Online remote charging and discharging fault prediction method and system for communication storage battery pack

Abstract

The invention provides an online remote charging and discharging fault prediction method and system for a communication storage battery pack, and relates to the technical field of storage battery pack fault prediction, and the method comprises the steps: preprocessing time series data, extracting complexity features based on multi-scale entropy analysis, and determining physical features based on a physical model. Performing normalization dimensionless processing on the complexity features and the physical features, and determining a primary feature set; on the basis of the primary feature set, constructing a weighted undirected graph, expanding to generate a maximum weight connected sub-graph, and constructing a preferred feature set by taking primary features in the maximum weight connected sub-graph as preferred features; and inputting to-be-predicted storage battery pack data into the fault prediction network, extracting corresponding feature values on an input layer based on the optimal feature set, calculating fault type probability distribution through multiple hidden layers, and selecting a fault type corresponding to the highest probability as a fault prediction result.

Description

Technical Field

[0001] The present invention relates to the technical field of battery pack fault prediction, and in particular to an online remote charging and discharging fault prediction method and system for a communication battery pack. Background Art

[0002] Communication power is usually powered by AC mains electricity, and equipped with communication battery packs as backup power to ensure that the base station can continue to operate when the mains power is cut off. However, communication battery packs are easily affected by various factors during long-term use, resulting in performance degradation or even failure. Battery pack failure will not only affect the normal operation of the base station, but may also cause safety accidents and economic losses. Therefore, timely and accurate prediction of communication battery pack failure is of great significance to ensure the reliability and continuity of the communication network.

[0003] Traditional battery pack fault prediction methods mainly rely on manual inspections and regular maintenance, which have problems such as large workload, low efficiency, and poor accuracy. In recent years, with the development of the Internet of Things and big data technologies, remote online monitoring and data analysis have gradually been applied to the field of battery pack fault prediction. Existing fault prediction methods are mainly aimed at single batteries or small-scale battery packs, and lack consideration of large-scale distributed communication battery packs. Communication power sources are widely distributed, and the environmental conditions and load characteristics of battery packs in different base stations are different, resulting in different failure modes and degradation laws.

[0004] In summary, how to achieve online remote charging and discharging fault prediction for large-scale communication battery groups while taking into account the differences and correlations between different base stations is an urgent problem to be solved. The present invention can solve the problems in the prior art. Summary of the invention

[0005] The embodiment of the present invention provides a method and system for predicting online remote charging and discharging faults of a communication battery pack, which can solve the problems in the prior art.

[0006] According to a first aspect of the embodiments of the present invention,

[0007] A method for predicting online remote charging and discharging faults of a communication battery pack is provided, comprising:

[0008] Collecting time series data of the charging and discharging process of the communication battery pack, preprocessing the time series data, extracting complexity features based on multi-scale entropy analysis, determining physical features through a parameter identification algorithm based on a physical model corresponding to the battery pack, normalizing the complexity features and the physical features into a dimensionless process, and determining a primary feature set of the charging and discharging process of the battery pack;

[0009] Based on the primary feature set, a correlation coefficient matrix is ​​constructed, and the weighted edges of the undirected graph are determined with the correlation coefficients as weight values. A weighted undirected graph is constructed with the primary features as undirected graph nodes; based on the weighted undirected graph, a maximum weighted connected subgraph is generated by iteratively calculating the weighted edge connectivity sum of each undirected graph node, and the primary features corresponding to each subgraph node in the maximum weighted connected subgraph are used as preferred features to construct a preferred feature set;

[0010] The battery group data to be predicted is input into the fault prediction network. The corresponding feature values ​​are extracted based on the preferred feature set in the input layer. The feature representation is abstracted layer by layer through multiple hidden layers, and high-order features are extracted. The probability distribution of the fault type is calculated through the output layer activation function, and the fault type corresponding to the highest probability is selected as the fault prediction result.

[0011] In an optional embodiment,

[0012] Preprocess the time series data, extract the complexity features based on multi-scale entropy analysis, determine the physical features through parameter identification algorithm based on the physical model corresponding to the battery pack, normalize the complexity features and the physical features into dimensionless features, and determine that the primary feature set of the battery pack charging and discharging process includes:

[0013] Detect and remove outliers from time series data, and obtain standard time series data through smoothing filtering and normalization.

[0014] Performing a coarse-graining operation on the standard time series data to obtain scale subsequences at different scales, calculating sample entropy for each scale subsequence, determining a multi-scale entropy curve, extracting slope features, curvature features, entropy value features and area features from the multi-scale entropy curve, and constructing complexity features;

[0015] Based on the electrochemical characteristics and thermodynamic behavior of the battery pack, the corresponding physical model is constructed. The physical model parameters are evaluated and the physical characteristics are determined through the parameter identification algorithm based on the recursive least squares method.

[0016] The complexity characteristics and the physical characteristics are normalized and dimensionless, the numerical range and dimension differences are eliminated, and a primary feature set is determined.

[0017] In an optional embodiment,

[0018] Based on the primary feature set, a correlation coefficient matrix is ​​constructed, and the weighted edges of the undirected graph are determined with the correlation coefficient as the weight value. The weighted undirected graph is constructed with the primary features as the undirected graph nodes; based on the weighted undirected graph, the weighted edge connectivity sum of each undirected graph node is iteratively calculated to expand and generate a maximum weighted connected subgraph, and the primary features corresponding to each connected node in the maximum weighted connected subgraph are used as the preferred features. Constructing the preferred feature set includes:

[0019] Based on the primary feature set, the correlation coefficient is determined by calculating the covariance and variance between every two primary features. If the absolute value of the correlation coefficient is greater than a preset correlation coefficient threshold, the corresponding correlation coefficient is retained, otherwise it is removed, and a correlation coefficient matrix is ​​constructed based on the retained correlation coefficients;

[0020] Taking each primary feature in the primary feature set as an undirected graph node, traversing each correlation coefficient in the correlation coefficient matrix, taking the correlation coefficient as a weight value, determining an undirected graph weight edge between two corresponding undirected graph nodes, and constructing a weighted undirected graph;

[0021] Traverse each undirected graph node in the weighted undirected graph, calculate the undirected weight sum of all undirected graph weight edges connected to the undirected graph node, select the undirected graph node corresponding to the maximum value of the undirected weight sum as the first connected node, initialize the connected subgraph, and start iterative expansion of the connected subgraph:

[0022] Traversing each connected node of the connected subgraph, determining a corresponding undirected graph node from the weighted undirected graph, finding all undirected graph nodes that are directly connected to the undirected graph node and are not in the connected subgraph, determining a candidate node set, calculating a weighted edge connectivity sum in the weighted undirected graph for each candidate node in the candidate node set and all connected nodes in the connected subgraph, selecting a candidate node corresponding to the maximum value of the weighted edge connectivity sum as a new connected node, and adding it to the connected subgraph;

[0023] Until the candidate node set is empty, a maximum weight connected subgraph is generated;

[0024] The primary feature corresponding to each connected node in the maximum weight connected subgraph is taken as the preferred feature to construct a preferred feature set.

[0025] In an optional embodiment,

[0026] The construction process of the weighted undirected graph also includes:

[0027] Based on the correlation coefficient matrix and a preset correlation coefficient threshold, binarizing the correlation coefficient matrix to determine a binarized correlation matrix;

[0028] Based on the binary correlation matrix, the Euclidean distance between every two features is calculated, a feature distance matrix is ​​determined, the maximum value and the minimum value in the feature distance matrix are extracted, the feature distance ratio is determined, and a diversity measurement matrix is ​​generated;

[0029] Each primary feature in the primary feature set is taken as an undirected graph node, each correlation coefficient in the correlation coefficient matrix and each diversity metric in the diversity metric matrix are traversed, a weight value is calculated through the correlation coefficient and the diversity metric, an undirected graph weight edge between two corresponding undirected graph nodes is determined, and a weighted undirected graph is constructed.

[0030] In an optional embodiment,

[0031] The battery group data to be predicted is input into the fault prediction network. Based on the preferred feature set, the corresponding feature values ​​are extracted in the input layer. The feature representation is abstracted layer by layer through multiple hidden layers, and high-order features are extracted. The fault type probability distribution is calculated through the output layer activation function, and the fault type corresponding to the highest probability is selected as the fault prediction result, including:

[0032] The fault prediction network is constructed based on a deep belief network, and includes an input layer, at least one hidden layer and an output layer, wherein the number of nodes in the input layer is equal to the number of features in the preferred feature set, the hidden layer is composed of multiple layers of stacked probabilistic neural units, and the number of nodes in the output layer corresponds one-to-one to the fault type;

[0033] Based on the unsupervised learning method, primary training is performed layer by layer on each layer of the probabilistic neural unit of the fault prediction network. The probabilistic neural unit of the previous layer outputs the activation probability of the hidden layer, which enters the input layer of the probabilistic neural unit of the current layer. In the hidden layer of the probabilistic neural unit of the current layer, the weight and bias factor of the probabilistic neural unit of the current layer are adjusted by minimizing the energy function to generate the probability distribution corresponding to the input data of the probabilistic neural unit of the current layer.

[0034] When the primary training of all the probabilistic neural units is completed, the fault prediction network is secondary adjusted based on a supervised learning method, and the fault prediction network is initialized based on the weights and bias factors corresponding to each layer of the probabilistic neural units determined by the primary training;

[0035] Adding a classifier to the output layer, taking different fault types as labels, minimizing the classification loss function through an optimization algorithm, updating the weight and bias factor of each layer in the fault prediction network, and obtaining an optimal fault prediction network;

[0036] The battery group data to be predicted is input into the optimal fault prediction network, the fault type probability distribution is input, and the fault type with the highest probability is selected as the fault prediction result.

[0037] In an optional embodiment,

[0038] The previous layer of probabilistic neural units outputs the hidden layer activation probability and enters the input layer of the current layer of probabilistic neural units. In the hidden layer of the current layer of probabilistic neural units, the weights and bias factors of the current layer of probabilistic neural units are adjusted by minimizing the energy function to generate the probability distribution corresponding to the input data of the current layer of probabilistic neural units:

[0039] The energy function has the following formula:

[0040]

[0041] Where l represents the hidden layer number of the fault prediction network, v l represents the input layer state, h l represents the hidden layer state, E(v l ,h l ) represents the energy function, W l represents the connection weight matrix between the input layer and the hidden layer, a l represents the input layer bias vector, b l represents the hidden layer bias vector, U l ij represents the interaction weight between node i and node j in the input layer, v l i represents the input layer state of input layer node i, v l j represents the input layer state of input layer node j, V l km represents the interaction weight between node k and node m in the hidden layer, h l k represents the hidden layer state of hidden layer node k, h l m represents the hidden layer state of the hidden layer node m, and λ represents the regularization term balance parameter.

[0042] In an optional embodiment,

[0043] Based on the fault prediction network, constructing a distributed prediction network includes:

[0044] A distributed prediction network is constructed with a communication battery pack and corresponding edge computing nodes as the branch network and a central server as the core network;

[0045] The central network is constructed based on the fault prediction network, and the central network is distributed to the branch networks; data collection and training are performed independently on each branch network, and the model parameters in each branch network after training are uploaded to the central network. The central network aggregates the model parameters to obtain updated model parameters, and distributes them to the branch networks;

[0046] Multiple rounds of iterations are performed in sequence until a preset iteration upper limit is reached, and a final central network and a corresponding final branch network are determined, and the final branch network performs fault prediction.

[0047] According to a second aspect of the embodiments of the present invention,

[0048] Provided is a communication battery pack online remote charging and discharging fault prediction system, comprising:

[0049] The first unit is used to collect time series data of the charging and discharging process of the communication battery pack, pre-process the time series data, extract complexity features based on multi-scale entropy analysis, determine physical features through parameter identification algorithm based on the physical model corresponding to the battery pack, normalize the complexity features and the physical features into dimensionless processing, and determine the primary feature set of the charging and discharging process of the battery pack;

[0050] The second unit is used to construct a correlation coefficient matrix based on the primary feature set, determine the weight edge of the undirected graph with the correlation coefficient as the weight value, and construct a weighted undirected graph with the primary features as the undirected graph nodes; based on the weighted undirected graph, iteratively calculate the weight edge connectivity sum of each undirected graph node, expand and generate a maximum weight connected subgraph, and use the primary features corresponding to each subgraph node in the maximum weight connected subgraph as the preferred features to construct a preferred feature set;

[0051] The third unit is used to input the battery group data to be predicted into the fault prediction network, extract the corresponding feature values ​​based on the preferred feature set in the input layer, abstract the feature representation layer by layer through multiple hidden layers, extract high-order features, and calculate the probability distribution of fault types through the output layer activation function, and select the fault type corresponding to the highest probability as the fault prediction result.

[0052] According to a third aspect of the embodiments of the present invention,

[0053] An electronic device is provided, comprising:

[0054] processor;

[0055] a memory for storing processor-executable instructions;

[0056] The processor is configured to call the instructions stored in the memory to execute the aforementioned method.

[0057] A fourth aspect of the embodiments of the present invention is:

[0058] A computer-readable storage medium is provided, on which computer program instructions are stored. When the computer program instructions are executed by a processor, the aforementioned method is implemented.

[0059] In the embodiment of the present invention, a multi-scale entropy analysis method is used to extract complex features from the pre-processed time series data, capture the nonlinear and non-stationary characteristics of the signal, and help to deeply understand the dynamic changes of the battery pack charging and discharging process; according to the physical model corresponding to the battery pack, it can provide an in-depth understanding of the battery state and performance; by constructing a correlation coefficient matrix and using a weighted undirected graph, the correlation between primary features can be systematically analyzed, the strongest correlation between features can be identified, and the most useful features for fault prediction can be effectively screened out, which can ensure that the selected feature set is not only representative, but also can retain the information content of the original data to the greatest extent; by selecting the maximum weight connected subgraph, redundant features can be effectively removed, and the correlation coefficient matrix helps to identify highly correlated features and avoid duplication. The influence of information improves the generalization ability and prediction accuracy of the model; building a branch network allows parallel processing of data from each communication battery group, which means that the system can process multiple data streams at the same time, improving the overall efficiency and processing power of the system; a two-way communication channel is established between the central network and the branch network, so that the model parameters can be transmitted and updated between the central network and the branch network. The collaborative training method allows the central network to summarize information from multiple branch networks to obtain a more comprehensive data perspective and improve the accuracy and generalization ability of the model; after completing the independent training of the branch network, the central network will collect and aggregate the model parameters of all branch networks to generate updated global model parameters. The parameter aggregation mechanism can effectively integrate the training results of each branch network and improve the overall performance of the model. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] Figure 1 It is a flow chart of a method for predicting online remote charging and discharging faults of a communication battery pack according to an embodiment of the present invention;

[0061] Figure 2 It is a structural schematic diagram of the online remote charging and discharging fault prediction system of the communication battery pack according to the embodiment of the present invention. DETAILED DESCRIPTION

[0062] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0063] The technical solution of the present invention is described in detail with specific embodiments below. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described in detail in some embodiments.

[0064] Figure 1 FIG. 1 is a flow chart of a method for predicting online remote charging and discharging faults of a communication battery pack according to an embodiment of the present invention. Figure 1 As shown, the method includes:

[0065] S101. Collect time series data of the charging and discharging process of the communication battery pack, pre-process the time series data, extract complexity features based on multi-scale entropy analysis, determine physical features through parameter identification algorithm based on the physical model corresponding to the battery pack, normalize the complexity features and the physical features into dimensionless processing, and determine the primary feature set of the charging and discharging process of the battery pack;

[0066] The multiscale entropy analysis specifically refers to a method for describing the complexity of time series data, which calculates the entropy value of the data at different time scales. Multiscale entropy analysis can provide complexity information of data at different time scales by calculating entropy values ​​at different time scales;

[0067] The parameter identification algorithm specifically refers to determining the parameters in the system model by calculation based on the dynamic model of the system and the measured data. In the present invention, the parameter identification algorithm can be used to determine the electrochemical parameters of the battery, such as resistance, capacitance, etc.;

[0068] The normalized dimensionless processing specifically refers to dimensionlessly processing the data and distributing the data within a certain range, so that the dimensions of different features can be consistent, which is convenient for subsequent data analysis and processing.

[0069] In this embodiment, a multiscale entropy analysis method is used to extract complexity features from the preprocessed time series data. Multiscale entropy is a method for measuring the complexity of time series, which can capture the nonlinear and non-stationary characteristics of signals, and is helpful for in-depth understanding of the dynamic changes in the charging and discharging process of the battery pack; according to the physical model corresponding to the battery pack, the physical characteristics are determined by a parameter identification algorithm. The physical characteristics may involve the electrochemical parameters, internal resistance, voltage curve characteristics, etc. of the battery, which can provide an in-depth understanding of the battery state and performance; the extracted complexity features and physical features are normalized so that they have the same dimension and range, which is convenient for subsequent feature selection and model building; the normalized complexity features and physical features are combined to form a primary feature set of the charging and discharging process of the battery pack. The primary feature set will reflect the key characteristics of the battery pack in the charging and discharging process, and provide an important basis for subsequent fault prediction, state assessment and other tasks.

[0070] In an optional embodiment, the time series data is preprocessed, the complexity features are extracted based on multi-scale entropy analysis, the physical features are determined by a parameter identification algorithm based on a physical model corresponding to the battery pack, the complexity features and the physical features are normalized dimensionlessly, and the primary feature set of the battery pack charge and discharge process is determined to include:

[0071] Detect and remove outliers from time series data, and obtain standard time series data through smoothing filtering and normalization.

[0072] Performing a coarse-graining operation on the standard time series data to obtain scale subsequences at different scales, calculating sample entropy for each scale subsequence, determining a multi-scale entropy curve, extracting slope features, curvature features, entropy value features and area features from the multi-scale entropy curve, and constructing complexity features;

[0073] Based on the electrochemical characteristics and thermodynamic behavior of the battery pack, the corresponding physical model is constructed. The physical model parameters are evaluated and the physical characteristics are determined through the parameter identification algorithm based on the recursive least squares method.

[0074] The complexity characteristics and the physical characteristics are normalized and dimensionless, the numerical range and dimension differences are eliminated, and a primary feature set is determined.

[0075] Collect the original time series data and conduct preliminary checks, using the 3σ principle statistical method to identify outliers in the data, where the 3σ principle refers to the range of data points exceeding the mean plus or minus three times the standard deviation; process the detected outliers, either by directly deleting them or replacing them with reasonable alternative values, preferably using the mean or median of the previous and next data points;

[0076] Select appropriate filtering methods, such as moving average filtering, median filtering, and Kalman filtering, to remove high-frequency noise and short-term fluctuations, and smooth the time series data; normalize the smoothed time series data, scale the data to a uniform range, preferably [-1, 1], and eliminate the dimensional differences between different time series. The normalization method may include maximum and minimum value normalization or zero mean unit variance normalization; use the data after outlier removal, smoothing filtering, and normalization as standard time series data;

[0077] Determine different scale factors to perform coarse-graining operations on standard time series data. Use the set scale factors to divide the standard time series data into subsequences at different scales, and map each subsequence to a scale point by taking the average, maximum or minimum value. For the scaled subsequences at each scale, calculate the sample entropy to determine the complexity and regularity of the time series at that scale. Select sample entropy values ​​under different scale factors, draw multi-scale entropy curves, and determine the changing trend of the complexity of the time series at different scales.

[0078] Extract complexity features from the multiscale entropy curve, calculate the slope of the multiscale entropy curve in different scale intervals, and obtain the rate at which the entropy value changes with scale; calculate the curvature of the multiscale entropy curve at different scale points to obtain the degree of nonlinearity of the entropy value changing with scale; extract the maximum entropy value, minimum entropy value and average entropy value of the multiscale entropy curve to obtain the overall complexity level of the time series; calculate the area enclosed by the multiscale entropy curve to obtain the cumulative effect of complexity at different scales; combine the extracted slope features, curvature features, entropy value features and area features into a complexity feature vector;

[0079] According to the specific characteristics of the battery pack, select an equivalent circuit model, preferably an RC circuit model, set the structure and equations of the model based on the model type, and define various parameters in the model, such as resistance, capacitance, thermal conductivity, etc.; collect experimental data or actual operation data, including key variables such as current, voltage and temperature;

[0080] Initialize the model parameter values, use the recursive least squares algorithm to continuously update the parameters to minimize the error between the predicted value and the actual observed value, and determine whether the parameter update converges, that is, whether the parameter change is within the preset tolerance range; extract key parameters from the identified physical model, and calculate the physical characteristics based on the key parameters;

[0081] The complexity features and physical features are standardized, and then the numerical ranges of different features are scaled to the same interval, preferably [-1, 1], through normalization, to eliminate the dimensional differences between different features and ensure that different features are in the same numerical range. This makes it easier to perform feature fusion and comparative analysis, and the normalized complexity features and physical features are fused to construct a primary feature set.

[0082] In this embodiment, by coarse-graining different scale factors, the complexity and regularity of the time series at different scales are obtained, the dynamic characteristics of the time series are fully captured, and it is helpful to extract the complexity characteristics more accurately; the slope, curvature, maximum entropy value, minimum entropy value, average entropy value and the area enclosed by the curve are extracted from the multi-scale entropy curve to fully reflect the complexity and change law of the time series and improve the description ability and discrimination of the feature vector; by selecting an appropriate equivalent circuit model and using the recursive least squares algorithm to perform parameter identification on the data, the key physical parameters of the battery pack are extracted to accurately reflect the physical characteristics and operating status of the battery pack; the complexity characteristics and physical characteristics are standardized and normalized to ensure that the numerical ranges of different characteristics are consistent, which is convenient for feature fusion and comparative analysis. The fused primary feature set contains the dynamic complexity and physical characteristics of the battery pack, providing rich and effective feature data for subsequent fault prediction and status assessment.

[0083] S102. Based on the primary feature set, a correlation coefficient matrix is ​​constructed, and the weighted edges of the undirected graph are determined with the correlation coefficients as weight values. A weighted undirected graph is constructed with the primary features as undirected graph nodes; based on the weighted undirected graph, the weighted edge connectivity sum of each undirected graph node is iteratively calculated to expand and generate a maximum weighted connected subgraph, and the primary features corresponding to each subgraph node in the maximum weighted connected subgraph are used as preferred features to construct a preferred feature set;

[0084] The correlation coefficient matrix specifically refers to a symmetric matrix used to represent the linear correlation degree between the features in the primary feature set, and the elements in the matrix represent the correlation coefficients between the features.

[0085] The weighted undirected graph specifically refers to a graph in which nodes represent primary features, each primary feature serves as a node in the graph, and edges represent the relationship between features. If the absolute value of the correlation coefficient between two features is greater than a preset threshold, an edge is drawn between the two nodes, and the edge has a weight, and the weight of the edge is the correlation coefficient between the features.

[0086] In this embodiment, by constructing a correlation coefficient matrix and using a weighted undirected graph, the correlation between primary features can be systematically analyzed. By iteratively calculating the weighted edge connectivity, the strongest association between features can be identified, and the most useful features for fault prediction can be effectively screened out, ensuring that the selected feature set is not only representative, but also able to retain the information content of the original data to the greatest extent; by selecting the maximum weight connected subgraph, redundant features can be effectively removed, and the correlation coefficient matrix helps identify highly correlated features, avoids the influence of duplicate information, and improves the generalization ability and prediction accuracy of the model. Reducing redundant features helps simplify the model and reduce computational complexity and storage requirements.

[0087] In an optional embodiment, based on the primary feature set, a correlation coefficient matrix is ​​constructed, the correlation coefficient is used as the weight value, the undirected graph weight edge is determined, and the primary features are used as undirected graph nodes to construct a weighted undirected graph; based on the weighted undirected graph, the weighted edge connectivity of each undirected graph node is iteratively calculated to expand and generate a maximum weight connected subgraph, and the primary feature corresponding to each connected node in the maximum weight connected subgraph is used as the preferred feature, and constructing the preferred feature set includes:

[0088] Based on the primary feature set, the correlation coefficient is determined by calculating the covariance and variance between every two primary features. If the absolute value of the correlation coefficient is greater than a preset correlation coefficient threshold, the corresponding correlation coefficient is retained, otherwise it is removed, and a correlation coefficient matrix is ​​constructed based on the retained correlation coefficients;

[0089] Taking each primary feature in the primary feature set as an undirected graph node, traversing each correlation coefficient in the correlation coefficient matrix, taking the correlation coefficient as a weight value, determining an undirected graph weight edge between two corresponding undirected graph nodes, and constructing a weighted undirected graph;

[0090] Traverse each undirected graph node in the weighted undirected graph, calculate the undirected weight sum of all undirected graph weight edges connected to the undirected graph node, select the undirected graph node corresponding to the maximum value of the undirected weight sum as the first connected node, initialize the connected subgraph, and start iterative expansion of the connected subgraph:

[0091] Traversing each connected node of the connected subgraph, determining a corresponding undirected graph node from the weighted undirected graph, finding all undirected graph nodes that are directly connected to the undirected graph node and are not in the connected subgraph, determining a candidate node set, calculating a weighted edge connectivity sum in the weighted undirected graph for each candidate node in the candidate node set and all connected nodes in the connected subgraph, selecting a candidate node corresponding to the maximum value of the weighted edge connectivity sum as a new connected node, and adding it to the connected subgraph;

[0092] Until the candidate node set is empty, a maximum weight connected subgraph is generated;

[0093] The primary feature corresponding to each connected node in the maximum weight connected subgraph is taken as the preferred feature to construct a preferred feature set.

[0094] Extract all primary features from the normalized primary feature set obtained in the previous stage, calculate the covariance between every two primary features, evaluate their linear relationship, and calculate the variance of each primary feature to indicate the degree of discreteness of the feature data;

[0095] Use the covariance and variance between the primary features to calculate the correlation coefficient and quantify the degree of linear correlation between the features; define a correlation coefficient threshold to screen significantly correlated feature pairs. If the absolute value of the correlation coefficient is greater than the preset correlation coefficient threshold, the corresponding correlation coefficient is retained; otherwise, the corresponding correlation coefficient is eliminated; create a matrix in which rows and columns represent each primary feature in the primary feature set, fill the retained correlation coefficients into the corresponding positions of the matrix, and construct a correlation coefficient matrix.

[0096] Each primary feature in the primary feature set is regarded as a node in the undirected graph, each correlation coefficient in the correlation coefficient matrix is ​​traversed, and the correlation coefficient is used as the weight value of the edge between two nodes in the undirected graph. If the correlation coefficient exists and is retained, a weighted edge is established between the corresponding undirected graph nodes, and all undirected graph nodes and undirected weighted edges are combined to construct a complete weighted undirected graph.

[0097] Traverse each undirected graph node in the weighted undirected graph, calculate the sum of all undirected graph weight edges connected to the undirected graph node, that is, the undirected weight sum, select the undirected graph node with the largest undirected weight sum as the first connected node, add the first connected node to the connected subgraph, and start iterative expansion:

[0098] Traverse each connected node in the current connected subgraph, find all undirected graph nodes that are directly connected to each connected node and are not in the connected subgraph from the weighted undirected graph, form a candidate node set, for each candidate node in the candidate node set, calculate the weighted edge connectivity sum between the candidate node and all connected nodes in the connected subgraph, select the candidate node with the largest weighted edge connectivity sum, and use the selected candidate node as a new connected node and add it to the connected subgraph;

[0099] Repeat the above steps until the candidate node set is empty.

[0100] The primary features corresponding to each connected node are extracted from the maximum weight connected subgraph, and the extracted primary features are combined to construct the final optimal feature set.

[0101] Alternatively, assume that there is a feature set F containing 5 features, and the correlation coefficient matrix R is as follows:

[0102]

[0103] And set the threshold r 0 is 0.5;

[0104] Initialize an empty undirected graph G with 5 nodes corresponding to the 5 features in the feature set F. Traverse each element in the correlation coefficient matrix R and add edges that meet the threshold conditions, including:

[0105] r 12 =0.9,>r 0 , add edge (1, 2) to G with weight 0.9;

[0106] r 15 =0.5, =r 0 , add edge (1, 5) to G with weight 0.5;

[0107] r 24 =0.7,>r 0 , add edge (2, 4) to G with weight 0.7;

[0108] r 25 =0.6,>r 0 , add edge (2, 5) to G with weight 0.6;

[0109] r 34 =0.8,>r 0 , add edge (3, 4) to G with weight 0.8;

[0110] r 45 =0.5, =r 0 , add edge (4, 5) to G with weight 0.5.

[0111] Determine the weighted undirected graph G;

[0112] Initialize an empty connected subgraph S and an empty candidate feature set C, calculate the weight and edge sum of each node, and select the node with the largest weight and the largest edge sum as the initial node:

[0113] Node 1, 0.9+0.5=1.4;

[0114] Node 2, 0.9+0.7+0.6=2.2 (maximum);

[0115] Node 3, 0.8 = 0.8;

[0116] Node 4, 0.7+0.8+0.5=2.0;

[0117] Node 5, 0.5+0.6+0.5=1.6;

[0118] Select node 2 and initialize S={2}, C={2}.

[0119] Initialize an empty candidate node set N, traverse the nodes in the current candidate feature set C, find the nodes that are directly connected to it and are not in S, and calculate the weight and edge sum:

[0120] Currently, there is only node 2 in C. The nodes directly connected to node 2 and not in S are 1, 4, and 5. Calculate them respectively:

[0121] Node 1, r 21 =0.9; Node 4, r 24 =0.7; node 5, r 25 =0.6;

[0122] Select the node with the largest weight, that is, select node 1, add S and C, and get S = {2, 1}, C = {2, 1};

[0123] The nodes directly connected to node 2 and not in S are 4 and 5, and the node directly connected to node 1 and not in S is 5. Update N = {4, 5};

[0124] Iterate again and calculate: Node 4, r 24 =0.7; node 5, r 15 +r 25 =0.5+0.6=1.1(maximum);

[0125] Select node 5, add S and C, now S = {2, 1, 5}, C = {2, 1, 5};

[0126] The node that is directly connected to nodes 2 and 5 and is not in S is 4, update N = {4};

[0127] Iterate again and calculate: Node 4, r 24 +r 45 =0.7+0.5=1.2;

[0128] Select node 4, add S and C, now S = {2, 1, 5, 4}, C = {2, 1, 5, 4};

[0129] There is no node directly connected to nodes 2, 5, and 4 that is not in S, so update N to be empty;

[0130] When the candidate node set N is empty, the iteration stops and the final maximum weight connected subgraph S = {2, 1, 5, 4} is obtained.

[0131] In this embodiment, by calculating the correlation coefficient between primary features, quantifying the degree of linear correlation between features, and using the correlation coefficient threshold to screen significantly correlated feature pairs, features that have a significant impact on the model can be efficiently selected, which helps to reduce the number of model input features and improve the model training speed and efficiency; by constructing a weighted undirected graph and a maximum weight connected subgraph, features that have a strong correlation with the target variable are selected, which can improve the prediction performance of the model, and the features in the preferred feature set can better reflect the essential characteristics of the data and enhance the generalization ability and prediction effect of the model; by gradually selecting features with high correlation and constructing a maximum weight connected subgraph, the features in the preferred feature set are made more explanatory, and the contribution of each feature to the final prediction result can be explained by its position and connectivity in the weighted undirected graph, which helps to better understand the decision-making process of the model.

[0132] In an optional embodiment, the process of constructing the weighted undirected graph further includes:

[0133] Based on the correlation coefficient matrix and a preset correlation coefficient threshold, binarizing the correlation coefficient matrix to determine a binarized correlation matrix;

[0134] Based on the binary correlation matrix, the Euclidean distance between every two features is calculated, a feature distance matrix is ​​determined, the maximum value and the minimum value in the feature distance matrix are extracted, the feature distance ratio is determined, and a diversity measurement matrix is ​​generated;

[0135] Each primary feature in the primary feature set is taken as an undirected graph node, each correlation coefficient in the correlation coefficient matrix and each diversity metric in the diversity metric matrix are traversed, a weight value is calculated through the correlation coefficient and the diversity metric, an undirected graph weight edge between two corresponding undirected graph nodes is determined, and a weighted undirected graph is constructed.

[0136] Extract all primary features from the primary feature set obtained in the previous step, and calculate the correlation coefficient between each two primary features to quantify the linear correlation between the features;

[0137] Define a correlation coefficient threshold to screen out significantly correlated feature pairs, traverse each correlation coefficient in the correlation coefficient matrix, and if the absolute value of the correlation coefficient is greater than the preset correlation coefficient threshold, set it to 1, otherwise set it to 0, and form a binary correlation matrix with the processed results;

[0138] Create a matrix with rows and columns representing each primary feature in the primary feature set. For every two primary features, calculate the Euclidean distance based on their binary correlation values, and fill the calculated Euclidean distance into the corresponding position of the feature distance matrix; extract the maximum Euclidean distance value and the minimum Euclidean distance value from the feature distance matrix, and for each value in the feature distance matrix, calculate the proportion of the value between the maximum value and the minimum value.

[0139] Create a matrix with the same structure as the feature distance matrix, fill the feature distance ratio results into the diversity measurement matrix, and complete the construction of the diversity measurement matrix;

[0140] Each primary feature in the primary feature set is taken as a node in the undirected graph, each correlation coefficient in the correlation coefficient matrix is ​​traversed, each diversity metric in the diversity metric matrix is ​​traversed, and the weight value of the edge between two undirected graph nodes is calculated based on the correlation coefficient and the diversity metric. The calculation of the weight value can consider the combination of the absolute value of the correlation coefficient and the diversity metric.

[0141] For each pair of primary features, if their correlation coefficient and diversity measure exist, a weighted edge is established between the corresponding undirected graph nodes, and all undirected graph nodes and undirected weighted edges are combined to construct a complete weighted undirected graph.

[0142] In this embodiment, by calculating the correlation coefficient between features and performing binarization processing, significantly correlated feature pairs can be effectively screened out, the accuracy of feature selection can be improved, it is ensured that the final selected features have a strong correlation with the target variable, and the prediction performance of the model is improved; the binary correlation matrix can eliminate feature pairs that are not significantly correlated, reduce feature redundancy, reduce the problem of multicollinearity, ensure the stability and reliability of the model, and avoid overfitting caused by redundant features; by calculating the Euclidean distance between features and constructing a diversity measurement matrix, the differences between features can be quantified, ensuring that the selected features have higher diversity, capturing more potential patterns of data, and improving the generalization ability of the model.

[0143] S103. Input the battery pack data to be predicted into the fault prediction network, extract the corresponding feature values ​​based on the preferred feature set in the input layer, abstract the feature representation layer by layer through multiple hidden layers, extract high-order features, and calculate the probability distribution of fault types through the output layer activation function, and select the fault type corresponding to the highest probability as the fault prediction result.

[0144] The multi-layer hidden layer specifically refers to a neural network structure located between the input layer and the output layer, which consists of multiple hidden layers, each of which contains a number of neurons connected by weights and biases. The main function of the multi-layer hidden layer is to gradually form a higher-order and more abstract feature representation by abstracting and extracting the features of the input data layer by layer;

[0145] In this embodiment, feature representations are abstracted layer by layer through multiple hidden layers to extract deeper features. High-order features can better capture complex patterns and potential relationships in battery pack data, thereby improving the accuracy of fault prediction. The probability distribution of fault types is calculated through the activation function of the output layer, which can accurately measure the possibility of each fault type, and select the fault type corresponding to the highest probability as the prediction result to ensure the accuracy and reliability of the prediction. The high-order feature representations extracted by multiple hidden layers can comprehensively reflect the status and potential fault information of the battery pack, thereby improving the robustness and reliability of the system under complex and changeable working conditions.

[0146] In an optional embodiment, the battery pack data to be predicted is input into the fault prediction network, and the corresponding feature values ​​are extracted based on the preferred feature set at the input layer. The feature representation is abstracted layer by layer through multiple hidden layers to extract high-order features, and the probability distribution of the fault type is calculated through the output layer activation function, and the fault type corresponding to the highest probability is selected as the fault prediction result, which includes:

[0147] The fault prediction network is constructed based on a deep belief network, and includes an input layer, at least one hidden layer and an output layer, wherein the number of nodes in the input layer is equal to the number of features in the preferred feature set, the hidden layer is composed of multiple layers of stacked probabilistic neural units, and the number of nodes in the output layer corresponds one-to-one to the fault type;

[0148] Based on the unsupervised learning method, primary training is performed layer by layer on each layer of the probabilistic neural unit of the fault prediction network. The probabilistic neural unit of the previous layer outputs the activation probability of the hidden layer, which enters the input layer of the probabilistic neural unit of the current layer. In the hidden layer of the probabilistic neural unit of the current layer, the weight and bias factor of the probabilistic neural unit of the current layer are adjusted by minimizing the energy function to generate the probability distribution corresponding to the input data of the probabilistic neural unit of the current layer.

[0149] When the primary training of all the probabilistic neural units is completed, the fault prediction network is secondary adjusted based on a supervised learning method, and the fault prediction network is initialized based on the weights and bias factors corresponding to each layer of the probabilistic neural units determined by the primary training;

[0150] Adding a classifier to the output layer, taking different fault types as labels, minimizing the classification loss function through an optimization algorithm, updating the weight and bias factor of each layer in the fault prediction network, and obtaining an optimal fault prediction network;

[0151] The battery group data to be predicted is input into the optimal fault prediction network, the fault type probability distribution is input, and the fault type with the highest probability is selected as the fault prediction result.

[0152] The energy function specifically refers to a scalar value used to describe the state of the network, reflecting the quality of the current network state, which is particularly important in probabilistic neural units. The energy function helps understand and optimize the weights and biases of the network by introducing the concept of energy value to minimize energy and make the network better fit the data distribution; the energy function assigns an energy value to each state of the network, and a lower energy value corresponds to a better state, that is, the network is more consistent with the probability distribution of the input data; the energy function is used to define the joint probability distribution of probabilistic neural units, and the probability of a state in the network is inversely proportional to its energy. The lower the energy, the higher the probability of the state; by minimizing the energy function, the weights and biases of the network are adjusted so that the network can learn the distribution of the input data more accurately;

[0153] The number of features in the preferred feature set is used as the number of nodes in the input layer, and the hidden layer is constructed using multi-layer stacked probabilistic neural units. The number of units in each layer can be set according to requirements. The number of nodes in the output layer is set to be consistent with the number of fault types, and each node corresponds to one fault type.

[0154] Starting from the input layer, each layer of probabilistic neural units is trained layer by layer, and the output of the probabilistic neural units in the previous layer is used as the hidden layer activation probability and input into the input layer of the probabilistic neural units in the current layer. In the hidden layer of the probabilistic neural units in the current layer, the weights and bias factors are adjusted by minimizing the energy function. The energy function measures the difference between the probability distribution of the current layer and the input data, and generates the probability distribution corresponding to the input data of the probabilistic neural units in the current layer. Repeat the training until the primary training of all hidden layers is completed.

[0155] Based on the weights and bias factors of each layer of probabilistic neural units obtained through unsupervised primary training, the entire fault prediction network is initialized, a classifier is added to the output layer, different fault types are used as labels, and a classification loss function is defined to measure the difference between the prediction result and the true label. The loss function preferably includes a cross entropy loss function;

[0156] Select an optimization algorithm, preferably a gradient descent method, to minimize the classification loss function, update the weights and bias factors of each layer in the fault prediction network through back propagation, and iteratively perform the optimization operation until the classification loss function converges or reaches a preset number of iterations, thereby obtaining a trained optimal fault prediction network;

[0157] The battery pack data to be predicted is input into the trained optimal fault prediction network, and the data is forward propagated through the input layer, hidden layer, and output layer. The probability distribution corresponding to each fault type is obtained in the output layer, and the fault type with the highest probability value is selected as the final fault prediction result.

[0158] In this embodiment, the use of the preferred feature set and the multi-layer stacking structure enables the model to more effectively extract and represent the key features of the input data, thereby improving the prediction accuracy; the layer-by-layer unsupervised primary training can learn the essential characteristics of the data with less reliance on labels, enhance the robustness of the model, and reduce the prediction errors caused by data noise and anomalies; through phased training and regularization, the risk of overfitting is reduced, the generalization ability of the model is improved, and its performance is stable under different environments and data conditions; combined with classifiers and optimization algorithms, the fault prediction network can realize automated fault diagnosis, reduce manual intervention, and improve prediction efficiency and accuracy.

[0159] In an optional embodiment, the previous layer of probability neural units outputs the hidden layer activation probability, which enters the input layer of the current layer of probability neural units. In the hidden layer of the current layer of probability neural units, the weights and bias factors of the current layer of probability neural units are adjusted by minimizing the energy function, and the probability distribution corresponding to the input data of the current layer of probability neural units is generated, including:

[0160] The energy function has the following formula:

[0161]

[0162] Where l represents the hidden layer number of the fault prediction network, v l represents the input layer state, h l represents the hidden layer state, E(v l ,h l ) represents the energy function, W l represents the connection weight matrix between the input layer and the hidden layer, a l represents the input layer bias vector, b l represents the hidden layer bias vector, U l ij represents the interaction weight between node i and node j in the input layer, v l i represents the input layer state of input layer node i, v l j represents the input layer state of input layer node j, V l km represents the interaction weight between node k and node m in the hidden layer, h l k represents the hidden layer state of hidden layer node k, h lm represents the hidden layer state of the hidden layer node m, and λ represents the regularization term balance parameter.

[0163] The formula of the energy function consists of three parts: a linear term, a quadratic term and a regularization term. The linear term is used to calculate the direct contribution of the input layer and the hidden layer, as well as the respective biases to the energy, and describes the direct impact of the input and hidden layer states and their interactions; the quadratic term is used to calculate the quadratic interaction between the input layer and the nodes in the hidden layer, and captures the more complex relationships between the nodes in the layer; the regularization term is used to control the complexity of the model and prevent overfitting. By introducing the L2 norm of the weights and biases, it is ensured that the parameters are not too large, so that the model has better generalization ability;

[0164] According to the formula, the energy function combines the linear terms and quadratic terms between the input layer and the hidden layer. By adjusting the weights and biases, it can abstract the feature representation layer by layer and extract the high-order features of the input data, which helps the network learn the complex patterns and features in the data; by introducing the nonlinear activation function and quadratic terms in the hidden layer, the energy function can model the nonlinear relationship in the data so that the network can better adapt to the complex data distribution and characteristics; the regularization term in the energy function helps to control the complexity of the model, avoid overfitting the training data during the training process, and improve the generalization ability of the network; the energy function adjusts the weights and biases through the minimization process to minimize the energy of the model and generate a probability distribution corresponding to the input data, which is achieved through optimization algorithms such as gradient descent and can effectively learn the representation of the data.

[0165] In an optional embodiment, based on the fault prediction network, constructing a distributed prediction network includes:

[0166] A distributed prediction network is constructed with a communication battery pack and corresponding edge computing nodes as the branch network and a central server as the core network;

[0167] The central network is constructed based on the fault prediction network, and the central network is distributed to the branch networks; data collection and training are performed independently on each branch network, and the model parameters in each branch network after training are uploaded to the central network. The central network aggregates the model parameters to obtain updated model parameters, and distributes them to the branch networks;

[0168] Multiple rounds of iterations are performed in sequence until a preset iteration upper limit is reached, and a final central network and a corresponding final branch network are determined, and the final branch network performs fault prediction.

[0169] Define branch networks, regard each communication battery group and the corresponding edge computing node as a branch network, define a central network, regard the central server as the central network, which is used to coordinate and aggregate the model parameters of the branch networks, establish a communication mechanism, and establish a two-way communication channel between the central network and the branch networks for the transmission and update of model parameters.

[0170] The central network model built based on the fault prediction network is distributed to each branch network, and the model parameters are initialized in the branch network, including the network structure, weights, and bias factors, etc., to ensure that the initial model parameters of all branch networks are consistent with the central network.

[0171] In each branch network, data collection and preprocessing are performed independently, the collected data is input into the model of the branch network, forward propagation calculation is performed, the loss function is calculated, the performance of the model on the current data is evaluated, the model parameters of the branch network are updated using the gradient descent method, and the weights and bias factors are adjusted through back propagation. ， The branch network model is trained for multiple rounds of iterations until the preset number of iterations or performance indicators are reached.

[0172] After each branch network completes independent training, the trained model parameters are uploaded to the central network. The central network receives the model parameters from all branch networks and aggregates the received model parameters, usually by averaging or weighted averaging, to obtain updated global model parameters. The updated global model parameters are distributed to all branch networks, and the branch networks replace the original local model parameters with the received global model parameters.

[0173] Perform multiple rounds of iterative training. In each round, the branch network trains independently and uploads model parameters, and the central network aggregates parameters and distributes updates. The iterative process continues until the preset iteration limit is reached.

[0174] After reaching the iteration limit, the central network obtains the final global model parameters and distributes them to all branch networks. Each branch network updates the local model with the received global model parameters to obtain the final branch network model.

[0175] In this embodiment, building a branch network allows parallel processing of data from each communication battery pack, which means that the system can process multiple data streams simultaneously, improving the overall efficiency and processing power of the system; a two-way communication channel is established between the central network and the branch network, so that model parameters can be transmitted and updated between the central network and the branch network, and the collaborative training method allows the central network to aggregate information from multiple branch networks to obtain a more comprehensive data perspective and improve the accuracy and generalization ability of the model; after completing the independent training of the branch network, the central network will collect and aggregate the model parameters of all branch networks to generate updated global model parameters, and the parameter aggregation mechanism can effectively integrate the training results of each branch network and improve the overall performance of the model; through the iterative training process, the system can continuously receive new data and model parameters from the branch network, and update the global model based on this information, so that the model can be continuously optimized with the growth of time and data, and adapt to the changing environment and data distribution; the distributed decision-making and feedback mechanism can help monitor and manage the status of the communication battery pack in real time, and take timely measures to deal with potential faults and problems.

[0176] Figure 2 FIG. 1 is a schematic diagram of the structure of the online remote charging and discharging fault prediction system for a communication storage battery pack according to an embodiment of the present invention. Figure 2 As shown, the system comprises:

[0177] The first unit is used to collect time series data of the charging and discharging process of the communication battery pack, pre-process the time series data, extract complexity features based on multi-scale entropy analysis, determine physical features through parameter identification algorithm based on the physical model corresponding to the battery pack, normalize the complexity features and the physical features into dimensionless processing, and determine the primary feature set of the charging and discharging process of the battery pack;

[0178] The second unit is used to construct a correlation coefficient matrix based on the primary feature set, determine the weight edge of the undirected graph with the correlation coefficient as the weight value, and construct a weighted undirected graph with the primary features as the undirected graph nodes; based on the weighted undirected graph, iteratively calculate the weight edge connectivity sum of each undirected graph node, expand and generate a maximum weight connected subgraph, and use the primary features corresponding to each subgraph node in the maximum weight connected subgraph as the preferred features to construct a preferred feature set;

[0179] The third unit is used to input the battery group data to be predicted into the fault prediction network, extract the corresponding feature values ​​based on the preferred feature set in the input layer, abstract the feature representation layer by layer through multiple hidden layers, extract high-order features, and calculate the probability distribution of fault types through the output layer activation function, and select the fault type corresponding to the highest probability as the fault prediction result.

[0180] According to a third aspect of the embodiments of the present invention,

[0181] An electronic device is provided, comprising:

[0182] processor;

[0183] a memory for storing processor-executable instructions;

[0184] The processor is configured to call the instructions stored in the memory to execute the aforementioned method.

[0185] A fourth aspect of the embodiments of the present invention is:

[0186] A computer-readable storage medium is provided, on which computer program instructions are stored. When the computer program instructions are executed by a processor, the aforementioned method is implemented.

[0187] The present invention may be a method, an apparatus, a system and / or a computer program product. The computer program product may include a computer-readable storage medium carrying computer-readable program instructions for executing various aspects of the present invention.

[0188] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for predicting online remote charging and discharging faults of a communication battery pack, characterized in that: include: Collecting time series data of the charging and discharging process of the communication battery pack, preprocessing the time series data, extracting complexity features based on multi-scale entropy analysis, determining physical features through a parameter identification algorithm based on a physical model corresponding to the battery pack, normalizing the complexity features and the physical features into a dimensionless process, and determining a primary feature set of the charging and discharging process of the battery pack; Based on the primary feature set, a correlation coefficient matrix is ​​constructed, and the weighted edges of the undirected graph are determined using the correlation coefficient as the weight value. The primary features are used as the undirected graph nodes to construct a weighted undirected graph. Based on the weighted undirected graph, by iteratively calculating the weighted edge connectivity sum of each undirected graph node, a maximum weight connected subgraph is generated by expansion, and the primary features corresponding to each subgraph node in the maximum weight connected subgraph are taken as preferred features to construct a preferred feature set; The battery group data to be predicted is input into the fault prediction network. The corresponding feature values ​​are extracted based on the preferred feature set in the input layer. The feature representation is abstracted layer by layer through multiple hidden layers, and high-order features are extracted. The probability distribution of the fault type is calculated through the output layer activation function, and the fault type corresponding to the highest probability is selected as the fault prediction result.

2. The method according to claim 1, characterized in that: Preprocess the time series data, extract the complexity features based on multi-scale entropy analysis, determine the physical features through parameter identification algorithm based on the physical model corresponding to the battery pack, normalize the complexity features and the physical features into dimensionless features, and determine that the primary feature set of the battery pack charging and discharging process includes: Detect and remove outliers from time series data, and obtain standard time series data through smoothing filtering and normalization. Performing a coarse-graining operation on the standard time series data to obtain scale subsequences at different scales, calculating sample entropy for each scale subsequence, determining a multi-scale entropy curve, extracting slope features, curvature features, entropy value features and area features from the multi-scale entropy curve, and constructing complexity features; Based on the electrochemical characteristics and thermodynamic behavior of the battery pack, the corresponding physical model is constructed. The physical model parameters are evaluated and the physical characteristics are determined through the parameter identification algorithm based on the recursive least squares method. The complexity characteristics and the physical characteristics are normalized and dimensionless, the numerical range and dimension differences are eliminated, and a primary feature set is determined.

3. The method according to claim 1, characterized in that Based on the primary feature set, a correlation coefficient matrix is ​​constructed, and the weighted edges of the undirected graph are determined using the correlation coefficient as the weight value. The primary features are used as the undirected graph nodes to construct a weighted undirected graph. Based on the weighted undirected graph, the weighted edge connectivity sum of each undirected graph node is iteratively calculated to expand and generate a maximum weight connected subgraph, and the primary feature corresponding to each connected node in the maximum weight connected subgraph is used as the preferred feature. Constructing the preferred feature set includes: Based on the primary feature set, the correlation coefficient is determined by calculating the covariance and variance between every two primary features. If the absolute value of the correlation coefficient is greater than a preset correlation coefficient threshold, the corresponding correlation coefficient is retained, otherwise it is removed, and a correlation coefficient matrix is ​​constructed based on the retained correlation coefficients; Taking each primary feature in the primary feature set as an undirected graph node, traversing each correlation coefficient in the correlation coefficient matrix, taking the correlation coefficient as a weight value, determining an undirected graph weight edge between two corresponding undirected graph nodes, and constructing a weighted undirected graph; Traverse each undirected graph node in the weighted undirected graph, calculate the undirected weight sum of all undirected graph weight edges connected to the undirected graph node, select the undirected graph node corresponding to the maximum value of the undirected weight sum as the first connected node, initialize the connected subgraph, and start iterative expansion of the connected subgraph: Traversing each connected node of the connected subgraph, determining a corresponding undirected graph node from the weighted undirected graph, finding all undirected graph nodes that are directly connected to the undirected graph node and are not in the connected subgraph, determining a candidate node set, calculating a weighted edge connectivity sum in the weighted undirected graph for each candidate node in the candidate node set and all connected nodes in the connected subgraph, selecting a candidate node corresponding to the maximum value of the weighted edge connectivity sum as a new connected node, and adding it to the connected subgraph; Until the candidate node set is empty, a maximum weight connected subgraph is generated; The primary feature corresponding to each connected node in the maximum weight connected subgraph is taken as the preferred feature to construct a preferred feature set.

4. The method according to claim 3, characterized in that The construction process of the weighted undirected graph also includes: Based on the correlation coefficient matrix and a preset correlation coefficient threshold, binarizing the correlation coefficient matrix to determine a binarized correlation matrix; Based on the binary correlation matrix, the Euclidean distance between every two features is calculated, a feature distance matrix is ​​determined, the maximum value and the minimum value in the feature distance matrix are extracted, the feature distance ratio is determined, and a diversity measurement matrix is ​​generated; Each primary feature in the primary feature set is taken as an undirected graph node, each correlation coefficient in the correlation coefficient matrix and each diversity metric in the diversity metric matrix are traversed, a weight value is calculated through the correlation coefficient and the diversity metric, an undirected graph weight edge between two corresponding undirected graph nodes is determined, and a weighted undirected graph is constructed.

5. The method according to claim 1, characterized in that The battery group data to be predicted is input into the fault prediction network. Based on the preferred feature set, the corresponding feature values ​​are extracted in the input layer. The feature representation is abstracted layer by layer through multiple hidden layers, and high-order features are extracted. The fault type probability distribution is calculated through the output layer activation function, and the fault type corresponding to the highest probability is selected as the fault prediction result, including: The fault prediction network is constructed based on a deep belief network, and includes an input layer, at least one hidden layer and an output layer, wherein the number of nodes in the input layer is equal to the number of features in the preferred feature set, the hidden layer is composed of multiple layers of stacked probabilistic neural units, and the number of nodes in the output layer corresponds one-to-one to the fault type; Based on the unsupervised learning method, primary training is performed layer by layer on each layer of the probabilistic neural unit of the fault prediction network. The probabilistic neural unit of the previous layer outputs the activation probability of the hidden layer, which enters the input layer of the probabilistic neural unit of the current layer. In the hidden layer of the probabilistic neural unit of the current layer, the weight and bias factor of the probabilistic neural unit of the current layer are adjusted by minimizing the energy function to generate the probability distribution corresponding to the input data of the probabilistic neural unit of the current layer. When the primary training of all the probabilistic neural units is completed, the fault prediction network is secondary adjusted based on a supervised learning method, and the fault prediction network is initialized based on the weights and bias factors corresponding to each layer of the probabilistic neural units determined by the primary training; Adding a classifier to the output layer, taking different fault types as labels, minimizing the classification loss function through an optimization algorithm, updating the weight and bias factor of each layer in the fault prediction network, and obtaining an optimal fault prediction network; The battery group data to be predicted is input into the optimal fault prediction network, the fault type probability distribution is input, and the fault type with the highest probability is selected as the fault prediction result.

6. The method according to claim 5, characterized in that The previous layer of probabilistic neural units outputs the hidden layer activation probability and enters the input layer of the current layer of probabilistic neural units. In the hidden layer of the current layer of probabilistic neural units, the weights and bias factors of the current layer of probabilistic neural units are adjusted by minimizing the energy function to generate the probability distribution corresponding to the input data of the current layer of probabilistic neural units: The energy function has the following formula: Where l represents the hidden layer number of the fault prediction network, v l represents the input layer state, h l represents the hidden layer state, E(v l ,h l ) represents the energy function, W l represents the connection weight matrix between the input layer and the hidden layer, a l represents the input layer bias vector, b l represents the hidden layer bias vector, U l ij represents the interaction weight between node i and node j in the input layer, v l i represents the input layer state of input layer node i, v l j represents the input layer state of input layer node j, V l km represents the interaction weight between node k and node m in the hidden layer, h l k represents the hidden layer state of hidden layer node k, h l m represents the hidden layer state of the hidden layer node m, and λ represents the regularization term balance parameter.

7. The method according to claim 5, characterized in that Based on the fault prediction network, constructing a distributed prediction network includes: A distributed prediction network is constructed with a communication battery pack and corresponding edge computing nodes as the branch network and a central server as the core network; The central network is constructed based on the fault prediction network, and the central network is distributed to the branch networks; data collection and training are performed independently on each branch network, and the model parameters in each branch network after training are uploaded to the central network. The central network aggregates the model parameters to obtain updated model parameters, and distributes them to the branch networks; Multiple rounds of iterations are performed in sequence until a preset iteration upper limit is reached, and a final central network and a corresponding final branch network are determined, and the final branch network performs fault prediction.

8. A communication battery pack online remote charging and discharging fault prediction system, used to implement the method described in any one of claims 1 to 7, characterized in that: include: The first unit is used to collect time series data of the charging and discharging process of the communication battery pack, pre-process the time series data, extract complexity features based on multi-scale entropy analysis, determine physical features through parameter identification algorithm based on the physical model corresponding to the battery pack, normalize the complexity features and the physical features into dimensionless processing, and determine the primary feature set of the charging and discharging process of the battery pack; The second unit is used to construct a correlation coefficient matrix based on the primary feature set, determine the weight edge of the undirected graph with the correlation coefficient as the weight value, and construct a weighted undirected graph with the primary features as the undirected graph nodes; Based on the weighted undirected graph, by iteratively calculating the weighted edge connectivity sum of each undirected graph node, a maximum weight connected subgraph is generated by expansion, and the primary features corresponding to each subgraph node in the maximum weight connected subgraph are taken as preferred features to construct a preferred feature set; The third unit is used to input the battery group data to be predicted into the fault prediction network, extract the corresponding feature values ​​based on the preferred feature set in the input layer, abstract the feature representation layer by layer through multiple hidden layers, extract high-order features, and calculate the probability distribution of fault types through the output layer activation function, and select the fault type corresponding to the highest probability as the fault prediction result.

9. An electronic device, characterized in that: include: processor; a memory for storing processor-executable instructions; The processor is configured to call the instructions stored in the memory to execute the method described in any one of claims 1 to 7.

10. A computer-readable storage medium having computer program instructions stored thereon, characterized in that: When the computer program instructions are executed by a processor, the method according to any one of claims 1 to 7 is implemented.

Images