Intelligent operation and maintenance management method and system for direct current charging pile

Abstract

The invention relates to a DC charging pile intelligent operation and maintenance management method and system, and belongs to the technical field of charging pile operation and maintenance. Firstly, environment indexes are established by counting road data and fault rate around the direct current charging pile; secondly, collecting charging direct-current voltage and direct-current signals, obtaining direct-current voltage and direct-current time sequence data through analog-to-digital conversion and an EEMD (ensemble empirical mode decomposition) algorithm, calculating sudden change, intensity, fluctuation and trend indexes according to the time sequence data, and constructing a reliability index; meanwhile, collecting power and electric energy quality parameters in real time, performing feature aggregation and pooling through a graph isomorphic neural network, extracting electric energy features in combination with a convolutional neural network, fusing the electric energy features through an attention mechanism, inputting the electric energy features into a multi-layer perceptron to obtain an error prediction value, and further establishing a performance index in combination with a true value; and finally, constructing an operation and maintenance management index by integrating the environment, reliability and performance indexes, and making an operation and maintenance strategy according to the operation and maintenance management index.

Description

Technical Field

[0001] This invention relates to an intelligent operation and maintenance management method and system for DC charging piles, belonging to the field of charging pile operation and maintenance technology. Background Technology

[0002] With the continuous expansion of the electric vehicle market, DC charging piles, as key equipment for fast charging, have become crucial for ensuring the stability and safety of charging services through their operation and maintenance management. Because DC charging piles operate in complex and ever-changing environments, they face challenges such as equipment aging and frequent failures. Therefore, effective operation and maintenance management is essential for ensuring the continuity and reliability of charging services. Traditional DC charging pile network operation and maintenance presents a series of challenges. First, traditional operation and maintenance often requires a significant investment of human resources and time, which is undoubtedly a huge burden for large-scale DC charging pile networks. Second, traditional operation and maintenance is slow to respond to problems and usually requires human intervention to resolve them. This inefficient operation and maintenance approach leads to instability and service interruptions in the DC charging pile network, causing significant inconvenience to users' charging experience.

[0003] For example, Chinese invention patent application CN117952592A discloses an intelligent management method for DC charging piles, including the following steps: acquiring historical operating data and corresponding historical environmental and maintenance data of DC charging piles in chronological order; conducting a health status assessment of DC charging piles based on the historical operating data and corresponding historical environmental and maintenance data; acquiring real-time operating data and real-time environmental data of DC charging piles; inputting the health status assessment results, real-time operating data, and real-time environmental data of DC charging piles into a pre-trained fault prediction model to predict fault risks; and managing and maintaining DC charging piles based on the fault risk prediction results. However, the aforementioned patent uses an improved Osprey optimization algorithm to optimize weights, which involves a complex iterative process, including physical fatigue functions and hovering strategies, requiring multiple iterations, increasing the latency of real-time prediction. Especially in large-scale DC charging pile networks, high-performance hardware support is required, placing high demands on the hardware.

[0004] In summary, there is an urgent need for a method and system that can achieve intelligent operation and maintenance management of DC charging piles without a complex iterative process. Summary of the Invention

[0005] To address the problems existing in the prior art, this invention proposes an intelligent operation and maintenance management method and system for DC charging piles.

[0006] The technical solution of the present invention is as follows: On the one hand, this invention proposes an intelligent operation and maintenance management method for DC charging piles, including the following steps: The road conditions and DC charging pile failure rates of the DC charging pile locations within the managed area are statistically analyzed. Based on the road conditions and DC charging pile failure rates, environmental indicators for DC charging piles are calculated, including road index and failure index. The DC voltage and DC current signals during the charging process of the DC charging pile are collected, and the time-series data corresponding to the DC voltage and DC current signals are obtained by analog-to-digital conversion and EEMD algorithm. Based on the time-series data, the operational reliability index of the DC charging pile is calculated, including the sudden change, intensity, fluctuation and trend index corresponding to the DC voltage and DC current. The system collects power and power quality parameters of DC charging piles in real time, extracts feature vectors of power quality parameters using graph isomorphic neural networks, and extracts feature vectors of power quality parameters using convolutional neural networks. The feature vectors of power quality parameters and power quality parameters are updated through an attention mechanism to obtain a set of feature vectors. The set of feature vectors is used as the input of a pre-trained multilayer perceptron to obtain the predicted value of the metering error of DC charging piles in the area to be managed. The operating performance index of DC charging piles is constructed based on the predicted value of metering error and the actual value of metering error. Based on the environment, operational reliability, and operational performance indicators of DC charging piles, an operation and maintenance management index for DC charging piles is calculated, and corresponding operation and maintenance management strategies are formulated based on the operation and maintenance management index.

[0007] Preferably, the calculation of the environmental indicators for DC charging piles based on the road conditions and the failure rate of DC charging piles is specifically as follows: The road conditions include road grade, traffic density, lane width, and number of lanes in one direction. A road index is calculated based on these road conditions, expressed by the formula: ; In the formula, Indicates road index, Represents the natural constant. Indicates traffic density. Indicates road grade, Indicates the number of lanes in one direction. Indicates the width of a single lane; The failure index is calculated based on the failure rate of DC charging piles, and expressed by the formula: ; In the formula, Indicates the failure index. This indicates the number of faults in DC charging stations. This indicates the total number of DC charging stations installed. Indicates the statistical time period. This indicates the total number of faults of the DC charging station within the statistical period. Index representing the number of DC charging pile failures. Indicates the DC charging pile has experienced the first occurrence. The time required to repair the fault; Based on the calculated road index and fault index, an environmental index for DC charging piles is constructed, expressed by the formula: ; In the formula, This indicates the environmental indicators of DC charging piles.

[0008] Preferably, the reliability index of DC charging pile operation is calculated based on time-series data as follows: The mutation index is expressed by the formula: ; In the formula, Indicates the mutation index. This represents the timing data corresponding to DC voltage or DC current signals in the frequency band. The total time from mutation to stable operation. Timing data representing DC voltage or DC current signals in the frequency band The time during which the system remains stable without any sudden changes; The strength index is expressed by the formula: ; In the formula, Indicates the intensity index. This indicates that the maximum value is used for calculation. This indicates that the minimum value is used for calculation. Indicates at the sampling point The DC voltage or DC current value at that location. This represents the average value of the DC voltage or DC current during the operation of the DC charging pile. The volatility index is expressed by the formula: ; In the formula, Indicates volatility index, Indicates the maximum number of sampling points. Indicates from sampling point to sampling point Perform cumulative calculation; The time-series data is divided into subsequences according to different preset lengths to obtain corresponding subsequences. A trend index is calculated based on the subsequences, expressed by the formula: ; In the formula, Indicates a trend index. This represents the linear regression function. Indicates the preset length. This represents the difference between the maximum and minimum values ​​of a subsequence. The standard deviation of the subsequence is represented. Indicates the total number of subsequences. Indicates the subsequence index; The abrupt change index, intensity index, fluctuation index, and trend index of the DC voltage and DC current signals are calculated respectively. Based on the calculation results, the corresponding operational reliability indexes are calculated respectively. The operational reliability index of the DC voltage signal is expressed by the formula: ; In the formula, This indicates the operational reliability of the DC voltage signal. Indicates taking Exponentiation, This represents the arctangent operation; The formula for calculating the operational reliability index of DC current signals is the same as that for DC voltage signals, expressed as follows: ; Based on the operational reliability indicators of DC voltage and DC current signals, a comprehensive operational reliability index for DC charging piles is constructed. This can be expressed as a formula: .

[0009] Preferably, the feature vectors for extracting power quality parameters using graph isomorphic neural networks are as follows: The power quality parameters include DC voltage amplitude, grid frequency, temperature, DC voltage harmonic components, three-phase DC voltage unbalance, and DC voltage and DC current. The power quality parameters are normalized, and each normalized power quality parameter is mapped to a corresponding node in a preset graph structure as the initial feature vector of the node. The preset graph structure contains multiple nodes, and each node represents a type of power quality parameter. Based on the data sequence of various power quality parameters, the behavioral similarity between any two nodes in the graph structure is dynamically calculated. Based on a preset similarity threshold, the connection relationship between nodes is determined, and node pairs whose behavioral similarity meets the preset conditions are identified as adjacent to each other. Based on the judgment results, a graph topology structure describing the correlation between power quality parameters is constructed. The constructed graph topology is input into the graph isomorphic neural network. Through the iterative calculation of the graph isomorphic neural network, each node aggregates the feature information of its neighboring nodes and, combined with the initial feature vector of the node, updates and generates a new feature vector containing neighborhood information. The new feature vectors of the nodes output by each layer of the graph isomorphic neural network are subjected to average pooling operation, and the features obtained after average pooling operation are integrated to output a feature vector set of power quality parameters.

[0010] Preferably, the feature vector for extracting power quality parameters using a convolutional neural network is as follows: The DC voltage and DC current features in power quality parameters are extracted using a convolutional neural network, expressed by the following formula: ; ; In the formula, This indicates that a CNN convolution operation is being performed. This represents the extracted DC voltage feature vector. This represents the extracted DC current feature vector. Indicates DC voltage. Indicates direct current; By combining the DC voltage characteristic vector and the DC current characteristic vector, a set of electrical energy parameter characteristic vectors is constructed. .

[0011] Preferably, updating the feature vectors of power quality parameters and energy quality parameters through an attention mechanism specifically involves: Using an attention mechanism, an influence matrix representing the correlation between the feature vectors of the power quality parameters and the power quality parameters is constructed based on the feature vector sets of the power quality parameters and the power quality parameters. Based on this influence matrix, each feature vector in the feature vector set of the power quality parameters and each feature vector in the feature vector set of the power quality parameters are updated, as expressed by the following formula: ; ; ; In the formula, This represents the updated power quality parameter feature vector. This represents the updated eigenvector of the power quality parameters. Represents the influence matrix. Represents the weight matrix. Represents the transpose of a matrix. This represents the hyperbolic tangent activation function. The eigenvector set representing the power quality parameters is the first... 1 eigenvector This represents the set of feature vectors representing power quality parameters. The eigenvector set representing the power quality parameter is the first... 1 eigenvector This represents the set of eigenvectors representing power quality parameters. The weight matrix represents the eigenvectors of power quality parameters. The weight matrix represents the eigenvectors of power quality parameters. Represents the graph topology.

[0012] Preferably, the specific steps for constructing the operational performance indicators include: Using the set of feature vectors as input to a pre-trained multilayer perceptron, the predicted metering error of DC charging piles in the managed area is obtained, expressed by the formula: ; In the formula, This represents the predicted value of the measurement error. This represents a multilayer perceptron. This represents the updated power quality parameter feature vector. This represents the updated eigenvector of the power quality parameters; The operating performance index of DC charging piles is constructed based on the predicted value of metering error and the actual value of metering error, expressed by the formula: ; ; In the formula, Indicates operational performance indicators, Represents probability. The standard deviation of the true value of measurement error. The variance of the true value of the measurement error. Indicates sample The predicted value of measurement error, Indicates sample The true value of the measurement error, Indicates the sample index. Indicates the true value of the measurement error. Indicates the collection of electrical energy. This represents actual electrical energy.

[0013] Preferably, based on the DC charging pile's environment, operational reliability, and operational performance indicators, the DC charging pile operation and maintenance management index is calculated, expressed by the formula: ; In the formula, , and These are the weighted parameters corresponding to the preset DC charging pile environment, operational reliability, and operational performance indicators. This represents the operation and maintenance management index of DC charging piles. Indicators of environmental indicators for DC charging piles Normalized values, Indicators of operational reliability Normalized values, Indicates operating performance indicators The normalized value.

[0014] On the other hand, the present invention also proposes an intelligent operation and maintenance management system for DC charging piles, including the following modules: The environmental index construction module is used to statistically analyze the road conditions and DC charging pile failure rate of the DC charging pile locations within the managed area, and calculate the DC charging pile environmental index based on the road conditions and DC charging pile failure rate, including the road index and the failure index. The operational reliability index construction module is used to collect DC voltage and DC current signals during the charging process of DC charging piles, and use analog-to-digital conversion and EEMD algorithm to obtain the time series data corresponding to DC voltage and DC current signals; based on the time series data, the operational reliability index of DC charging piles is calculated, including the sudden change, intensity, fluctuation and trend index corresponding to DC voltage and DC current; The operational performance index construction module is used to collect power and power quality parameters of DC charging piles in real time. It uses a graph isomorphic neural network to extract feature vectors of power quality parameters and a convolutional neural network to extract feature vectors of power quality parameters. The feature vectors of power quality parameters and power quality parameters are updated through an attention mechanism to obtain a feature vector set. The feature vector set is used as the input of a pre-trained multilayer perceptron to obtain the metering error prediction value of DC charging piles in the area to be managed. Based on the metering error prediction value and the actual metering error value, the operational performance index of DC charging piles is constructed. The operation and maintenance management module is used to calculate the operation and maintenance management index of DC charging piles based on the environment, operational reliability and operational performance indicators of DC charging piles, and to formulate corresponding operation and maintenance management strategies based on the operation and maintenance management index.

[0015] In another aspect, the present invention also proposes an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the method as described in any embodiment of the present invention.

[0016] The present invention has the following beneficial effects: (1) This invention is a method and system for intelligent operation and maintenance management of DC charging piles. By comprehensively evaluating the surrounding road conditions and its own fault history, environmental indicators are constructed, which effectively reflect the complexity of the external environment and operation and maintenance pressure of the equipment, improve the ability of operation and maintenance management to perceive geographical and facility conditions, and make the strategy formulation more in line with the actual scenario requirements.

[0017] (2) This invention is a method and system for intelligent operation and maintenance management of DC charging piles. It processes the DC voltage and DC current signals of charging through analog-to-digital conversion and EEMD algorithm, extracts the time sequence characteristics of DC components, and further constructs four indices: mutation, intensity, fluctuation and trend, thereby establishing reliability indicators and significantly improving the accuracy of capturing abnormal operating status and performance degradation of DC charging piles and the ability to provide early warning.

[0018] (3) This invention is a method and system for intelligent operation and maintenance management of DC charging piles. By introducing graph isomorphic neural networks and convolutional neural networks to extract the deep features of power and energy parameters respectively, and by using the attention mechanism to achieve multimodal feature fusion, the accurate prediction of metering error is achieved based on multilayer perceptron, and finally the performance index is constructed, which improves the accuracy and reliability of DC charging pile status assessment and the level of intelligent operation and maintenance decision-making. Attached Figure Description

[0019] Figure 1 This is a flowchart of the operation and maintenance management method provided in Embodiment 1 of the present invention. Detailed Implementation

[0020] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0021] It should be understood that the step numbers used in the text are for ease of description only and are not intended to limit the order in which the steps are performed.

[0022] It should be understood that the terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to limit the invention. As used in this specification and the appended claims, the singular forms “a,” “an,” and “the” are intended to include the plural forms unless the context clearly indicates otherwise.

[0023] The terms “comprising” and “including” indicate the presence of the described feature, whole, step, operation, element and / or component, but do not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components and / or collections thereof.

[0024] The term “and / or” refers to any combination of one or more of the associated listed items, as well as all possible combinations, and includes these combinations.

[0025] Example 1: See Figure 1This embodiment provides an intelligent operation and maintenance management method for DC charging piles, the method including: S100. Collect road conditions data for the locations of DC charging piles within the managed area, including road grade, traffic density, lane width, and number of lanes in one direction. Based on this data, calculate the road index for the locations of DC charging piles within the managed area, expressed by the formula: ; In the formula, Indicates road index, Represents the natural constant. Indicates traffic density. Indicates road grade, Indicates the number of lanes in one direction. Indicates the width of a single lane; It should be noted that traffic density is divided into three levels: Level 1 has 60 to 80 vehicles per kilometer in each one-way lane, Level 2 has 40 to 60 vehicles per kilometer in each one-way lane, Level 3 has 20 to 40 vehicles per kilometer in each one-way lane, and Level 4 has less than 20 vehicles per kilometer in each one-way lane. Furthermore, the road classifications are as follows: arterial roads are Class 1, secondary arterial roads are Class 2, and local roads are Class 3; the number of lanes in one direction is as follows: 4 lanes in one direction for Class 1, 3 lanes in one direction for Class 2, 2 lanes in one direction for Class 3, and 1 lane in one direction for Class 4; the width of a single lane is as follows: Level 1, single lane width and Level 2, single lane width and It is level 3.

[0026] S101. Obtain the failure rate of DC charging piles in the area to be managed, and calculate the failure index of DC charging piles in the area to be managed, expressed by the formula: ; In the formula, Indicates the failure index. This indicates the number of faults in DC charging stations. This indicates the total number of DC charging stations installed. Indicates the statistical time period. This indicates the total number of faults of the DC charging station within the statistical period. Index representing the number of DC charging pile failures. Indicates the DC charging pile has experienced the first occurrence. The time required to repair the fault.

[0027] S102. Based on the calculated road index and fault index, construct the environmental indicators for DC charging piles, expressed by the formula: ; In the formula, This indicates the environmental indicators of DC charging piles.

[0028] S200: Collects DC voltage and DC current output signals during the charging process of the DC charging pile, and obtains time-series data of DC voltage and DC current components using analog-to-digital conversion and EEMD algorithm; calculates the sudden changes, intensity, fluctuations, and trend indices of DC voltage and DC current based on the time-series data, where: The mutation index, expressed by the formula: ; In the formula, Indicates the mutation index. This represents the timing data corresponding to DC voltage or DC current signals in the frequency band. The total time from mutation to stable operation. This represents the timing data corresponding to DC voltage or DC current signals in the frequency band. The time during which the system remains stable without any sudden changes; The strength index, expressed by the formula: ; In the formula, Indicates the intensity index. This indicates that the maximum value is used for calculation. This indicates that the minimum value is used for calculation. Indicates at the sampling point DC voltage value of DC power supply, sampling point , This represents the average value of the DC voltage or DC current during the operation of the DC charging pile. The volatility index, expressed by the formula: ; In the formula, Indicates volatility index, Indicates the maximum number of sampling points. Indicates from sampling point to sampling point Perform cumulative calculation; The time-series data is divided into subsequences according to different preset lengths to obtain corresponding subsequences. A trend index is calculated based on the subsequences, expressed by the formula: ; In the formula, Indicates a trend index. This represents the linear regression function. Indicates the preset length. This represents the difference between the maximum and minimum values ​​of a subsequence. The standard deviation of the subsequence is represented. Indicates the total number of subsequences. Indicates the subsequence index; The abrupt change index, intensity index, fluctuation index, and trend index of the DC voltage and DC current signals are calculated respectively. Based on the calculation results, the corresponding operational reliability indexes are calculated respectively. The operational reliability index of the DC voltage signal is expressed by the formula: ; In the formula, This indicates the operational reliability of the DC voltage signal. Indicates taking Exponentiation, This represents the arctangent operation; The formula for calculating the operational reliability index of DC current signals is the same as that for DC voltage signals, expressed as follows: ; Based on the operational reliability indicators of DC voltage and DC current signals, a comprehensive operational reliability index for DC charging piles is constructed. This can be expressed as a formula: ; In the formula, This indicates the overall operational reliability index of DC charging piles.

[0029] S300: Real-time acquisition of power quality parameters and energy quality parameters of DC charging piles, including: Power quality parameters are expressed by the following formula: ; In the formula, Indicates the DC voltage amplitude. Indicates the power grid frequency. Indicates temperature. This represents the third harmonic component of DC voltage. This represents the fifth harmonic component of DC voltage. Indicates sampling point DC voltage amplitude at that location Indicates sampling point The power grid frequency at that location, Indicates sampling point Temperature value at that location, Indicates sampling point The third harmonic component of the DC voltage. Indicates sampling point The fifth harmonic component of the DC voltage; In addition, the amplitude of the sampling points in the three-phase DC voltage is collected, and the three-phase unbalance is calculated based on the amplitude, expressed by the formula: ; ; In the formula, Indicates the degree of three-phase DC voltage imbalance. Indicates sampling point Three-phase DC voltage imbalance Indicates the first Three-phase imbalance at each sampling point Indicates the first The instantaneous DC voltage amplitude of phase A collected at each sampling point. Indicates the first The instantaneous DC voltage amplitude of phase B collected at each sampling point. Indicates the first The instantaneous DC voltage amplitude of phase C collected at each sampling point. This represents the average value of the three-phase DC voltage.

[0030] Power quality parameters include: DC voltage DC current .

[0031] S301. Utilize graph isomorphic neural networks to extract features from the power quality parameters of DC charging piles. Specifically: The graph structure is constructed based on power quality parameters, and can be expressed by the following formula: ; ; In the formula, Representing the graph structure, Represents a set of nodes. Denotes the set of edges. Indicates the DC voltage amplitude. Indicates the power grid frequency. Indicates temperature. Indicates the 3rd harmonic, Indicates the 5th harmonic, Indicates the degree of three-phase imbalance; The power quality parameters are normalized, and each normalized power quality parameter is mapped to a corresponding node in a preset graph structure as the initial feature vector of the node. The preset graph structure contains multiple nodes, each representing a class of power quality parameters, as expressed by the formula: ; In the formula, Represents a node eigenvectors, This represents the normalized DC voltage amplitude. Represents a node eigenvectors, This represents the normalized power grid frequency. Represents a node eigenvectors, This represents the normalized temperature. Represents a node eigenvectors, This represents the third harmonic component of the normalized DC voltage. Represents a node eigenvectors, This represents the fifth harmonic component of the normalized DC voltage. Represents a node eigenvectors, This represents the normalized three-phase DC voltage imbalance. Represents the normalized sampling points DC voltage amplitude at that location Represents the normalized sampling points The power grid frequency at that location, Represents the normalized sampling points Temperature value at that location, Represents the normalized sampling points The third harmonic component of the DC voltage. Represents the normalized sampling points The fifth harmonic component of the DC voltage. Represents the normalized sampling points Three-phase DC voltage imbalance.

[0032] Based on the data sequence of various power quality parameters, the behavioral similarity between any two nodes in the graph structure is dynamically calculated. In this embodiment, through calculation and Euclidean distance between two nodes It reflects the behavioral similarity between any two nodes in the graph structure; It should be noted that, It is a formula for measuring the behavioral similarity between two nodes in a multidimensional space; if the feature of each node is a feature containing... The sequence of sampling points is to calculate the sum of these two sequence vectors. Behavioral similarity in 3D space; Based on a preset behavioral similarity threshold, determine the connection relationship between nodes: when express and Two nodes with high behavioral similarity are adjacent nodes; when express and Two nodes with low behavioral similarity do not constitute adjacent nodes; Furthermore, nodes whose behavioral similarity meets the preset conditions are identified as adjacent to each other, and a graph topology describing the relationship between power quality parameters is constructed based on the judgment results. The constructed graph topology is input into a graph isomorphic neural network. Through iterative computation of the graph isomorphic neural network, each node aggregates the feature information of its neighboring nodes and, combined with the node's initial feature vector, updates and generates a new feature vector containing neighborhood information, as expressed by the formula: ; In the formula, This represents the layer index of a graph isomorphic neural network. Indicates the first Nodes in a layered graph isomorphic neural network New feature vectors of nodes containing neighborhood information Indicates the first Nodes in a layered graph isomorphic neural network New feature vectors of nodes containing neighborhood information Indicates the first Multilayer perceptron of layer graph isomorphic neural network. Indicates the first Learnable parameters of a layered graph isomorphic neural network. Represents a node The set of all neighboring nodes, Indicates the first Nodes in a layered graph isomorphic neural network a certain neighbor node eigenvectors, Indicates neighboring nodes.

[0033] S302, the graph isomorphic neural network of the first... Layer output node The new feature vectors are subjected to average pooling, and the features obtained after average pooling are integrated to output the feature vector set of power quality parameters, expressed by the formula: ; In the formula, This represents the set of eigenvectors representing power quality parameters. This indicates the average pooling operation. This represents the summation operation; This represents the total number of layers in a graph isomorphic neural network. In this embodiment, This involves performing aggregation and average pooling operations on a three-layer graph isomorphic neural network to obtain the whole graph features at different levels, and outputting a set of power quality parameter feature vectors, expressed by the formula: ; In the formula, These represent the power quality parameter feature vectors output by the first to third layer graph isomorphic neural networks, respectively.

[0034] S303. Extract the feature vectors of power quality parameters using a convolutional neural network, specifically: The DC voltage and DC current features in power quality parameters are extracted using a convolutional neural network, expressed by the following formula: ; ; In the formula, This indicates that a CNN convolution operation is being performed. This represents the extracted DC voltage feature vector. This represents the extracted DC current feature vector. Indicates DC voltage. Indicates direct current; By combining the DC voltage characteristic vector and the DC current characteristic vector, a set of electrical energy parameter characteristic vectors is constructed. .

[0035] S304. Using an attention mechanism, based on the power quality parameter feature vector set and the power quality parameter feature vector set, construct an influence matrix representing the correlation between the feature vectors of the power quality parameters and the power quality parameters; based on the influence matrix, update each feature vector in the power quality parameter feature vector set and each feature vector in the power quality parameter feature vector set, as expressed by the formula: ; ; ; In the formula, This represents the updated power quality parameter feature vector. This represents the updated eigenvector of the power quality parameters. Represents the influence matrix. Represents the weight matrix. Represents the transpose of a matrix. This represents the hyperbolic tangent activation function. The eigenvector set representing the power quality parameters is the first... 1 eigenvector This represents the set of feature vectors representing power quality parameters. The eigenvector set representing the power quality parameter is the first... 1 eigenvector This represents the set of eigenvectors representing power quality parameters. The weight matrix represents the eigenvectors of power quality parameters. The weight matrix represents the eigenvectors of power quality parameters. Represents the graph topology.

[0036] S305. Using the set of feature vectors as input to the pre-trained multilayer perceptron, the predicted metering error value of the DC charging piles in the area to be managed is obtained, expressed by the formula: ; In the formula, This represents the predicted value of the measurement error. This represents a multilayer perceptron. This represents the updated power quality parameter feature vector. This represents the updated eigenvector of the power quality parameters; It should be noted that an attention mechanism has been introduced. and These features are concatenated to form a comprehensive feature vector, which is then input into the MLP model. The role of the MLP is to learn the mapping relationship between these complex features and the final measurement error value through multiple layers of nonlinear transformation.

[0037] The loss function is set as follows, expressed by the formula: ; In the formula, Represents the loss function. Indicates the number of training samples. Indicates sample The predicted value of measurement error, Indicates sample The true value of the measurement error; It should be noted that the weight parameters inside the MLP model are continuously adjusted through the backpropagation algorithm to minimize the loss function value. The smaller the loss function value, the closer the predicted value of the measurement error of the MLP model is to the true value of the measurement error, and the stronger the predictive ability of the model.

[0038] The operating performance index of DC charging piles is constructed based on the predicted value of metering error and the actual value of metering error, expressed by the formula: ; ; In the formula, Indicates operational performance indicators, Represents probability. The standard deviation of the true value of measurement error. The variance of the true value of the measurement error. Indicates sample The predicted value of measurement error, Indicates sample The true value of the measurement error, Indicates the sample index. Indicates the true value of the measurement error. Indicates the collection of electrical energy. This represents actual electrical energy.

[0039] S400, based on the DC charging pile environment, operational reliability, and operational performance indicators, calculates the DC charging pile operation and maintenance management index, expressed by the formula: ; In the formula, , and These are the weighted parameters corresponding to the preset DC charging pile environment, operational reliability, and operational performance indicators. This represents the operation and maintenance management index of DC charging piles. Indicators of environmental indicators for DC charging piles Normalized values, Indicators of operational reliability Normalized values, Indicates operating performance indicators Normalized values; Based on the operation and maintenance management index, corresponding operation and maintenance management strategies are formulated, expressed by the formula: ; In the formula, This represents the minimum value of the preset operation and maintenance management index threshold. This indicates the maximum value of the preset operation and maintenance management index threshold; It should be noted that when the operation and maintenance management index is less than the minimum threshold value, the DC charging pile does not require operation and maintenance; when the operation and maintenance management index is greater than or equal to the minimum threshold value and less than or equal to the maximum threshold value, the DC charging pile will be operated and maintained normally on a quarterly basis; when the operation and maintenance management index is greater than the maximum threshold value, the DC charging pile needs to be operated and maintained immediately.

[0040] Example 2: This embodiment provides an intelligent operation and maintenance management system for DC charging piles, the system including the following modules: The environmental index construction module is used to statistically analyze the road conditions and DC charging pile failure rate of the DC charging pile locations within the managed area, and calculate the DC charging pile environmental index based on the road conditions and DC charging pile failure rate, including the road index and the failure index. The operational reliability index construction module is used to collect DC voltage and DC current signals during the charging process of DC charging piles, and use analog-to-digital conversion and EEMD algorithm to obtain the time series data corresponding to DC voltage and DC current signals; based on the time series data, the operational reliability index of DC charging piles is calculated, including the sudden change, intensity, fluctuation and trend index corresponding to DC voltage and DC current; The operational performance index construction module is used to collect power and power quality parameters of DC charging piles in real time. It uses a graph isomorphic neural network to extract feature vectors of power quality parameters and a convolutional neural network to extract feature vectors of power quality parameters. The feature vectors of power quality parameters and power quality parameters are updated through an attention mechanism to obtain a feature vector set. The feature vector set is used as the input of a pre-trained multilayer perceptron to obtain the metering error prediction value of DC charging piles in the area to be managed. Based on the metering error prediction value and the actual metering error value, the operational performance index of DC charging piles is constructed. The operation and maintenance management module is used to calculate the operation and maintenance management index of DC charging piles based on the environment, operational reliability and operational performance indicators of DC charging piles, and to formulate corresponding operation and maintenance management strategies based on the operation and maintenance management index.

[0041] Example 3: This embodiment proposes an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, it implements the method described in any embodiment of the present invention.

[0042] Example 4: This embodiment proposes a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the method described in any embodiment of the present invention.

[0043] In this application embodiment, "at least one" refers to one or more, and "more than one" refers to two or more. "And / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent the existence of A alone, A and B simultaneously, or B alone. A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one of the following" and similar expressions refer to any combination of these items, including any combination of singular or plural items. For example, at least one of a, b, and c can represent: a, b, c, a and b, a and c, b and c, or a and b and c, where a, b, and c can be single or multiple.

[0044] Those skilled in the art will recognize that the units and algorithm steps described in the embodiments disclosed herein can be implemented using electronic hardware, computer software, or a combination of electronic hardware and software. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0045] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0046] In the several embodiments provided in this application, any function, if implemented as a software functional unit and sold or used as an independent product, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, essentially, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0047] The above description is merely an embodiment of the present invention and does not limit the patent scope of the present invention. Any equivalent structural or procedural transformations made based on the content of the present invention's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of the present invention.

Claims

1. A method for intelligent operation and maintenance management of DC charging piles, characterized in that, Includes the following steps: The road conditions and DC charging pile failure rates of the DC charging pile locations within the managed area are statistically analyzed. Based on the road conditions and DC charging pile failure rates, environmental indicators for DC charging piles are calculated, including road index and failure index. The DC voltage and DC current signals during the charging process of the DC charging pile are collected, and the time series data corresponding to the DC voltage and DC current signals are obtained by analog-to-digital conversion and EEMD algorithm. Based on the time series data, the operational reliability index of the DC charging pile is calculated, including the sudden change, intensity, fluctuation and trend index corresponding to the DC voltage and DC current. The system collects power and power quality parameters of DC charging piles in real time, extracts feature vectors of power quality parameters using graph isomorphic neural networks, and extracts feature vectors of power quality parameters using convolutional neural networks. The feature vectors of power quality parameters and power quality parameters are updated through an attention mechanism to obtain a set of feature vectors. The set of feature vectors is used as the input of a pre-trained multilayer perceptron to obtain the predicted value of the metering error of DC charging piles in the area to be managed. The operating performance index of DC charging piles is constructed based on the predicted value of metering error and the actual value of metering error. Based on the DC charging pile's environment, operational reliability, and operational performance indicators, a DC charging pile operation and maintenance management index is calculated, and corresponding operation and maintenance management strategies are formulated based on this index.

2. The intelligent operation and maintenance management method for DC charging piles according to claim 1, characterized in that, Based on the aforementioned road conditions and DC charging pile failure rate, the specific environmental indicators for DC charging piles are calculated as follows: The road conditions include road grade, traffic density, lane width, and number of lanes in one direction. A road index is calculated based on these road conditions, expressed by the formula: ； In the formula, Indicates road index, Represents the natural constant. Indicates traffic density. Indicates road grade, Indicates the number of lanes in one direction. Indicates the width of a single lane; The failure index is calculated based on the failure rate of DC charging piles, and expressed by the formula: ； In the formula, Indicates the failure index. This indicates the number of faults in DC charging stations. This indicates the total number of DC charging stations installed. Indicates the statistical time period. This indicates the total number of faults of the DC charging station within the statistical period. Index representing the number of DC charging pile failures. Indicates the DC charging pile has experienced the first occurrence. The time required to repair the fault; Based on the calculated road index and fault index, an environmental index for DC charging piles is constructed, expressed by the formula: ； In the formula, This indicates the environmental indicators of DC charging piles.

3. The intelligent operation and maintenance management method for DC charging piles according to claim 1, characterized in that, The specific reliability indicators for DC charging piles are calculated based on time-series data as follows: The mutation index is expressed by the formula: ； In the formula, Indicates the mutation index. This represents the timing data corresponding to DC voltage or DC current signals in the frequency band. The total time from mutation to stable operation. Timing data representing DC voltage or DC current signals in the frequency band The time during which the system remains stable without any sudden changes; The strength index is expressed by the formula: ； In the formula, Indicates the intensity index. This indicates that the maximum value is used for calculation. This indicates that the minimum value is used for calculation. Indicates at the sampling point The DC voltage or DC current value at that location. This represents the average value of the DC voltage or DC current during the operation of the DC charging pile. The volatility index is expressed by the formula: ； In the formula, Indicates volatility index, Indicates the maximum number of sampling points. Indicates from sampling point to sampling point Perform cumulative calculation; The time-series data is divided into subsequences according to different preset lengths to obtain corresponding subsequences. A trend index is calculated based on the subsequences, expressed by the formula: ； In the formula, Indicates a trend index. This represents the linear regression function. Indicates the preset length. This represents the difference between the maximum and minimum values ​​of a subsequence. The standard deviation of the subsequence is represented. Indicates the total number of subsequences. Indicates the subsequence index; The abrupt change index, intensity index, fluctuation index, and trend index of the DC voltage and DC current signals are calculated respectively. Based on the calculation results, the corresponding operational reliability indexes are calculated. The operational reliability index of the DC voltage signal is expressed by the formula: ； In the formula, This indicates the operational reliability of the DC voltage signal. Indicates taking Exponentiation, This represents the arctangent operation; The formula for calculating the operational reliability index of DC current signals is the same as that for DC voltage signals, expressed as follows: ; Based on the operational reliability indicators of DC voltage and DC current signals, a comprehensive operational reliability index for DC charging piles is constructed. This can be expressed as a formula: 。 4. The intelligent operation and maintenance management method for DC charging piles according to claim 1, characterized in that, The specific steps for extracting the feature vectors of power quality parameters using a graph isomorphic neural network are as follows: The power quality parameters include DC voltage amplitude, grid frequency, temperature, DC voltage harmonic components, and three-phase DC voltage imbalance. The power quality parameters are normalized, and each normalized power quality parameter is mapped to a corresponding node in a preset graph structure as the initial feature vector of the node. The preset graph structure contains multiple nodes, and each node represents a type of power quality parameter. Based on the data sequence of various power quality parameters, the behavioral similarity between any two nodes in the graph structure is dynamically calculated. Based on a preset similarity threshold, the connection relationship between nodes is determined, and node pairs whose behavioral similarity meets the preset conditions are identified as adjacent to each other. Based on the judgment results, a graph topology structure describing the correlation between power quality parameters is constructed. The constructed graph topology is input into the graph isomorphic neural network. Through the iterative calculation of the graph isomorphic neural network, each node aggregates the feature information of its neighboring nodes and, combined with the initial feature vector of the node, updates and generates a new feature vector containing neighborhood information. The new feature vectors of the nodes output by each layer of the graph isomorphic neural network are subjected to average pooling operation, and the features obtained after average pooling operation are integrated to output a feature vector set of power quality parameters.

5. The intelligent operation and maintenance management method for DC charging piles according to claim 4, characterized in that, The specific steps for extracting the feature vectors of power quality parameters using a convolutional neural network are as follows: The DC voltage and DC current features in power quality parameters are extracted using a convolutional neural network, expressed by the following formula: ； ； In the formula, This indicates that a CNN convolution operation is being performed. This represents the extracted DC voltage feature vector. This represents the extracted DC current feature vector. Indicates DC voltage. Indicates direct current; By combining the DC voltage characteristic vector and the DC current characteristic vector, a set of electrical energy parameter characteristic vectors is constructed. .

6. The intelligent operation and maintenance management method for DC charging piles according to claim 5, characterized in that, The feature vectors of power quality parameters and energy quality parameters are updated using an attention mechanism as follows: Using an attention mechanism, an influence matrix representing the correlation between the feature vectors of the power quality parameters and the power quality parameters is constructed based on the feature vector sets of the power quality parameters and the power quality parameters. Based on this influence matrix, each feature vector in the feature vector set of the power quality parameters and each feature vector in the feature vector set of the power quality parameters are updated, as expressed by the following formula: ； ； ； In the formula, This represents the updated power quality parameter feature vector. This represents the updated eigenvector of the power quality parameters. Represents the influence matrix. Represents the weight matrix. To represent the transpose of a matrix, This represents the hyperbolic tangent activation function. The eigenvector set representing the power quality parameters is the first... 1 eigenvector This represents the set of feature vectors representing power quality parameters. The eigenvector set representing the power quality parameter is the first... 1 eigenvector This represents the set of eigenvectors representing power quality parameters. The weight matrix represents the eigenvectors of power quality parameters. The weight matrix represents the eigenvectors of power quality parameters. Represents the graph topology.

7. The intelligent operation and maintenance management method for DC charging piles according to claim 1, characterized in that, The specific steps for constructing the aforementioned performance metrics include: Using the set of feature vectors as input to a pre-trained multilayer perceptron, the predicted metering error of DC charging piles in the managed area is obtained, expressed by the formula: ； In the formula, This represents the predicted value of the measurement error. This represents a multilayer perceptron. This represents the updated power quality parameter feature vector. This represents the updated eigenvector of the power quality parameters; The operating performance index of DC charging piles is constructed based on the predicted value of metering error and the actual value of metering error, expressed by the formula: ； ； In the formula, Indicates operational performance indicators, Represents probability. The standard deviation of the true value of measurement error. The variance of the true value of the measurement error. Indicates sample The predicted value of measurement error, Indicates sample The true value of the measurement error, Indicates the sample index. Indicates the true value of the measurement error. Indicates the collection of electrical energy. This represents actual electrical energy.

8. The intelligent operation and maintenance management method for DC charging piles according to claim 1, characterized in that, Based on the environmental, operational reliability, and operational performance indicators of DC charging piles, the DC charging pile operation and maintenance management index is calculated, expressed by the formula: ； In the formula, , and These are the weighted parameters corresponding to the preset DC charging pile environment, operational reliability, and operational performance indicators. This represents the operation and maintenance management index of DC charging piles. Indicators of environmental indicators for DC charging piles Normalized values, Indicators of operational reliability Normalized values, Indicates operating performance indicators The normalized value.

9. A smart operation and maintenance management system for DC charging piles, characterized in that, Includes the following modules: The environmental index construction module is used to statistically analyze the road conditions and DC charging pile failure rate of the DC charging pile locations within the managed area, and calculate the DC charging pile environmental index based on the road conditions and DC charging pile failure rate, including the road index and the failure index. The operational reliability index construction module is used to collect DC voltage and DC current signals during the charging process of DC charging piles, and use analog-to-digital conversion and EEMD algorithm to obtain the time series data corresponding to DC voltage and DC current signals; based on the time series data, the operational reliability index of DC charging piles is calculated, including the sudden change, intensity, fluctuation and trend index corresponding to DC voltage and DC current; The operational performance index construction module is used to collect power and power quality parameters of DC charging piles in real time. It uses a graph isomorphic neural network to extract feature vectors of power quality parameters and a convolutional neural network to extract feature vectors of power quality parameters. The feature vectors of power quality parameters and power quality parameters are updated through an attention mechanism to obtain a feature vector set. The feature vector set is used as the input of a pre-trained multilayer perceptron to obtain the metering error prediction value of DC charging piles in the area to be managed. Based on the metering error prediction value and the actual metering error value, the operational performance index of DC charging piles is constructed. The operation and maintenance management module is used to calculate the operation and maintenance management index of DC charging piles based on the environment, operational reliability and operational performance indicators of DC charging piles, and to formulate corresponding operation and maintenance management strategies based on the operation and maintenance management index.

10. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the method as described in any one of claims 1 to 8.

Images