Charger charging detection method, device, equipment and storage medium

Abstract

The present invention provides a charging detection method, device, equipment and storage medium for a charger, wherein the method comprises: obtaining startup and dynamic charging current and voltage data through a harmonic energy meter during the startup and charging process of a charging pile. The management system performs an initial harmonic characteristic analysis on the startup current and voltage to generate a reference harmonic characteristic data set, and performs a dynamic harmonic characteristic analysis on the dynamic current and voltage to generate a dynamic harmonic characteristic data set. Based on the reference, dynamic harmonic data and preset normal harmonic data, the working state of the charging pile is detected for abnormality, and a graded abnormal alarm signal is generated. Intelligent decisions are made based on the alarm signal, charging control instructions are issued, and the charging operation of the charging pile is dynamically adjusted. The present invention accurately detects abnormal states during the charging process by analyzing the harmonic characteristics of the charging current and voltage, thereby improving the accuracy of abnormal detection, making adjustments according to different abnormal situations, and improving the safety and reliability of the charging operation.

Description

Technical Field

[0001] The present invention relates to the field of charging technology, and in particular to a charging detection method, device, equipment and storage medium of a charger. Background Art

[0002] With the rapid development of the electric vehicle industry, the demand for charging piles, as the key infrastructure for charging electric vehicles, is increasing day by day. During the charging process, how to ensure the efficient operation, safety, reliability and fault detection capabilities of the charging piles has become the core issue to ensure the normal charging of electric vehicles. Traditional charging detection methods usually rely on simple current and voltage monitoring methods, which are difficult to accurately identify abnormal conditions in the charging process, such as current fluctuations, overloads, harmonic interference, etc., which may lead to charging failures or safety hazards. During the charging process, due to the influence of factors such as equipment and environment, current and voltage will produce varying degrees of harmonic interference. The presence of harmonics not only affects the working state of the charging equipment, but may also have a negative impact on the power grid. Therefore, how to accurately analyze and detect harmonics has become an important technical means to improve the safety and stability of charging equipment. However, the detection system of existing chargers usually lacks the ability to analyze complex harmonic characteristics and cannot provide intelligent decision support for abnormal situations, resulting in low charging management efficiency. Summary of the invention

[0003] The main purpose of the present invention is to solve the technical problem that the existing charger detection lacks the ability to analyze complex harmonic characteristics and cannot provide intelligent decision support for abnormal situations, resulting in low charging management efficiency;

[0004] A first aspect of the present invention provides a charging detection method for a charger, wherein the charger includes a single-channel single-plug charging pile, a harmonic electric energy meter, and a management system. The charging detection method for the charger includes:

[0005] When the charging operation of the single-channel single-plug charging pile is started and the charging operation is in progress, the starting charging current and starting charging voltage, dynamic charging current and dynamic charging voltage of the single-channel single-plug charging pile are obtained through the harmonic electric energy meter;

[0006] Performing initial harmonic characteristic analysis processing on the startup charging current and the startup charging voltage through the management system to obtain a reference harmonic characteristic data set, and performing dynamic harmonic characteristic analysis processing on the dynamic charging current and the dynamic charging voltage to obtain a dynamic harmonic characteristic data set;

[0007] The management system performs abnormal detection processing on the working state of the single-channel single-plug charging pile according to the reference harmonic feature data set, the dynamic harmonic feature data set and the preset normal harmonic data to obtain a graded abnormal alarm signal;

[0008] The management system performs intelligent decision-making processing on the charging operation of the single-channel single-plug charging pile according to the graded abnormal alarm signal, obtains a charging control instruction, and adjusts the charging operation of the single-channel single-plug charging pile according to the charging control instruction.

[0009] Optionally, in a first implementation of the first aspect of the present invention, the management system performs initial harmonic feature analysis processing on the startup charging current and the startup charging voltage to obtain a reference harmonic feature data set, and performs dynamic harmonic feature analysis processing on the dynamic charging current and the dynamic charging voltage to obtain a dynamic harmonic feature data set including:

[0010] The management system performs harmonic content calculation processing on the starting charging current and the starting charging voltage to obtain initial harmonic characteristic data, and extracts the main harmonic components of the initial harmonic characteristic data to obtain a reference harmonic characteristic data set;

[0011] Performing periodic harmonic measurement processing on the dynamic charging current and the dynamic charging voltage to obtain time series harmonic data;

[0012] Harmonic parameters in the time series harmonic data are extracted, and a change trend analysis process is performed on the harmonic parameters to obtain a dynamic harmonic feature data set.

[0013] Optionally, in a second implementation of the first aspect of the present invention, the extracting harmonic parameters from the time series harmonic data and performing a change trend analysis on the harmonic parameters to obtain a dynamic harmonic feature data set includes:

[0014] Performing sliding window segmentation processing on the time series harmonic data to obtain harmonic data subsets of multiple time windows, and extracting harmonic features in the harmonic data subsets to obtain a harmonic feature sequence;

[0015] Performing adjacent window difference calculation processing on the harmonic characteristic sequence to obtain harmonic parameter change rate data;

[0016] According to the harmonic parameter change rate data, the charging operation of the single-channel single-plug charging pile is divided into stages to obtain a charging stage identification result;

[0017] The charging stage identification result and the corresponding harmonic feature sequence are subjected to correlation analysis to obtain a dynamic harmonic feature data set.

[0018] Optionally, in a third implementation of the first aspect of the present invention, the management system performs abnormal detection processing on the working state of the single-channel single-plug charging pile according to the reference harmonic feature data set, the dynamic harmonic feature data set and the preset normal harmonic data to obtain a graded abnormal alarm signal, including:

[0019] The management system performs dynamic deviation calculation processing on the harmonic changes of the charging operation of the single-channel single-plug charging pile according to the dynamic harmonic characteristic data set and the preset normal harmonic data to obtain a harmonic deviation time series;

[0020] Judging whether there is an abnormality in the charging operation of the single-channel single-plug charging pile according to the harmonic deviation time series;

[0021] If so, the abnormal type of the working state of the single-channel single-plug charging pile is determined according to the harmonic deviation time series and the reference harmonic characteristic data set, and a graded abnormal alarm signal corresponding to the abnormal type is generated.

[0022] Optionally, in a fourth implementation of the first aspect of the present invention, the management system performs dynamic deviation calculation processing on the harmonic changes of the charging operation of the single-channel single-plug charging pile according to the dynamic harmonic feature data set and the preset normal harmonic data to obtain the harmonic deviation time series, including:

[0023] Performing time alignment processing on the dynamic harmonic feature data set and the preset normal harmonic data, and performing multi-scale deviation calculation on the dynamic harmonic feature data set and the preset normal harmonic data after time alignment to obtain a multi-scale harmonic deviation matrix;

[0024] Performing nonlinear feature extraction processing on the multi-scale harmonic deviation matrix to obtain a harmonic anomaly feature vector;

[0025] The charging process is dynamically clustered according to the harmonic anomaly feature vector to obtain a harmonic state evolution sequence, and the harmonic state evolution sequence is subjected to time series prediction processing to obtain a harmonic deviation time series.

[0026] Optionally, in a fifth implementation of the first aspect of the present invention, determining the abnormal type of the working state of the single-channel single-plug charging pile according to the harmonic deviation time series and the reference harmonic feature data set, and generating a graded abnormal alarm signal corresponding to the abnormal type includes:

[0027] Performing time-frequency analysis on the harmonic deviation time series to obtain a harmonic anomaly characteristic spectrum;

[0028] Matching and identifying the abnormal pattern of the single-channel single-plug charging pile according to the harmonic abnormality feature map and the reference harmonic feature data set to obtain an abnormality type determination result, wherein the abnormality type determination result includes the abnormality type and the probability distribution of the abnormality type;

[0029] Performing fuzzy reasoning processing on the abnormality type determination result to obtain a comprehensive abnormality evaluation index;

[0030] The abnormal state of the single-channel single-plug charging pile is graded and processed according to the comprehensive abnormality evaluation index to obtain a graded abnormality alarm signal.

[0031] Optionally, in a sixth implementation of the first aspect of the present invention, the intelligent decision-making process is performed on the charging operation of the single-channel single-plug charging pile according to the graded abnormal alarm signal by the management system to obtain a charging control instruction, and the charging operation of the single-channel single-plug charging pile is adjusted according to the charging control instruction, including:

[0032] The management system performs multimodal feature extraction processing on the charging process data of the charging operation of the single-channel single-plug charging pile according to the graded abnormal alarm signal to obtain a high-dimensional abnormal feature tensor;

[0033] Performing topological mapping processing on the charging state space according to the high-dimensional abnormal feature tensor to obtain an abnormal state topological map, and performing graph neural network analysis processing on the abnormal state topological map to obtain an abnormal propagation prediction model;

[0034] A meta-heuristic algorithm is used to solve the multi-objective charging optimization problem according to the abnormal propagation prediction model to obtain a Pareto optimal solution set, and a fuzzy decision-making process is performed on the Pareto optimal solution set to obtain a fuzzy charging control strategy;

[0035] According to the fuzzy charging control strategy, an adaptive model predictive control process is performed on the charging operation of the single-channel single-plug charging pile to obtain a charging control instruction, and the charging operation of the single-channel single-plug charging pile is adjusted according to the charging control instruction.

[0036] A second aspect of the present invention provides a charging detection device for a charger, wherein the charger comprises a single-channel single-plug charging pile, a harmonic energy meter, and a management system, and the charging detection device for the charger comprises:

[0037] A data detection module, used for obtaining the starting charging current and starting charging voltage, dynamic charging current and dynamic charging voltage of the single-channel single-plug charging pile through the harmonic electric energy meter when the charging operation of the single-channel single-plug charging pile is started and when the charging operation is in progress;

[0038] A harmonic analysis module, configured to perform initial harmonic characteristic analysis processing on the startup charging current and the startup charging voltage through the management system to obtain a reference harmonic characteristic data set, and perform dynamic harmonic characteristic analysis processing on the dynamic charging current and the dynamic charging voltage to obtain a dynamic harmonic characteristic data set;

[0039] an abnormality detection module, used to perform abnormality detection processing on the working state of the single-channel single-plug charging pile according to the reference harmonic characteristic data set, the dynamic harmonic characteristic data set and the preset normal harmonic data through the management system to obtain a graded abnormality alarm signal;

[0040] The intelligent decision-making module is used to perform intelligent decision-making processing on the charging operation of the single-channel single-plug charging pile according to the graded abnormal alarm signal through the management system, obtain a charging control instruction, and adjust the charging operation of the single-channel single-plug charging pile according to the charging control instruction.

[0041] A third aspect of the present invention provides a charging detection device for a charger, comprising: a memory and at least one processor, wherein instructions are stored in the memory, and the memory and the at least one processor are interconnected via a line; the at least one processor calls the instructions in the memory so that the charging detection device of the charger performs the steps of the above-mentioned charging detection method of the charger.

[0042] A fourth aspect of the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores instructions, which, when executed on a computer, enable the computer to execute the steps of the above-mentioned charger charging detection method.

[0043] The charging detection method, device, equipment and storage medium of the above-mentioned charger obtains the startup and dynamic charging current and voltage data through the harmonic energy meter during the startup and charging process of the charging pile. The management system performs an initial harmonic characteristic analysis on the startup current and voltage to generate a reference harmonic characteristic data set, and performs a dynamic harmonic characteristic analysis on the dynamic current and voltage to generate a dynamic harmonic characteristic data set. Based on the reference, dynamic harmonic data and preset normal harmonic data, the working state of the charging pile is detected for abnormalities, and a graded abnormal alarm signal is generated. Intelligent decisions are made based on the alarm signal, charging control instructions are issued, and the charging operation of the charging pile is dynamically adjusted. The present invention accurately detects abnormal conditions during the charging process by analyzing the harmonic characteristics of the charging current and voltage, thereby improving the accuracy of abnormal detection, making adjustments according to different abnormal conditions, and improving the safety and reliability of the charging operation.

[0044] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention are realized and obtained by the structures particularly pointed out in the description, claims and drawings.

[0045] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, preferred embodiments are given below and described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 It is a schematic diagram of a first embodiment of a charging detection method of a charger in an embodiment of the present invention;

[0047] Figure 2 It is a schematic diagram of an embodiment of a charging detection device of a charger in an embodiment of the present invention;

[0048] Figure 3 The figure is a schematic diagram of an embodiment of a charging detection device of a charger in an embodiment of the present invention. DETAILED DESCRIPTION

[0049] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the described embodiments are 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.

[0050] The terms "including" and "having" and any variations thereof mentioned in the embodiments of the present invention are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device end including a series of steps or units is not limited to the listed steps or units, but may optionally include other steps or units that are not listed, or may optionally include other steps or units that are inherent to these processes, methods, products or device ends.

[0051] To facilitate understanding of this embodiment, a charging detection method for a charger disclosed in an embodiment of the present invention is first described in detail. Figure 1 As shown, the method comprises the following steps:

[0052] 101. When the charging operation of the single-channel single-plug charging pile is started and the charging operation is in progress, the starting charging current and starting charging voltage, dynamic charging current and dynamic charging voltage of the single-channel single-plug charging pile are obtained through the harmonic electric energy meter;

[0053] In one embodiment of the present invention, it is first necessary to install an electric energy meter with a harmonic measurement function on a single-channel single-plug charging pile. This electric energy meter should be able to measure the harmonic content of voltage and current in real time, including total harmonic distortion (THD) and the amplitude of each harmonic. According to the research results, the focus is on harmonics 3, 5, and 7, because these harmonics contribute the most to the charging process of electric vehicles. The sampling frequency of the electric energy meter is set to no less than 10 kHz to ensure that harmonics within 50 can be accurately captured. At the same time, the electric energy meter is equipped with a communication function that can transmit the measurement data to the charging pile management system in real time. When the electric vehicle is plugged into the charging pile and starts charging, the harmonic electric energy meter immediately starts the measurement program. Within the first 3 minutes of charging, the system performs an initial harmonic feature analysis. The selection of this time period is based on the observation that the charging power reaches a stable state about 3 minutes after charging starts. In this initial stage, the system records the THD values ​​of voltage and current, the amplitudes of harmonics 3, 5, and 7 (relative to the percentage of the fundamental wave), and the charging power curve. These initial data constitute the harmonic characteristics of the starting charging current and the starting charging voltage. Throughout the charging process, the harmonic energy meter continuously monitors the harmonic content of the charging current and voltage. The system records complete harmonic data every 1 minute, including THD and the amplitude of at least 25 harmonics. The selection of this sampling interval is based on practical experience, which can capture the trend of harmonic changes without generating too much data. For each sampling point, the system calculates the THD value of voltage and current, the amplitude of the main harmonics (3rd, 5th, and 7th), the amplitude of even harmonics (especially 2nd and 4th), and the DC component. These data constitute the harmonic characteristics of dynamic charging current and dynamic charging voltage. In order to improve the accuracy and reliability of the measurement, the system uses sliding window technology to process harmonic data. Each sliding window contains 10 seconds of data, and the windows overlap by 50%. This method can smooth short-term fluctuations while retaining the trend information of harmonic changes. Fast Fourier transform (FFT) is performed on the data in each sliding window to obtain a more accurate harmonic spectrum. The system also takes into account that different stages of the charging process may present different harmonic characteristics. Therefore, the parameters of harmonic measurement are dynamically adjusted according to the changes in charging power. For example, at the end of charging (when the battery charge exceeds 80%), the system increases the sampling frequency to capture the faster harmonic changes that may occur at this stage. In order to deal with possible measurement errors and outliers, the system implements a data verification mechanism. After each measurement, the result is compared with the average of the previous measurements. If a significant deviation is found, the system will immediately perform additional measurements to confirm the reliability of the data. At the same time, the system continuously monitors the working status of the energy meter to ensure that it is always in normal working condition. Finally, all measured harmonic data, including dynamic data during the startup phase and the entire charging process, are transmitted to the charging pile management system in real time through a secure communication protocol.The management system associates this data with timestamps to form a complete charging harmonic characteristic time series, providing basic data support for subsequent anomaly detection and intelligent decision-making.

[0054] 102. Performing initial harmonic characteristic analysis processing on the startup charging current and the startup charging voltage through the management system to obtain a reference harmonic characteristic data set, and performing dynamic harmonic characteristic analysis processing on the dynamic charging current and the dynamic charging voltage to obtain a dynamic harmonic characteristic data set;

[0055] In one embodiment of the present invention, the management system performs initial harmonic characteristic analysis on the starting charging current and the starting charging voltage to obtain a reference harmonic characteristic data set, and performs dynamic harmonic characteristic analysis on the dynamic charging current and the dynamic charging voltage to obtain a dynamic harmonic characteristic data set, including: performing harmonic content calculation on the starting charging current and the starting charging voltage to obtain initial harmonic characteristic data, and extracting the main harmonic components of the initial harmonic characteristic data to obtain a reference harmonic characteristic data set; performing periodic harmonic measurement on the dynamic charging current and the dynamic charging voltage to obtain time series harmonic data; extracting harmonic parameters in the time series harmonic data, and performing change trend analysis on the harmonic parameters to obtain a dynamic harmonic characteristic data set.

[0056] Specifically, the management system first calculates the harmonic content of the startup charging current and the startup charging voltage. This process begins when the electric vehicle is connected to the charging pile and starts charging, and the system immediately starts data collection. The fast Fourier transform (FFT) algorithm is used to analyze the collected current and voltage waveforms to calculate the total harmonic distortion (THD) and the amplitude of each harmonic. During the calculation process, the system pays special attention to the 3rd, 5th, and 7th harmonics because these harmonics usually contribute the most during the charging process of electric vehicles. At the same time, the system also records the amplitude of even harmonics (especially 2nd and 4th), as well as the DC component, which are crucial for a comprehensive assessment of the harmonic characteristics of the charging process. After the calculation is completed, the system further processes the initial harmonic characteristic data obtained to extract the main harmonic components. This step first sorts the amplitudes of each harmonic and selects the first few harmonics with the largest amplitude as the main harmonic components. Usually, these main harmonics include the fundamental (50Hz or 60Hz) and the 3rd, 5th, and 7th harmonics. The system also calculates the percentage of these main harmonics relative to the fundamental wave to more intuitively reflect the relative strength of the harmonics. In addition, the system also records the THD value as an indicator of the overall harmonic level. These extracted main harmonic components and THD values ​​together constitute the benchmark harmonic feature data set, providing a reference benchmark for subsequent dynamic analysis. Next, the system performs periodic harmonic measurement processing on the dynamic charging current and dynamic charging voltage. This process continues throughout the charging cycle, and the system performs a complete harmonic measurement at fixed time intervals (for example, every minute). Each measurement uses the same FFT algorithm as the initial harmonic analysis to calculate the THD and the amplitude of each harmonic. This periodic measurement ensures that the system can capture the dynamic changes of harmonic characteristics during the charging process. In order to improve the reliability and continuity of the data, the system uses a sliding window technology, each measurement window contains 10 seconds of data, and there is a 50% overlap between adjacent windows. This method can both smooth short-term fluctuations and retain the trend information of harmonic changes. Based on periodic measurements, the system extracts key harmonic parameters from time series harmonic data and performs trend analysis on these parameters. The extracted parameters include THD, the amplitude of the main harmonics (3rd, 5th, and 7th), the amplitude of the even harmonics, and the DC component. The system calculates the rate of change of these parameters between consecutive measurement cycles and identifies significant change trends and abnormal fluctuations. At the same time, the system also analyzes the correlation between harmonic parameters, such as whether the proportional relationship between the amplitudes of harmonics of different orders remains stable. These analysis results together constitute a dynamic harmonic feature data set, which fully reflects the dynamic changes of harmonic characteristics during the charging process. Throughout the process, the system continuously monitors the working status of the energy meter to ensure the accuracy of the measurement data. If an abnormal value is found in a certain measurement, the system will immediately perform additional measurements to verify the reliability of the data.At the same time, the system also takes into account that different stages of the charging process may present different harmonic characteristics, so the sampling frequency is increased at the end of charging (for example, when the battery charge exceeds 80%) to capture the faster harmonic changes that may occur at this stage. Finally, the system transmits all measured and analyzed data, including the baseline harmonic characteristic data set and the dynamic harmonic characteristic data set, to the central management system in real time through a secure communication protocol. These data are associated with timestamps to form a complete charging harmonic characteristic time series, providing comprehensive and detailed data support for subsequent anomaly detection and intelligent decision-making. This comprehensive harmonic analysis method can not only accurately capture the harmonic changes during the charging process, but also provide important reference information for the safe operation of charging piles and the stability of the power grid.

[0057] Furthermore, the extraction of harmonic parameters from the time series harmonic data and the performance of a change trend analysis on the harmonic parameters to obtain a dynamic harmonic feature data set includes: performing a sliding window segmentation process on the time series harmonic data to obtain a harmonic data subset of multiple time windows, and extracting harmonic features from the harmonic data subset to obtain a harmonic feature sequence; performing adjacent window difference calculation on the harmonic feature sequence to obtain harmonic parameter change rate data; performing a stage division process on the charging operation of the single-channel single-plug charging pile according to the harmonic parameter change rate data to obtain a charging stage identification result; and performing a correlation analysis process on the charging stage identification result and the corresponding harmonic feature sequence to obtain a dynamic harmonic feature data set.

[0058] Specifically, firstly, the time series harmonic data is processed by sliding window segmentation. This step uses a fixed-size time window, such as 10 seconds, to slide along the time series in sequence. The step size of each slide is set to 50% of the window size, that is, 5 seconds, so that there is a 50% overlap between adjacent windows. This sliding window technology can smooth short-term fluctuations while retaining the continuity information of harmonic changes. Through this process, harmonic data subsets of multiple time windows are obtained, each subset containing 10 seconds of harmonic data. Next, harmonic features are extracted from each harmonic data subset. The extracted features include total harmonic distortion (THD), the amplitude of the main harmonics (especially the 3rd, 5th, and 7th harmonics), the amplitude of the even harmonics (such as the 2nd and 4th harmonics), and the DC component. For each time window, the mean and standard deviation of these features are calculated. In addition, the percentage of the amplitude of each harmonic relative to the fundamental is calculated to more intuitively reflect the relative strength of the harmonics. These extracted features constitute a harmonic feature sequence, and each time window corresponds to a set of feature values. Then, the harmonic feature sequence is processed by adjacent window difference calculation. This step calculates the change amount of each harmonic feature between two adjacent time windows. Specifically, for each feature (such as THD, amplitude of each harmonic, etc.), the difference between the current window and the previous window is calculated, and the difference is divided by the time interval to obtain the rate of change. This processing obtains the harmonic parameter change rate data, which reflects the dynamic change speed and direction of the harmonic characteristics during the charging process. Based on the obtained harmonic parameter change rate data, the system performs stage division processing on the charging operation of a single-channel single-plug charging pile. This step identifies the different stages of the charging process by analyzing the pattern of the change rate data. For example, the initial charging may be manifested as a rapid change of harmonic parameters, the middle period may be relatively stable, and the later period (such as when the battery charge exceeds 80%) may fluctuate greatly again. The system uses a clustering algorithm (such as K-means or hierarchical clustering) to analyze the change rate data and divide the entire charging process into several stages with obvious characteristics, such as the startup stage, the stable stage, and the end stage. Each stage is assigned a unique identifier to form a charging stage identification result. Finally, the system performs correlation analysis on the charging stage identification results and the corresponding harmonic feature sequences. This step matches and analyzes each charging stage with its corresponding harmonic feature. Specifically, the system calculates the statistical indicators of each harmonic feature in each charging stage, such as the mean, standard deviation, maximum and minimum values. At the same time, the changing trends of each harmonic parameter between different charging stages are analyzed, such as whether certain harmonics show a clear increase or decrease trend in a specific stage. In addition, the system also analyzes the correlation between harmonic parameters, such as whether the proportional relationship between the amplitudes of different harmonics remains stable in each charging stage. The results of this correlation analysis constitute the final dynamic harmonic feature data set.This dynamic harmonic feature dataset comprehensively reflects the dynamic changes of harmonic characteristics during the charging process, including the typical harmonic characteristics and change patterns of each charging stage. It not only includes the specific harmonic parameter values ​​at each time point, but also includes the change trend of these parameters over time and the characteristic performance at different charging stages.

[0059] 103. The management system performs abnormal detection processing on the working state of the single-channel single-plug charging pile according to the reference harmonic characteristic data set, the dynamic harmonic characteristic data set and the preset normal harmonic data to obtain a graded abnormal alarm signal;

[0060] In one embodiment of the present invention, the management system performs abnormal detection processing on the working state of the single-channel single-plug charging pile according to the baseline harmonic characteristic data set, the dynamic harmonic characteristic data set and the preset normal harmonic data to obtain a graded abnormal alarm signal, including: the management system performs dynamic deviation calculation processing on the harmonic changes of the charging operation of the single-channel single-plug charging pile according to the dynamic harmonic characteristic data set and the preset normal harmonic data to obtain a harmonic deviation time series; judges whether there is an abnormality in the charging operation of the single-channel single-plug charging pile according to the harmonic deviation time series; if so, determines the abnormal type of the working state of the single-channel single-plug charging pile according to the harmonic deviation time series and the baseline harmonic characteristic data set, and generates a graded abnormal alarm signal corresponding to the abnormal type.

[0061] Specifically, the management system first performs dynamic deviation calculation processing on the harmonic changes of the charging operation of the single-channel single-plug charging pile according to the dynamic harmonic feature data set and the preset normal harmonic data to obtain the harmonic deviation time series. This step involves comparing the actually measured harmonic data with the preset normal range and calculating the deviation value at each time point. Next, the system determines whether the charging operation of the single-channel single-plug charging pile is abnormal based on the harmonic deviation time series. This judgment process involves multiple steps and technologies. First, the system performs statistical analysis on the harmonic deviation time series and calculates the average deviation value and standard deviation of each harmonic parameter (such as total harmonic distortion THD, main harmonic amplitude, etc.). These statistical indicators provide the basis for subsequent abnormal judgment. Then, the system uses a multi-threshold judgment method to identify potential abnormalities. For each harmonic parameter, multiple threshold levels are set, such as slight deviation, moderate deviation, and severe deviation. The setting of these thresholds is based on the preset normal harmonic data and historical operating experience. The system compares the actual deviation value with these thresholds, and if the deviation of a parameter continues to exceed a specific threshold, it is marked as a potential abnormality. In order to improve the accuracy of abnormality detection, the system also considers time factors. Short-term deviations may be caused by instantaneous fluctuations and may not necessarily represent real anomalies. Therefore, the system introduces time window analysis, and only when the deviation persists within a certain time window is it considered an actual anomaly. The length of this time window can be dynamically adjusted according to different harmonic parameters and the degree of deviation. In addition, the system also uses trend analysis to identify progressive anomalies. By calculating the rate of change and acceleration of harmonic deviations, the system is able to capture those anomalies that have not yet exceeded the threshold but show a clear upward trend. This approach helps to detect potential problems early and achieve preventive maintenance. Considering the possible mutual influence between different harmonic parameters, the system also uses correlation analysis. For example, the anomaly of some harmonics may cause a chain reaction of other harmonics. The system identifies this complex abnormal pattern by analyzing the correlation between the deviations of different harmonic parameters. In order to deal with the noise and outliers that may exist in the data, the system uses filtering technology. Through methods such as median filtering or Kalman filtering, short-term fluctuations and isolated anomalies are removed, making anomaly detection more stable and reliable. In the judgment process, the system also considers the impact of the charging stage. Different charging stages (such as the startup stage, the stable stage, and the end stage) may have different harmonic characteristics. Based on the previously obtained charging stage identification results, the system adopts different judgment criteria for different stages to adapt to the dynamic characteristics of the charging process. Finally, the system comprehensively considers all the above factors and uses machine learning algorithms (such as support vector machines or decision trees) to make the final abnormal judgment. This algorithm has been trained to comprehensively consider factors such as the deviation, time persistence, change trend and correlation of multiple harmonic parameters, so as to make more accurate abnormal judgments.

[0062] Furthermore, the management system performs dynamic deviation calculation processing on the harmonic changes of the charging operation of the single-channel single-plug charging pile according to the dynamic harmonic feature data set and the preset normal harmonic data to obtain a harmonic deviation time series, including: time alignment processing on the dynamic harmonic feature data set and the preset normal harmonic data, and multi-scale deviation calculation on the dynamic harmonic feature data set after time alignment and the preset normal harmonic data to obtain a multi-scale harmonic deviation matrix; nonlinear feature extraction processing is performed on the multi-scale harmonic deviation matrix to obtain a harmonic anomaly feature vector; dynamic clustering processing is performed on the charging process according to the harmonic anomaly feature vector to obtain a harmonic state evolution sequence, and time series prediction processing is performed on the harmonic state evolution sequence to obtain a harmonic deviation time series.

[0063] Specifically, firstly, the dynamic harmonic feature data set and the preset normal harmonic data need to be time-aligned. The time alignment process uses the dynamic time warping (DTW) algorithm, which can handle the nonlinear time distortion that may exist in the time series data, so that the two sets of data can be optimally matched on the time axis. After the time alignment is completed, the system-aligned dynamic harmonic feature data set and the preset normal harmonic data are multi-scale deviation calculated. This step uses wavelet transform technology to decompose the harmonic data into components of different time scales. Usually, discrete wavelet transform (DWT) or continuous wavelet transform (CWT) is selected, and the appropriate wavelet basis function, such as Daubechies wavelet or Morlet wavelet, is selected according to the characteristics of the data. For each time scale, the deviation between the actual harmonic data and the preset normal range is calculated. This multi-scale analysis can capture harmonic anomalies on different time scales, and can be identified from rapid fluctuations to long-term trends. The calculation results form a multi-scale harmonic deviation matrix, in which each row represents a time scale and each column represents a time point. Next, the multi-scale harmonic deviation matrix is ​​subjected to nonlinear feature extraction. This step aims to extract low-dimensional features that can effectively characterize harmonic anomalies from the high-dimensional deviation matrix. The processing process uses nonlinear dimensionality reduction techniques, such as kernel principal component analysis (Kernel PCA). These methods can retain the nonlinear structure in the data and better capture complex harmonic anomaly patterns. The result of feature extraction is a harmonic anomaly feature vector, which contains key information that can summarize the harmonic anomaly state. Then, the system dynamically clusters the charging process based on the harmonic anomaly feature vector. This step uses online clustering algorithms, such as online K-means or streaming DBSCAN (density clustering). These algorithms can process incoming data in real time and assign each time point in the charging process to different clusters according to its harmonic anomaly characteristics. The clustering results reflect the dynamic changes of the harmonic state during the charging process, forming a harmonic state evolution sequence. This sequence describes how the harmonic state transfers from one cluster to another during the charging process, reflecting the dynamic evolution of the harmonic anomaly. Finally, the system performs time series prediction processing on the harmonic state evolution sequence to obtain the harmonic deviation time series. This step uses time series prediction models, such as long short-term memory networks (LSTM) or gated recurrent units (GRU). These deep learning models can capture long-term dependencies in time series data and are suitable for processing complex time series data such as harmonic states. The prediction model predicts the changes in harmonic states in the future based on the historical harmonic state evolution sequence. The prediction results are compared with the actual observed harmonic states to generate the final harmonic deviation time series. This harmonic deviation time series not only contains the harmonic anomaly information at the current moment, but also contains the prediction of future harmonic changes.It combines the results of multi-scale analysis, nonlinear feature extraction, dynamic clustering and time series prediction to provide a comprehensive and detailed description of harmonic anomalies. This method can capture complex harmonic anomaly patterns, including transient anomalies, gradual anomalies and periodic anomalies. At the same time, by predicting future harmonic changes, the system can identify potential anomalies in advance and achieve preventive maintenance and control.

[0064] Furthermore, determining the abnormal type of the working state of the single-channel single-plug charging pile according to the harmonic deviation time series and the benchmark harmonic characteristic data set, and generating a graded abnormal alarm signal corresponding to the abnormal type includes: performing time-frequency analysis processing on the harmonic deviation time series to obtain a harmonic abnormal characteristic spectrum; matching and identifying the abnormal pattern of the single-channel single-plug charging pile according to the harmonic abnormal characteristic spectrum and the benchmark harmonic characteristic data set to obtain an abnormal type determination result, wherein the abnormal type determination result includes the abnormal type and the probability distribution of the abnormal type; performing fuzzy reasoning processing on the abnormal type determination result to obtain a comprehensive abnormal evaluation index; and performing graded classification processing on the abnormal state of the single-channel single-plug charging pile according to the comprehensive abnormal evaluation index to obtain a graded abnormal alarm signal.

[0065] Specifically, firstly, the harmonic deviation time series is subjected to time-frequency analysis to obtain the harmonic anomaly feature map. This step uses short-time Fourier transform (STFT) or wavelet transform (WT) technology to convert the time domain signal into a time-frequency domain representation. STFT is suitable for analyzing the local frequency characteristics of the signal, while WT can provide multi-resolution analysis and is suitable for processing non-stationary signals. During the processing, the system selects a suitable window function (such as Hanning window or Gaussian window) and window size to balance the time and frequency resolution. For each time window, the spectrum is calculated and the results are combined into a time-frequency map. This harmonic anomaly feature map intuitively shows the distribution characteristics of harmonic anomalies in the time and frequency dimensions. Next, the system matches and identifies the abnormal pattern of the single-channel single-plug charging pile according to the harmonic anomaly feature map and the benchmark harmonic feature data set to obtain the abnormal type determination result. This step uses pattern recognition technology, such as support vector machine (SVM) or convolutional neural network (CNN). The system first uses the benchmark harmonic feature data set to train the model and establish feature templates of different anomaly types. Then, the obtained harmonic anomaly feature map is input into the trained model for classification. The model output includes the anomaly type and its probability distribution, reflecting the possibility of different anomaly types. This probabilistic output takes into account the uncertainty of anomaly type judgment in actual situations and provides richer information for subsequent processing. Then, the system performs fuzzy reasoning on the anomaly type judgment results to obtain a comprehensive anomaly evaluation index. This step uses a fuzzy logic system to convert precise probability values ​​into fuzzy sets. The fuzzy reasoning system consists of three main parts: fuzzification, inference rules, and defuzzification. In the fuzzification stage, the system maps the probability values ​​of the anomaly types to predefined fuzzy sets (such as "low probability", "medium probability", and "high probability"). The inference rules are based on expert knowledge and historical data, such as "If the probability of type A is high and the probability of type B is medium, the comprehensive anomaly degree is severe." The defuzzification process converts the results of fuzzy reasoning into specific numerical values ​​as comprehensive anomaly evaluation indicators. This fuzzy reasoning method can better handle the uncertainty and ambiguity in anomaly judgment and provide evaluation results that are more in line with human intuition. Finally, the system classifies the abnormal state of the single-channel single-plug charging pile according to the comprehensive anomaly evaluation index to obtain a graded abnormal alarm signal. This step uses a hierarchical clustering algorithm or a rule-based classification method. The system predefines multiple abnormality levels, such as minor, moderate, severe, and critical. Then, according to the value of the comprehensive abnormality assessment index, the current abnormal state is divided into the corresponding level. In the classification process, the system considers factors such as the severity, duration, and potential impact of the abnormality. For example, even if the current abnormality assessment index is not high, it may be classified as a higher level if it lasts for a long time or shows a rapid deterioration trend. For each abnormality level, the system generates a corresponding alarm signal, including the abnormality type, severity, possible cause, and recommended measures.The whole process forms a complete anomaly detection and alarm generation chain. Starting from time-frequency analysis, through pattern recognition, fuzzy reasoning to the final classification, the system gradually transforms complex harmonic data into clear and actionable alarm information. This method not only takes into account the complexity and diversity of harmonic anomalies, but also introduces uncertainty processing and human expert knowledge, making the anomaly detection results more reliable and practical. At the same time, the hierarchical alarm mechanism can help operation and maintenance personnel quickly understand the severity and urgency of the anomaly and take corresponding measures in a targeted manner.

[0066] 104. The management system performs intelligent decision-making on the charging operation of the single-channel single-plug charging pile according to the graded abnormal alarm signal to obtain a charging control instruction, and adjusts the charging operation of the single-channel single-plug charging pile according to the charging control instruction.

[0067] In one embodiment of the present invention, the intelligent decision-making processing of the charging operation of the single-channel single-plug charging pile by the management system according to the graded abnormal alarm signal to obtain a charging control instruction, and adjusting the charging operation of the single-channel single-plug charging pile according to the charging control instruction includes: performing multimodal feature extraction processing on the charging process data of the charging operation of the single-channel single-plug charging pile according to the graded abnormal alarm signal by the management system to obtain a high-dimensional abnormal feature tensor; performing topological mapping processing on the charging state space according to the high-dimensional abnormal feature tensor to obtain an abnormal state topological map, and performing graph neural network analysis processing on the abnormal state topological map to obtain an abnormal propagation prediction model; performing meta-heuristic algorithm solution processing on the multi-objective charging optimization problem according to the abnormal propagation prediction model to obtain a Pareto optimal solution set, and performing fuzzy decision processing on the Pareto optimal solution set to obtain a fuzzy charging control strategy; performing adaptive model predictive control processing on the charging operation of the single-channel single-plug charging pile according to the fuzzy charging control strategy to obtain a charging control instruction, and adjusting the charging operation of the single-channel single-plug charging pile according to the charging control instruction.

[0068] Specifically, the management system first extracts multimodal features from the charging process data of the charging operation of a single-channel single-plug charging pile according to the graded abnormal alarm signal to obtain a high-dimensional abnormal feature tensor. This step uses deep learning techniques such as convolutional neural networks (CNN) and long short-term memory networks (LSTM) to process data in the time domain, frequency domain, and energy domain, respectively. CNN is used to extract spatial features, such as local patterns of harmonic spectra, while LSTM captures long-term dependencies of time series. In addition, the system also integrates environmental information, such as temperature and humidity data around the charging pile. These different modal features are fused into a unified high-dimensional abnormal feature tensor through tensor decomposition and reconstruction methods, such as Tucker decomposition or CANDECOMP / PARAFAC (CP) decomposition. This tensor comprehensively characterizes the abnormal state during the charging process and contains comprehensive information on time, frequency, and environmental factors. Next, the system performs topological mapping on the charging state space based on the high-dimensional abnormal feature tensor to obtain an abnormal state topology map. This step uses manifold learning algorithms, such as t-SNE (t-distributed stochastic neighbor embedding) or UMAP (Uniform Manifold Approximation and Projection). These algorithms can map high-dimensional data to low-dimensional space while preserving local relationships between data points. The mapping result forms a visual topological map, where each point represents a charge state and the distance between points reflects the similarity of the state. This topological map intuitively shows the relationship and evolution path between different abnormal states. Then, the system performs graph neural network analysis on the abnormal state topological map to obtain an abnormal propagation prediction model. This step uses graph convolutional networks (GCN) or graph attention networks (GAT). These models take the topological map as input and learn the mutual influence between nodes (abnormal states). Through multi-layer graph convolution or attention mechanisms, the model can capture complex nonlinear relationships and long-distance dependencies. During the training process, the abnormal state evolution sequence observed in the historical data is used as a label. After training, the model can predict the probability and degree of other states being affected under a given abnormal state, thereby simulating the propagation process of the abnormality. Next, the system uses a metaheuristic algorithm to solve the multi-objective charging optimization problem based on the abnormal propagation prediction model and obtains the Pareto optimal solution set. This step considers multiple objectives, such as minimizing charging time, maximizing battery life, minimizing harmonic interference, etc. The system uses multi-objective optimization algorithms such as NSGA-II (Non-dominated Sorting Genetic Algorithm II) or MOEA / D (Multi-objective Evolutionary Algorithm Based on Decomposition). These algorithms generate a series of non-dominated solutions, namely the Pareto optimal solution set, by simulating the evolutionary process. Each solution represents a possible charging strategy that strikes a balance between different objectives.Then, the system performs fuzzy decision processing on the Pareto optimal solution set to obtain a fuzzy charging control strategy. This step uses the fuzzy analytic hierarchy process (Fuzzy AHP) or fuzzy TOPSIS method. First, based on expert knowledge and historical experience, a fuzzy decision matrix is ​​established to define the relative importance of different objectives. Then, each solution in the Pareto optimal solution set is mapped to a fuzzy set, such as "high efficiency", "low risk", etc. Through fuzzy inference rules, the system evaluates the comprehensive performance of each solution and finally selects the best charging control strategy. This fuzzy decision method can handle the uncertainty and ambiguity in the decision-making process and obtain a control strategy that is more in line with the actual situation. Finally, the system performs adaptive model predictive control processing on the charging operation of a single-channel single-plug charging pile according to the fuzzy charging control strategy to obtain charging control instructions. This step uses nonlinear model predictive control (NMPC) technology. The system first establishes a dynamic model of the charging process, including factors such as battery status, charging power, and harmonic generation. Then, in each control cycle, the NMPC algorithm predicts the system behavior in the future based on the current state and the fuzzy charging control strategy, and optimizes the sequence of control actions. The optimization process takes into account various constraints, such as maximum charging power, harmonic restrictions, etc. The system selects the first action in the optimization sequence as the current charging control instruction, including specific parameters such as charging current and voltage adjustment. The system adjusts the charging operation of the single-channel single-plug charging pile according to the generated charging control instruction. This includes real-time adjustment of charging current and voltage, and pausing or resuming the charging process when necessary. At the same time, the system continuously monitors the feedback of the charging process, including actual charging parameters, battery status, and generated harmonics. These feedback data are used to update the system model and optimize the control strategy to form a closed-loop control system.

[0069] In this embodiment, during the startup and charging process of the charging pile, the startup and dynamic charging current and voltage data are obtained through the harmonic energy meter. The management system performs an initial harmonic characteristic analysis on the startup current and voltage to generate a reference harmonic characteristic data set, and performs a dynamic harmonic characteristic analysis on the dynamic current and voltage to generate a dynamic harmonic characteristic data set. Based on the reference, dynamic harmonic data and preset normal harmonic data, the working state of the charging pile is detected for abnormalities, and a graded abnormal alarm signal is generated. Intelligent decisions are made based on the alarm signal, charging control instructions are issued, and the charging operation of the charging pile is dynamically adjusted. The present invention accurately detects abnormal conditions during the charging process by analyzing the harmonic characteristics of the charging current and voltage, thereby improving the accuracy of abnormal detection, making adjustments according to different abnormal conditions, and improving the safety and reliability of the charging operation.

[0070] The above describes the charging detection method of the charger in the embodiment of the present invention. The following describes the charging detection device of the charger in the embodiment of the present invention. The charger includes a single-channel single-plug charging pile, a harmonic energy meter, and a management system. Figure 2 In one embodiment of the present invention, a charging detection device of a charger includes:

[0071] The data detection module 201 is used to obtain the starting charging current and starting charging voltage, dynamic charging current and dynamic charging voltage of the single-channel single-plug charging pile through the harmonic electric energy meter when the charging operation of the single-channel single-plug charging pile is started and when the charging operation is in progress;

[0072] The harmonic analysis module 202 is used to perform initial harmonic characteristic analysis processing on the startup charging current and the startup charging voltage through the management system to obtain a reference harmonic characteristic data set, and perform dynamic harmonic characteristic analysis processing on the dynamic charging current and the dynamic charging voltage to obtain a dynamic harmonic characteristic data set;

[0073] The abnormality detection module 203 is used to perform abnormality detection processing on the working state of the single-channel single-plug charging pile according to the reference harmonic characteristic data set, the dynamic harmonic characteristic data set and the preset normal harmonic data through the management system to obtain a graded abnormality alarm signal;

[0074] The intelligent decision module 204 is used to perform intelligent decision processing on the charging operation of the single-channel single-plug charging pile according to the graded abnormal alarm signal through the management system, obtain the charging control instruction, and adjust the charging operation of the single-channel single-plug charging pile according to the charging control instruction.

[0075] In an embodiment of the present invention, the charging detection device of the charger runs the charging detection method of the charger, and the charging detection device of the charger obtains the startup and dynamic charging current and voltage data through the harmonic energy meter during the startup and charging process of the charging pile. The management system performs an initial harmonic characteristic analysis on the startup current and voltage to generate a reference harmonic characteristic data set, and performs a dynamic harmonic characteristic analysis on the dynamic current and voltage to generate a dynamic harmonic characteristic data set. Based on the reference, dynamic harmonic data and preset normal harmonic data, the working state of the charging pile is detected for abnormality, and a graded abnormal alarm signal is generated. Intelligent decisions are made based on the alarm signal, charging control instructions are issued, and the charging operation of the charging pile is dynamically adjusted. The present invention accurately detects abnormal conditions during the charging process by analyzing the harmonic characteristics of the charging current and voltage, improves the accuracy of abnormal detection, makes adjustments according to different abnormal conditions, and improves the safety and reliability of the charging operation.

[0076] above Figure 2 The charging detection device of the charger in the embodiment of the present invention is described in detail from the perspective of modular functional entities. The charging detection device of the charger in the embodiment of the present invention is described in detail from the perspective of hardware processing.

[0077] Figure 3is a schematic diagram of the structure of a charging detection device for a charger provided in an embodiment of the present invention. The charging detection device 300 for the charger may have relatively large differences due to different configurations or performances, and may include one or more processors (central processing units, CPU) 310 (for example, one or more processors) and a memory 320, and one or more storage media 330 (for example, one or more mass storage device terminals) storing application programs 333 or data 332. Among them, the memory 320 and the storage medium 330 may be short-term storage or permanent storage. The program stored in the storage medium 330 may include one or more modules (not shown in the figure), and each module may include a series of instruction operations in the charging detection device 300 of the charger. Furthermore, the processor 310 may be configured to communicate with the storage medium 330, and execute a series of instruction operations in the storage medium 330 on the charging detection device 300 of the charger to implement the steps of the charging detection method of the charger described above.

[0078] The charging detection device 300 of the charger may also include one or more power supplies 340, one or more wired or wireless network interfaces 350, one or more input and output interfaces 360, and / or one or more operating systems 331, such as Windows Serve, Mac OS X, Unix, Linux, FreeBSD, etc. It will be appreciated by those skilled in the art that Figure 3 The structure of the charging detection device of the charger shown does not constitute a limitation on the charging detection device of the charger provided by the present invention, and may include more or fewer components than shown in the figure, or combine certain components, or arrange the components differently.

[0079] The present invention also provides a computer-readable storage medium, which may be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium. Instructions are stored in the computer-readable storage medium. When the instructions are executed on a computer, the computer executes the steps of the charging detection method of the charger.

[0080] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the above-described system, device, or unit can refer to the corresponding process in the aforementioned method embodiment and will not be repeated here.

[0081] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art or the whole or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions to enable a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk and other media that can store program code.

[0082] As described above, the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features thereof may be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A charging detection method for a charger, characterized in that: The charger includes a single-channel single-plug charging pile, a harmonic energy meter, and a management system. The charging detection method of the charger includes: When the charging operation of the single-channel single-plug charging pile is started and the charging operation is in progress, the starting charging current and starting charging voltage, dynamic charging current and dynamic charging voltage of the single-channel single-plug charging pile are obtained through the harmonic electric energy meter; Performing initial harmonic characteristic analysis processing on the startup charging current and the startup charging voltage through the management system to obtain a reference harmonic characteristic data set, and performing dynamic harmonic characteristic analysis processing on the dynamic charging current and the dynamic charging voltage to obtain a dynamic harmonic characteristic data set; The management system performs abnormal detection processing on the working state of the single-channel single-plug charging pile according to the reference harmonic feature data set, the dynamic harmonic feature data set and the preset normal harmonic data to obtain a graded abnormal alarm signal; The management system performs intelligent decision-making processing on the charging operation of the single-channel single-plug charging pile according to the graded abnormal alarm signal to obtain a charging control instruction, and adjusts the charging operation of the single-channel single-plug charging pile according to the charging control instruction, wherein the management system performs intelligent decision-making processing on the charging operation of the single-channel single-plug charging pile according to the graded abnormal alarm signal to obtain a charging control instruction, and adjusts the charging operation of the single-channel single-plug charging pile according to the charging control instruction includes: the management system performs multimodal feature extraction processing on the charging process data of the charging operation of the single-channel single-plug charging pile according to the graded abnormal alarm signal to obtain a high-dimensional abnormal feature map. The method comprises the following steps: performing topological mapping processing on the charging state space according to the high-dimensional abnormal feature tensor to obtain an abnormal state topological map, and performing graph neural network analysis processing on the abnormal state topological map to obtain an abnormal propagation prediction model; performing metaheuristic algorithm solution processing on the multi-objective charging optimization problem according to the abnormal propagation prediction model to obtain a Pareto optimal solution set, and performing fuzzy decision processing on the Pareto optimal solution set to obtain a fuzzy charging control strategy; performing adaptive model predictive control processing on the charging operation of the single-channel single-plug charging pile according to the fuzzy charging control strategy to obtain a charging control instruction, and adjusting the charging operation of the single-channel single-plug charging pile according to the charging control instruction.

2. The charging detection method of a charger according to claim 1, characterized in that: The management system performs initial harmonic characteristic analysis on the startup charging current and the startup charging voltage to obtain a reference harmonic characteristic data set, and performs dynamic harmonic characteristic analysis on the dynamic charging current and the dynamic charging voltage to obtain a dynamic harmonic characteristic data set, including: The management system performs harmonic content calculation processing on the starting charging current and the starting charging voltage to obtain initial harmonic characteristic data, and extracts the main harmonic components of the initial harmonic characteristic data to obtain a reference harmonic characteristic data set; Performing periodic harmonic measurement processing on the dynamic charging current and the dynamic charging voltage to obtain time series harmonic data; Harmonic parameters in the time series harmonic data are extracted, and a change trend analysis process is performed on the harmonic parameters to obtain a dynamic harmonic feature data set.

3. The charging detection method of a charger according to claim 2, characterized in that: The step of extracting harmonic parameters from the time series harmonic data and performing a change trend analysis on the harmonic parameters to obtain a dynamic harmonic feature data set includes: Performing sliding window segmentation processing on the time series harmonic data to obtain harmonic data subsets of multiple time windows, and extracting harmonic features in the harmonic data subsets to obtain a harmonic feature sequence; Performing adjacent window difference calculation processing on the harmonic characteristic sequence to obtain harmonic parameter change rate data; According to the harmonic parameter change rate data, the charging operation of the single-channel single-plug charging pile is divided into stages to obtain a charging stage identification result; The charging stage identification result and the corresponding harmonic feature sequence are subjected to correlation analysis to obtain a dynamic harmonic feature data set.

4. The charging detection method of a charger according to claim 1, characterized in that: The management system performs abnormal detection processing on the working state of the single-channel single-plug charging pile according to the reference harmonic feature data set, the dynamic harmonic feature data set and the preset normal harmonic data to obtain a graded abnormal alarm signal, including: The management system performs dynamic deviation calculation processing on the harmonic changes of the charging operation of the single-channel single-plug charging pile according to the dynamic harmonic characteristic data set and the preset normal harmonic data to obtain a harmonic deviation time series; Judging whether there is an abnormality in the charging operation of the single-channel single-plug charging pile according to the harmonic deviation time series; If so, the abnormal type of the working state of the single-channel single-plug charging pile is determined according to the harmonic deviation time series and the reference harmonic characteristic data set, and a graded abnormal alarm signal corresponding to the abnormal type is generated.

5. The charging detection method of a charger according to claim 4, characterized in that: The management system performs dynamic deviation calculation processing on the harmonic change of the charging operation of the single-channel single-plug charging pile according to the dynamic harmonic feature data set and the preset normal harmonic data to obtain the harmonic deviation time series, including: Performing time alignment processing on the dynamic harmonic feature data set and the preset normal harmonic data, and performing multi-scale deviation calculation on the dynamic harmonic feature data set and the preset normal harmonic data after time alignment to obtain a multi-scale harmonic deviation matrix; Performing nonlinear feature extraction processing on the multi-scale harmonic deviation matrix to obtain a harmonic anomaly feature vector; The charging process is dynamically clustered according to the harmonic anomaly feature vector to obtain a harmonic state evolution sequence, and the harmonic state evolution sequence is subjected to time series prediction processing to obtain a harmonic deviation time series.

6. The charging detection method of a charger according to claim 4, characterized in that: The determining the abnormal type of the working state of the single-channel single-plug charging pile according to the harmonic deviation time series and the reference harmonic characteristic data set, and generating a graded abnormal alarm signal corresponding to the abnormal type includes: Performing time-frequency analysis on the harmonic deviation time series to obtain a harmonic anomaly characteristic spectrum; Matching and identifying the abnormal pattern of the single-channel single-plug charging pile according to the harmonic abnormality feature map and the reference harmonic feature data set to obtain an abnormality type determination result, wherein the abnormality type determination result includes the abnormality type and the probability distribution of the abnormality type; Performing fuzzy reasoning processing on the abnormality type determination result to obtain a comprehensive abnormality evaluation index; The abnormal state of the single-channel single-plug charging pile is graded and processed according to the comprehensive abnormality evaluation index to obtain a graded abnormality alarm signal.

7. A charging detection device for a charger, characterized in that: The charger includes a single-channel single-plug charging pile, a harmonic energy meter, and a management system. The charging detection device of the charger includes: A data detection module, used for obtaining the starting charging current and starting charging voltage, dynamic charging current and dynamic charging voltage of the single-channel single-plug charging pile through the harmonic electric energy meter when the charging operation of the single-channel single-plug charging pile is started and when the charging operation is in progress; A harmonic analysis module, configured to perform initial harmonic characteristic analysis processing on the startup charging current and the startup charging voltage through the management system to obtain a reference harmonic characteristic data set, and perform dynamic harmonic characteristic analysis processing on the dynamic charging current and the dynamic charging voltage to obtain a dynamic harmonic characteristic data set; an abnormality detection module, used to perform abnormality detection processing on the working state of the single-channel single-plug charging pile according to the reference harmonic characteristic data set, the dynamic harmonic characteristic data set and the preset normal harmonic data through the management system to obtain a graded abnormality alarm signal; The intelligent decision-making module is used to perform intelligent decision-making processing on the charging operation of the single-channel single-plug charging pile according to the hierarchical abnormal alarm signal through the management system to obtain a charging control instruction, and adjust the charging operation of the single-channel single-plug charging pile according to the charging control instruction, wherein the intelligent decision-making processing on the charging operation of the single-channel single-plug charging pile according to the hierarchical abnormal alarm signal through the management system to obtain a charging control instruction, and adjust the charging operation of the single-channel single-plug charging pile according to the charging control instruction includes: performing multimodal feature extraction processing on the charging process data of the charging operation of the single-channel single-plug charging pile according to the hierarchical abnormal alarm signal through the management system to obtain a high-dimensional Abnormal feature tensor; performing topological mapping processing on the charging state space according to the high-dimensional abnormal feature tensor to obtain an abnormal state topological map, and performing graph neural network analysis processing on the abnormal state topological map to obtain an abnormal propagation prediction model; performing meta-heuristic algorithm solution processing on the multi-objective charging optimization problem according to the abnormal propagation prediction model to obtain a Pareto optimal solution set, and performing fuzzy decision processing on the Pareto optimal solution set to obtain a fuzzy charging control strategy; performing adaptive model predictive control processing on the charging operation of the single-channel single-plug charging pile according to the fuzzy charging control strategy to obtain a charging control instruction, and adjusting the charging operation of the single-channel single-plug charging pile according to the charging control instruction.

8. A charging detection device for a charger, characterized in that: The charging detection device of the charger comprises: a memory and at least one processor, wherein the memory stores instructions; The at least one processor calls the instructions in the memory to enable the charging detection device of the charger to perform the steps of the charging detection method of the charger as described in any one of claims 1-6.

9. A computer-readable storage medium having instructions stored thereon, characterized in that: When the instructions are executed by the processor, the steps of the charging detection method of the charger as described in any one of claims 1 to 6 are implemented.

Images