding output, and the input data and output data of the random forest algorithm are fused to obtain fused data; the convolutional neural network algorithm in the combined model is configured as follows: the feature extraction result is obtained based on the input fused data; the echo state network algorithm in the combined model is configured as follows: a state matrix is ​​generated based on the feature extraction result, and the state matrix is ​​used to obtain the final output result through the output weight matrix.

[0021] Optionally, the input and output data of the random forest algorithm are fused to obtain fused data, including: concatenating the input and output data of the random forest algorithm to obtain fused data; the echo state network algorithm is trained through the following steps: initializing the weight matrix and hyperparameters in the echo state network algorithm; the data in the training set is processed by the convolutional neural network part to generate corresponding feature extraction results, the feature extraction results are input to the echo state network part to train the connection weights from the hidden layer to the output layer; the hidden layer obtains the final output result through the output weight matrix of the output layer.

[0022] Optionally, the validated combined model is used to output a prediction result based on the input environmental state quantity and its corresponding measurement quantity, including: acquiring real-time data of the environmental state quantity and its corresponding measurement quantity; constructing the real-time data into a state matrix form; inputting the real-time data in the state matrix form into the validated combined model, and using the output of the validated combined model as the corresponding prediction result.

[0023] Optionally, the device further includes a state determination module, which is used to: after outputting a prediction result based on the input environmental state quantity and its corresponding metering quantity, fit the prediction result output by the verified combined model into a dynamic prediction curve; calculate the residual between the dynamic prediction curve and the actual curve obtained by fitting real-time data within a set time interval, and obtain a residual dynamic change curve based on the residual; and determine the equipment state of the substation according to the fluctuation of the residual dynamic change curve.

[0024] This application also provides an electronic device, including: at least one processor; and a memory connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the at least one processor implements the aforementioned substation metering system state prediction method by executing the instructions stored in the memory.

[0025] This application also provides a machine-readable storage medium storing instructions that, when executed by a processor, configure the processor to perform the aforementioned substation metering system state prediction method.

[0026] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the aforementioned substation metering system state prediction method.

[0027] The above technical solution has the following beneficial effects:

[0028] (1) By using closed-loop clustering algorithm and random forest algorithm, the accuracy and robustness of clustering results are significantly improved, thereby improving the stability of subsequent models.

[0029] (2) The model combining random forest with CNN-ESN extracts high-dimensional features through CNN, processes time series data through ESN, and improves model stability through random forest. It can show higher prediction accuracy and robustness in complex data and sequence prediction tasks.

[0030] (3) It effectively improves the level of intelligent operation and maintenance of substations and enhances the safety and reliability of power grid operation, which has important application value and promotion prospects.

[0031] Other features and advantages of the embodiments of this application will be described in detail in the following detailed description section. Attached Figure Description

[0032] The accompanying drawings are provided to further illustrate the embodiments of this application and form part of the specification. They are used together with the following detailed description to explain the embodiments of this application, but do not constitute a limitation on the embodiments of this application. In the drawings:

[0033] Figure 1 This illustration schematically shows the steps of a substation metering system state prediction method according to an embodiment of this application.

[0034] Figure 2 This illustration schematically shows a technical route diagram of the substation metering system state prediction method according to the embodiments of this application;

[0035] Figure 3 The illustration shows a schematic diagram of the implementation of the data preprocessing step in the substation metering system state prediction method according to the embodiments of this application;

[0036] Figure 4 The illustration shows a schematic diagram of the implementation of the closed-loop clustering algorithm in the substation metering system state prediction method according to the embodiments of this application;

[0037] Figure 5This illustration schematically shows a data processing diagram of the combined model in the substation metering system state prediction method according to an embodiment of this application;

[0038] Figure 6 This illustration schematically shows a structural diagram of a substation metering system state prediction device according to an embodiment of this application;

[0039] Figure 7 The diagram schematically illustrates the internal structure of an electronic device according to an embodiment of this application. Detailed Implementation

[0040] The specific embodiments of this application will be described in detail below with reference to the accompanying drawings. It should be understood that the specific embodiments described herein are for illustration and explanation only and are not intended to limit the embodiments of this application.

[0041] Figure 1 The illustration shows a schematic diagram of the steps in a substation metering system state prediction method according to an embodiment of this application. For example... Figure 1 As shown, the method includes:

[0042] S01. Obtain historical data of the substation, including environmental status quantities and their corresponding metering quantities;

[0043] S02. Preprocess the historical data to obtain preprocessed data;

[0044] S03. The preprocessed data is processed by a closed-loop clustering algorithm to obtain data classifications under different scenarios, and the preprocessed data after data classification is divided into a training set and a validation set.

[0045] S04. The combined model is trained using the training set, and the trained combined model is validated using the validation set; the combined model includes the random forest algorithm, the convolutional neural network algorithm, and the echo state network algorithm; the validated combined model is used to output the corresponding prediction results based on the input environmental state variables and their corresponding quantifiers.

[0046] In the above embodiments, steps S01 to S04 are mainly implemented in the model construction scenario. The model usage steps in this application mainly involve deploying the validated combined model online, which is then used to output corresponding prediction results based on the input environmental state variables and their corresponding measurement variables.

[0047] Through the above implementation methods, data preprocessing and closed-loop clustering algorithms effectively classified different scenarios of the training data, improving the scenario-specificity of the training process. Random forest algorithms were used for data classification, enhancing model stability. Simultaneously, the CNN-ESN model, combining convolutional neural networks and echo state networks, demonstrated higher prediction accuracy and robustness in complex data and sequence prediction tasks. This improved the accuracy of prediction results. The application of these technologies in substation equipment monitoring and maintenance effectively enhances the intelligence level of substation operation and maintenance, improves the safety and reliability of power grid operation, and has significant application value and promising prospects for widespread adoption.

[0048] Figure 2 This illustration schematically shows a technical route diagram of the substation metering system state prediction method according to an embodiment of this application. For example... Figure 2 As shown, the substation metering system state prediction method in this embodiment mainly includes: data collection step, data preprocessing step, cluster analysis step, unbalanced data processing step, random forest-neural network algorithm training, residual calculation step, and implementation evaluation step.

[0049] In some embodiments of this application, acquiring historical data of a substation, including environmental state quantities and their corresponding metering quantities, includes: collecting historical data of the substation, wherein the historical data includes environmental state quantities and their corresponding metering quantities; distinguishing the environmental state quantities and their corresponding metering quantities by different scenarios, and combining the environmental state quantities and metering quantities with corresponding relationships into a state matrix form. For example, collecting the environmental state quantities and metering quantities of the substation, distinguishing by different scenarios, collecting the environmental state quantities and metering quantities under each scenario, and combining them into a state matrix form, denoted as M, assuming there are k sets of scenarios and corresponding variables.

[0050] In some embodiments of this application, the historical data is preprocessed to obtain preprocessed data. Figure 3 The illustration shows a schematic diagram of the implementation of the data preprocessing step in the substation metering system state prediction method according to the embodiments of this application. For example... Figure 3 As shown, this includes: constructing a device status matrix based on the historical data. For example, in establishing the device status matrix, the environmental state quantities and measurement quantities considering the time series are denoted as a, a, and a, respectively. i,p and b j,pLet , where 'a' represents environmental state variables, of which there are i; 'b' represents measurement variables, of which there are j, where i and j range from [0, k]; and 'p' represents different time points within a day. To ensure a reasonable composition of the device state matrix, the sampling nodes for each time period within a day are consistent with the final determined number of scenarios k. The final device state matrix structure is as follows:

[0051]

[0052] The final device state matrix M is a k-order square matrix. and Let k represent the environmental state variables and the measurement variables, where k is the number of scenarios, i is the number of environmental state variables, j is the number of measurement variables, and p is the time series.

[0053] The elements in the device status matrix are evaluated for numerical range, and values ​​outside the preset range are identified as outliers. For example, this embodiment uses a simple statistical analysis method, performing descriptive analysis on the maximum and minimum values ​​of the dataset to identify unreasonable data, thereby filtering and locating outliers. The maximum and minimum values ​​are within a reasonable range, determined by reality and the actual condition of the device.

[0054]

[0055] Where max(a) i ),min(a i ) are respectively represented by data a i The maximum and minimum values; max(b i ),min(b i ) are respectively represented as data b i The maximum and minimum values.

[0056] After deleting the outliers, new values ​​are generated using Lagrange interpolation to fill the positions of the deleted outliers. The principle of Lagrange interpolation is to substitute the points corresponding to the missing data into the Lagrange interpolation formula to obtain approximate data at that location. The elements in the device state matrix are then standardized using linear transformations; for example:

[0057]

[0058] Where k represents the dimension of the data, a m a n b m b n This represents the data within the dataset, with the missing values ​​on the left side of the equals sign.

[0059] The elements in the device state matrix are standardized by linear transformation. For example, data standardization can be performed using the MAX-MIN standardization method. MAX-MIN standardization involves repeating the linear transformation process on the data so that each eigenvalue falls within the range [0, 1]. The solution for MAX-MIN standardization is as follows:

[0060]

[0061] In the formula: max(a i ), min(a i ), max(b i ), min(b i The maximum and minimum values ​​of the data are represented by . After standardization, the data in the state matrix are within [0,1], and the matrix is ​​denoted as M'.

[0062] Principal component analysis is used to reduce the dimensionality of the standardized equipment state matrix to obtain the preprocessed data. For example, the covariance matrix of the standardized data is calculated: M'M' T Next, eigenvalue decomposition is performed on the covariance matrix to obtain several eigenvalues. Then, the eigenvectors corresponding to the x largest eigenvalues ​​are selected as principal components. These principal components explain the largest variance in the data. The selected x eigenvectors are combined to form a dimensionality-reduced matrix P, where each column is an eigenvector.

[0063] P = [v1, v2, ..., v k ]

[0064] Multiply the standardized data M' by the dimension reduction matrix P to project the data from the high-dimensional space to the low-dimensional space, and denote the data in the standardized and dimension-reduced matrices as c. i,j The matrix is ​​denoted as N:

[0065] N = M'P.

[0066] The above methods achieve data preprocessing. Compared with the limited processing capacity of existing data preprocessing techniques for large datasets, the implementation method in this application further improves data quality and feature utilization efficiency through anomaly filtering and feature dimensionality reduction, which helps to build higher-performance machine learning models. This improvement has significant advantages in processing complex datasets, especially scenarios containing noisy and high-dimensional data.

[0067] In some embodiments of this application, the preprocessed data is processed using a closed-loop clustering algorithm to obtain data classifications for different scenarios, including: inputting the preprocessed data into a closed-loop clustering algorithm model to divide all scenarios into several categories; based on the measurement parameters of scenarios within the same category after clustering as the prediction object, training convolutional neural network models with the goal of minimizing the root mean square error to obtain prediction models that correspond one-to-one with the clustered categories; processing the data for each scenario in all scenarios as follows: using the prediction model to make predictions to obtain multiple sets of prediction results equal to the number of prediction models; if one of the multiple sets of prediction results has the smallest root mean square error, then moving the scenario corresponding to that set of prediction results to the corresponding category; repeating the aforementioned steps of obtaining prediction models and data processing for each scenario until the termination condition is reached, the termination condition including the clustering result no longer changing or reaching the maximum number of iterations; and using the clustering result at the termination condition as the data classification for the different scenarios. Figure 4 The illustration schematically shows an implementation diagram of the closed-loop clustering algorithm in the substation metering system state prediction method according to an embodiment of this application. For example... Figure 4 As shown, it can be implemented, for example, by the following steps:

[0068] Step a, Initialization. Input matrix N into the improved closed-loop K-means clustering algorithm model, and use the K-means algorithm to divide all k scenes in the region into v categories (1≤v≤k);

[0069] Step b, Prediction. Based on the clustering results, using the metrics of scenes within the same category as the prediction objects, train convolutional neural network models with the objective of minimizing the root mean squared error (RMSE) to obtain v prediction models corresponding to v categories:

[0070]

[0071] Among them, c i,j For the data in the characteristic matrix T, This represents the average value of the data.

[0072] Step c, Feedback. For each scenario, apply v prediction models to obtain v sets of prediction results. If the r-th group (1≤r≤v) has the smallest RMSE, then move the scenario to the r-th category. This process is repeated for all k scenarios to obtain new clustering results.

[0073] Step d: Iteration. Repeat steps b and c until the clustering results no longer change or the maximum number of iterations is reached.

[0074] This implementation method uses a closed-loop clustering algorithm for data classification, which improves the stability of classification compared to existing clustering algorithms.

[0075] In some embodiments of this application, the method further includes: generating cluster labels based on the clustering results when the termination condition is met and adding them to the preprocessed data; and performing imbalanced processing on the preprocessed data after classifying data under different scenarios. After clustering is completed, the form of the device state matrix will not change significantly; only a label will be generated in the last column of the matrix to represent each cluster. Then, imbalanced learning is performed on the matrix to address potential data distribution imbalances, and finally, the new state matrix is ​​denoted as T.

[0076] The imbalance handling provided in this implementation improves the predictive ability of subsequent models for a few scenarios, thereby improving the overall model performance.

[0077] In some embodiments of this application, the data in the training set is input into a random forest algorithm to obtain the corresponding output. The input and output data of the random forest algorithm are then fused to obtain a processed training set. This includes: inputting the data in the training set into the random forest algorithm to obtain output data corresponding to the input data, where the output data is the predicted output of the input data; and concatenating the input and output data of the random forest algorithm using a concatenation method to obtain the processed training set. For example, after clustering, the new state matrix T has been labeled according to the clustering results. After dividing the data into a training set T1 and a test set T2, the training set is input into the random forest algorithm. The predicted output of the random forest algorithm on the training data is obtained as the input features of the neural network. These features are usually the prediction results of each tree or the distribution of leaf nodes. After classification by the random forest algorithm, a new feature matrix T is obtained. RF After being combined with the input matrix, the resulting matrix T combined The next stage of the model will be implemented:

[0078] T combined =[T1 T RF ]

[0079] By using the random forest algorithm in this implementation, the classification of training data becomes more accurate and stable, thus improving the stability of subsequent models.

[0080] Figure 5 This diagram schematically illustrates the data processing of a combined model in the substation metering system state prediction method according to an embodiment of this application. For example... Figure 5As shown, the training set is used to train the combined model, and the validation set is used to validate the trained combined model. The combined model includes a random forest algorithm, a convolutional neural network algorithm, and an echo state network algorithm. In some embodiments of this application, the training of each part of the combined model includes: initializing the weight matrix and hyperparameters of the echo state network part in the combined model; initializing the CNN-ESN model includes: designing convolutional layers, pooling layers, etc. of the CNN to extract spatial features of the data; initializing the ESN includes initializing the weight matrix of the ESN, including input weights, echo state weights, and output weights. The hidden layer size of the ESN, i.e., the number of echo state units, and hyperparameters such as connection sparsity are determined. After processing, the data in the training set is processed by the convolutional neural network part to generate corresponding feature extraction results. The feature extraction results are input to the echo state network part to train the connection weights from the hidden layer to the output layer; T combined Input a CNN-ESN model, train the CNN-ESN model, and extract high-dimensional features. The hidden layers obtain the final output result through the output weight matrix of the output layer. For example, after the input training data is processed by the CNN, a feature extraction result X will be generated. cnn The input features are fed into the ESN model to update the dynamic state of the ESN and generate a state matrix.

[0081] h(t) = tanh(W) in X cnn (t))+Wh(t-1))

[0082] Among them, X cnn (t) are the input features, h(x) are the hidden states of the ESN, and W in The input weights are h(t), and the echo state weights are W(t). The state matrix h(t) is processed by the output weight matrix W(t). out The final output result Y is obtained:

[0083] Y = W out h(t)

[0084] Where Y is the actual output.

[0085] After training is complete, the performance of the trained CNN-ESN model is evaluated using the validation set T2 and then fine-tuned. The combined evaluated and fine-tuned model can be deployed for application purposes, where it is used to output prediction results based on the input environmental state variables and their corresponding metrics.

[0086] In this embodiment, the CNN-ESN model extracts high-dimensional features through CNN and processes time series data through ESN, which can demonstrate higher prediction accuracy and robustness in complex data and sequence prediction tasks.

[0087] In some embodiments of this application, the validated combined model is used to output corresponding prediction results based on input environmental state quantities and their corresponding metering quantities. This includes: acquiring real-time data of the environmental state quantities and their corresponding metering quantities; constructing the real-time data into a state matrix; inputting the real-time data in the state matrix form into the validated combined model; and using the output of the validated combined model as the corresponding prediction result. When the trained and validated combined model is deployed online, it can obtain prediction results based on real-time data in the same form as the training data. This embodiment provides the deployment and application of a trained and validated combined model. The combined model is applied during substation operation to acquire and analyze these environmental state quantities and metering data in real time. The prediction results output by the combined model are of great significance for preventing equipment failures and optimizing operation and maintenance decisions.

[0088] In some embodiments of this application, after outputting the corresponding prediction results based on the input environmental state quantities and their corresponding metering quantities, the method further includes: fitting the prediction results output by the verified combined model into a dynamic prediction curve; calculating the residual between the dynamic prediction curve and the actual curve obtained by fitting real-time data within a set time interval, and obtaining a residual dynamic change curve based on the residual; determining the equipment status of the substation based on the fluctuation of the residual dynamic change curve. Through dynamic curve fitting, the prediction output of the CNN-ESN model is represented as a dynamic prediction curve y(t), and the actual curve is obtained by fitting real-time data. The residual between the prediction curve and the actual curve is calculated within a set time interval. After calculating the residual, the residual dynamic change curve is plotted. When the amplitude of the residual dynamic curve frequently exceeds a set threshold within a set time interval, it is determined that a problem has occurred in the equipment status, thereby achieving real-time monitoring and early warning functions. This embodiment can automatically compare the prediction output of the CNN-ESN model with the actual data, realizing real-time monitoring, prediction, and evaluation of the substation equipment status. This method can effectively improve the intelligence level of substation operation and maintenance, enhance the safety and reliability of power grid operation, and has important application value and promotion prospects.

[0089] Based on the same inventive concept, this application also provides a substation metering system status prediction device. Figure 6 A schematic diagram illustrating the structure of a substation metering system state prediction device according to an embodiment of this application is shown. Figure 6As shown, the device includes: a data acquisition module for acquiring historical data of the substation, including environmental state quantities and their corresponding metering quantities; a preprocessing module for preprocessing the historical data to obtain preprocessed data; a data clustering module for processing the preprocessed data using a closed-loop clustering algorithm to obtain data classifications under different scenarios, and dividing the preprocessed data after data classification into a training set and a validation set; and a combined model module for training a combined model using the training set and validating the trained combined model using the validation set; the combined model includes a random forest algorithm, a convolutional neural network algorithm, and an echo state network algorithm; the validated combined model is used to output corresponding prediction results based on the input environmental state quantities and their corresponding metering quantities.

[0090] In some optional embodiments of this application, obtaining historical data of a substation, including environmental state quantities and their corresponding metering quantities, includes: collecting historical data of the substation, wherein the historical data includes environmental state quantities and their corresponding metering quantities; distinguishing the environmental state quantities and their corresponding metering quantities by different scenarios, and combining the environmental state quantities and metering quantities with corresponding relationships into a state matrix form.

[0091] In some optional embodiments of this application, the historical data is preprocessed to obtain preprocessed data, including: constructing a device state matrix based on the historical data; determining the numerical range of the elements in the device state matrix and identifying values ​​outside the preset numerical range as outliers; deleting the outliers and then using Lagrange interpolation to generate new values ​​to fill the positions of the deleted outliers; standardizing the elements in the device state matrix through linear transformation; and using principal component analysis to reduce the dimensionality of the standardized device state matrix to obtain the preprocessed data.

[0092] In some optional embodiments of this application, the preprocessed data is processed by a closed-loop clustering algorithm to obtain data classifications under different scenarios, including: inputting the preprocessed data into a closed-loop clustering algorithm model to divide all scenarios into several categories; based on the measurement of scenarios within the same category after clustering as the prediction object, training convolutional neural network models with the goal of minimizing the root mean square error to obtain prediction models that correspond one-to-one with the clustered categories; processing the data of each scenario in all scenarios as follows: using the prediction model to make predictions to obtain multiple sets of prediction results equal to the number of prediction models; if one of the multiple sets of prediction results has the smallest root mean square error, then moving the scenario corresponding to that set of prediction results to the corresponding category; repeating the aforementioned steps of obtaining prediction models and data processing steps for each scenario until the termination condition is reached, the termination condition including the clustering result no longer changing or reaching the maximum number of iterations; and using the clustering result at the termination condition as the data classification under the different scenarios.

[0093] In some optional embodiments of this application, the apparatus further includes an imbalance processing module, which is used to: generate cluster labels based on the clustering results when the termination condition is met and add them to the preprocessed data; and perform imbalance processing on the preprocessed data after data classification under different scenarios.

[0094] In some optional embodiments of this application, the random forest algorithm in the combined model is configured as follows: the data in the training set is input into the random forest algorithm to obtain the corresponding output, and the input data and output data of the random forest algorithm are fused to obtain fused data; the convolutional neural network algorithm in the combined model is configured as follows: the feature extraction result is obtained based on the input fused data; the echo state network algorithm in the combined model is configured as follows: a state matrix is ​​generated based on the feature extraction result, and the state matrix is ​​used to obtain the final output result through the output weight matrix.

[0095] In some optional embodiments of this application, the input and output data of the random forest algorithm are fused to obtain fused data, including: concatenating the input and output data of the random forest algorithm to obtain fused data; the echo state network algorithm is trained through the following steps: initializing the weight matrix and hyperparameters in the echo state network algorithm; the data in the training set is processed by the convolutional neural network part to generate corresponding feature extraction results, the feature extraction results are input to the echo state network part to train the connection weights from the hidden layer to the output layer; the hidden layer obtains the final output result through the output weight matrix of the output layer.

[0096] In some optional embodiments of this application, the validated combined model is used to output a corresponding prediction result based on the input environmental state quantity and its corresponding measurement quantity, including: acquiring real-time data of the environmental state quantity and its corresponding measurement quantity; constructing the real-time data into a state matrix form; inputting the real-time data in the state matrix form into the validated combined model, and using the output of the validated combined model as the corresponding prediction result.

[0097] In some optional embodiments of this application, the device further includes a state determination module, which is used to: after outputting a prediction result based on the input environmental state quantity and its corresponding metering quantity, fit the prediction result output by the verified combined model into a dynamic prediction curve; calculate the residual between the dynamic prediction curve and the actual curve obtained by fitting real-time data within a set time interval, and obtain a residual dynamic change curve based on the residual; and determine the equipment state of the substation according to the fluctuation of the residual dynamic change curve.

[0098] The specific limitations of each functional module in the aforementioned substation metering system state prediction device can be found in the limitations of the substation metering system state prediction method described above, and will not be repeated here. Each module in the above system can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in the electronic device, or stored in the memory of the electronic device as software, so that the processor can call and execute the corresponding operations of each module. This also achieves the advantages of improving the intelligence level of substation operation and maintenance, and enhancing the safety and reliability of power grid operation.

[0099] In some embodiments of this application, an electronic device is also provided, comprising: at least one processor; and a memory connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, which executes the aforementioned substation metering system state prediction method. Its internal structure diagram can be shown as follows. Figure 7 As shown. Figure 7This diagram schematically illustrates the internal structure of an electronic device according to an embodiment of this application. The electronic device includes a processor A01, a network interface A02, a memory (not shown), and a database (not shown) connected via a system bus. The processor A01 provides computational and control capabilities. The memory includes internal memory A03 and a non-volatile storage medium A04. The non-volatile storage medium A04 stores an operating system B01, a computer program B02, and a database (not shown). The internal memory A03 provides an environment for the operation of the operating system B01 and the computer program B02 stored in the non-volatile storage medium A04. The network interface A02 is used for communication with external terminals via a network connection. When the computer program B02 is executed by the processor A01, it implements a substation metering system state prediction method.

[0100] Those skilled in the art will understand that Figure 7 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the electronic device to which the present application is applied. The specific electronic device may include more or fewer components than shown in the figure, or combine certain components, or have different component arrangements.

[0101] In one embodiment provided in this application, a machine-readable storage medium is provided, on which instructions are stored, which, when executed by a processor, cause the processor to be configured to perform the aforementioned substation metering system state prediction method.

[0102] In one embodiment provided in this application, a computer program product is provided, including a computer program that, when executed by a processor, implements the aforementioned substation metering system state prediction method.

[0103] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0104] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0105] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0106] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0107] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0108] Memory may include non-persistent memory in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, like read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0109] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can store information using any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic magnetic disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0110] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0111] The above are merely embodiments of this application and are not intended to limit the scope of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of the claims of this application.

Claims

1. A method for predicting the state of a substation metering system, characterized in that, The method includes: Acquire historical data of the substation, including environmental status parameters and their corresponding metering parameters; The historical data is preprocessed to obtain preprocessed data; The preprocessed data is processed using a closed-loop clustering algorithm to obtain data classifications for different scenarios, including: inputting the preprocessed data into a closed-loop clustering algorithm model to divide all scenarios into several categories; based on the metrics of scenarios within the same category after clustering as prediction objects, training convolutional neural network models with the objective of minimizing the root mean square error to obtain prediction models that correspond one-to-one with the clustered categories; processing the data for each scenario in all scenarios as follows: using the prediction model to make predictions to obtain multiple sets of prediction results equal to the number of prediction models; if one of the multiple sets of prediction results has the smallest root mean square error, then moving the scenario corresponding to that set of prediction results to the corresponding category; repeating the aforementioned steps of obtaining prediction models and data processing for each scenario until the termination condition is met, the termination condition including the clustering result no longer changing or reaching the maximum number of iterations; using the clustering result at the termination condition as the data classification for the different scenarios. The preprocessed data after classification is divided into training set and validation set; The training set is used to train the combined model, and the validation set is used to validate the trained combined model. The combined model includes a random forest algorithm, a convolutional neural network algorithm, and an echo state network algorithm. The validated combined model is used to output prediction results based on the input environmental state variables and their corresponding quantifiers.

2. The method according to claim 1, characterized in that, Acquire historical data of the substation, including environmental status parameters and their corresponding metering parameters, including: Collect historical data of the substation, including environmental status quantities and their corresponding metering quantities; The environmental state quantities and their corresponding measurement quantities are distinguished according to different scenarios, and the environmental state quantities and measurement quantities with corresponding relationships are combined into a state matrix form.

3. The method according to claim 1, characterized in that, The historical data is preprocessed to obtain preprocessed data, including: A device status matrix is ​​constructed based on the historical data; The elements in the device status matrix are judged for their numerical range, and values ​​that are not in the preset numerical range are identified as abnormal values. After deleting the outliers, new values ​​are generated using Lagrange interpolation to fill the positions of the deleted outliers; The elements in the device state matrix are standardized by linear transformation; Principal component analysis is used to reduce the dimensionality of the standardized equipment state matrix to obtain the preprocessed data.

4. The method according to claim 1, characterized in that, The method further includes: Cluster labels are generated based on the clustering results when the termination condition is met and added to the preprocessed data; and Imbalanced processing is applied to the preprocessed data after classification in different scenarios.

5. The method according to claim 1, characterized in that, The random forest algorithm in the combined model is configured such that: the data in the training set is input into the random forest algorithm to obtain the corresponding output, and the input data and output data of the random forest algorithm are fused to obtain fused data; The convolutional neural network algorithm in the combined model is configured to: obtain feature extraction results based on the input fused data; The echo state network algorithm in the combined model is configured to generate a state matrix based on the feature extraction results, and the state matrix is ​​used to obtain the final output result through the output weight matrix.

6. The method according to claim 5, characterized in that, The fused data is obtained by fusing the input and output data of the random forest algorithm, including: concatenating the input and output data of the random forest algorithm to obtain the fused data; The Echo State Network algorithm is trained through the following steps: initializing the weight matrix and hyperparameters in the Echo State Network algorithm; generating corresponding feature extraction results from the data in the training set through the convolutional neural network algorithm, and inputting the feature extraction results into the Echo State Network algorithm to train the connection weights from the hidden layer to the output layer; the hidden layer obtains the final output result through the output weight matrix of the output layer.

7. The method according to claim 1, characterized in that, The validated combined model was used to output prediction results based on the input environmental state variables and their corresponding metrics, including: Acquire real-time data of environmental state quantities and their corresponding metering quantities; The real-time data is constructed into a state matrix form; The real-time data in the form of the state matrix is ​​input into the validated combined model, and the output of the validated combined model is used as the corresponding prediction result.

8. The method according to claim 1, characterized in that, After outputting the corresponding prediction results based on the input environmental state quantities and their corresponding measurement quantities, the method further includes: The prediction results output by the validated combined model are fitted into a dynamic prediction curve; Within a set time interval, calculate the residual between the dynamic prediction curve and the actual curve obtained by fitting real-time data, and obtain the residual dynamic change curve based on the residual. The equipment status of the substation is determined based on the fluctuation of the residual dynamic change curve.

9. A state prediction device for a substation metering system, characterized in that, The device includes: The data acquisition module is used to acquire historical data of the substation, including environmental status quantities and their corresponding metering quantities. The preprocessing module is used to preprocess the historical data to obtain preprocessed data; The data clustering module is used to process the preprocessed data using a closed-loop clustering algorithm to obtain data classifications for different scenarios. This includes: inputting the preprocessed data into a closed-loop clustering algorithm model to divide all scenarios into several categories; training convolutional neural network models with the objective of minimizing the root mean square error (RMSE) based on the metrics of scenarios within the same clustered category as the prediction objects, obtaining prediction models that correspond one-to-one with the clustered categories; processing the data for each scenario in all scenarios as follows: using the prediction models to make predictions, obtaining multiple sets of prediction results equal to the number of prediction models; if one set of prediction results has the smallest RMSE, moving the scenario corresponding to that set of prediction results to the corresponding category; repeating the aforementioned steps of obtaining prediction models and data processing for each scenario until a termination condition is reached, including the clustering result no longer changing or reaching the maximum number of iterations; using the clustering result at the termination condition as the data classification for the different scenarios; dividing the preprocessed data after data classification into a training set and a validation set; and... The combined model module is used to train the combined model using the training set and to validate the trained combined model using the validation set; the combined model includes a random forest algorithm, a convolutional neural network algorithm, and an echo state network algorithm; the validated combined model is used to output corresponding prediction results based on the input environmental state variables and their corresponding quantifiers.

10. The apparatus according to claim 9, characterized in that, Acquire historical data of the substation, including environmental status parameters and their corresponding metering parameters, including: Collect historical data of the substation, including environmental status quantities and their corresponding metering quantities; The environmental state quantities and their corresponding measurement quantities are distinguished according to different scenarios, and the environmental state quantities and measurement quantities with corresponding relationships are combined into a state matrix form.

11. The apparatus according to claim 9, characterized in that, The historical data is preprocessed to obtain preprocessed data, including: A device status matrix is ​​constructed based on the historical data; The elements in the device status matrix are judged for their numerical range, and values ​​that are not in the preset numerical range are identified as abnormal values. After deleting the outliers, new values ​​are generated using Lagrange interpolation to fill the positions of the deleted outliers; The elements in the device state matrix are standardized by linear transformation; Principal component analysis is used to reduce the dimensionality of the standardized equipment state matrix to obtain the preprocessed data.

12. The apparatus according to claim 9, characterized in that, The device further includes an imbalance processing module, which is used for: Cluster labels are generated based on the clustering results when the termination condition is met and added to the preprocessed data; and Imbalanced processing is applied to the preprocessed data after classification in different scenarios.

13. The apparatus according to claim 9, characterized in that, The random forest algorithm in the combined model is configured such that: the data in the training set is input into the random forest algorithm to obtain the corresponding output, and the input data and output data of the random forest algorithm are fused to obtain fused data; The convolutional neural network algorithm in the combined model is configured to: obtain feature extraction results based on the input fused data; The echo state network algorithm in the combined model is configured to generate a state matrix based on the feature extraction results, and the state matrix is ​​used to obtain the final output result through the output weight matrix.

14. The apparatus according to claim 13, characterized in that, The fused data is obtained by fusing the input and output data of the random forest algorithm, including: concatenating the input and output data of the random forest algorithm to obtain the fused data; The Echo State Network algorithm is trained through the following steps: initializing the weight matrix and hyperparameters in the Echo State Network algorithm; generating corresponding feature extraction results from the data in the training set through the convolutional neural network algorithm, and inputting the feature extraction results into the Echo State Network algorithm to train the connection weights from the hidden layer to the output layer; the hidden layer obtains the final output result through the output weight matrix of the output layer.

15. The apparatus according to claim 9, characterized in that, The validated combined model was used to output prediction results based on the input environmental state variables and their corresponding metrics, including: Acquire real-time data of environmental state quantities and their corresponding metering quantities; The real-time data is constructed into a state matrix form; The real-time data in the form of the state matrix is ​​input into the validated combined model, and the output of the validated combined model is used as the corresponding prediction result.

16. The apparatus according to claim 9, characterized in that, The device further includes a state determination module, which is used to: after outputting a corresponding prediction result based on the input environmental state quantity and its corresponding measurement quantity, The prediction results output by the validated combined model are fitted into a dynamic prediction curve; Within a set time interval, calculate the residual between the dynamic prediction curve and the actual curve obtained by fitting real-time data, and obtain the residual dynamic change curve based on the residual. The equipment status of the substation is determined based on the fluctuation of the residual dynamic change curve.

17. An electronic device, characterized in that, include: At least one processor; A memory connected to the at least one processor; The memory stores instructions executable by the at least one processor, which executes the steps of the substation metering system state prediction method according to any one of claims 1 to 8 by executing the instructions stored in the memory.

18. A computer-readable storage medium having a computer program / instructions stored thereon, characterized in that, When the computer program / instructions are executed by the processor, they implement the steps of the substation metering system state prediction method according to any one of claims 1 to 8.

19. A computer program product comprising a computer program / instructions, characterized in that, When the computer program / instructions are executed by the processor, they implement the steps of the substation metering system state prediction method according to any one of claims 1 to 8.

Images