Intelligent control system and method for alternating current charging pile of electric vehicle

Abstract

The invention relates to the field of charging pile control, and particularly discloses an intelligent control system and method for an alternating current charging pile of an electric vehicle. The method comprises the following steps: firstly, acquiring a relay image acquired by a camera, a relay detection chip output detection signal acquired by a sensor and alternating current charging pile relay voltage values acquired by a voltage sensor at a plurality of preset time points, then performing feature extraction and correlation analysis on the three parts by using a deep learning technology, and finally, performing classification through a classifier. Therefore, the abnormal or fault condition of the relay can be found in time, the safety risk and the equipment damage can be avoided, and the safety and the reliability of the charging pile can be improved.

Description

Technical Field

[0001] The present application relates to the field of charging pile control, and more specifically, to an intelligent control system and method for an AC charging pile of an electric vehicle. Background Art

[0002] Electricity is the main energy supply for new energy vehicles. Charging piles function similarly to gas pumps in gas stations. They can be fixed on the ground or on the wall and installed in public buildings (public buildings, shopping malls, public parking lots, etc.) and residential parking lots or charging stations. They can charge various types of electric vehicles according to different voltage levels. The input end of the charging pile is directly connected to the AC power grid, and the output end is equipped with a charging plug for charging electric vehicles.

[0003] When the AC power relay of an AC charging pile is working under high current conditions, the contacts may stick and fail to release quickly, causing the charging pile to be unable to control normal power outages and require maintenance. More seriously, when other faults occur at the same time, there is a risk of electric shock to operators and overall loss of control. Therefore, it is necessary to be unable to cut off the AC power supply in an emergency, resulting in the risk of electric shock on the user side.

[0004] Therefore, an intelligent control system and method for an AC charging pile of an electric vehicle is desired. Summary of the invention

[0005] In order to solve the above technical problems, this application is proposed. The embodiment of the application provides an intelligent control system and method for an AC charging pile of an electric vehicle, which first obtains a relay image collected by a camera, a relay detection chip output detection signal collected by a sensor, and an AC charging pile relay voltage value at multiple predetermined time points collected by a voltage sensor, and then uses deep learning technology to perform feature extraction and correlation analysis on the three, and finally uses a classifier to determine whether an emergency stop is required for the AC charging pile of the electric vehicle, so as to timely detect relay abnormalities or faults, avoid safety risks and equipment damage, and thus improve the safety and reliability of the charging pile.

[0006] According to one aspect of the present application, an intelligent control system for an AC charging pile of an electric vehicle is provided, comprising:

[0007] The AC charging pile data acquisition module is used to acquire the relay image collected by the camera, the relay detection chip output detection signal collected by the sensor, and the AC charging pile relay voltage values ​​at multiple predetermined time points collected by the voltage sensor;

[0008] An AC charging pile data extraction module, used to extract a relay appearance channel attention feature vector and a relay detection information association feature vector from the relay image collected by the camera, the relay detection chip output detection signal collected by the sensor, and the AC charging pile relay voltage values ​​at multiple predetermined time points collected by the voltage sensor;

[0009] The AC charging pile emergency stop processing judgment module is used to judge whether the electric vehicle AC charging pile needs to be emergency stopped based on the relay appearance channel attention feature vector and the relay detection information association feature vector.

[0010] According to another aspect of the present application, there is provided an intelligent control method for an AC charging pile of an electric vehicle, comprising:

[0011] Acquire a relay image acquired by a camera, a relay detection chip output detection signal acquired by a sensor, and an AC charging pile relay voltage value at a plurality of predetermined time points acquired by a voltage sensor;

[0012] Extracting a relay appearance channel attention feature vector and a relay detection information association feature vector from the relay image collected by the camera, the relay detection chip output detection signal collected by the sensor, and the AC charging pile relay voltage values ​​at multiple predetermined time points collected by the voltage sensor;

[0013] Based on the relay appearance channel attention feature vector and the relay detection information association feature vector, it is determined whether an emergency stop is required for the AC charging pile of the electric vehicle.

[0014] Compared with the prior art, the present application provides an intelligent control system and method for an AC charging pile for an electric vehicle, which first obtains a relay image captured by a camera, a detection signal output by a relay detection chip captured by a sensor, and an AC charging pile relay voltage value at multiple predetermined time points captured by a voltage sensor, and then uses deep learning technology to perform feature extraction and correlation analysis on the three, and finally uses a classifier to determine whether the electric vehicle AC charging pile needs to be emergency stopped, thereby promptly discovering relay abnormalities or faults, avoiding safety risks and equipment damage, and thereby improving the safety and reliability of the charging pile. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] By describing the embodiments of the present application in more detail in conjunction with the accompanying drawings, the above and other purposes, features and advantages of the present application will become more apparent. The accompanying drawings are used to provide a further understanding of the embodiments of the present application and constitute a part of the specification. Together with the embodiments of the present application, they are used to explain the present application and do not constitute a limitation of the present application. In the accompanying drawings, the same reference numerals generally represent the same components or steps.

[0016] Figure 1 The block diagram is a block diagram of an intelligent control system for an AC charging pile of an electric vehicle according to an embodiment of the present application.

[0017] Figure 2 The block diagram is a data extraction module of an AC charging pile in an intelligent control system of an AC charging pile for an electric vehicle according to an embodiment of the present application.

[0018] Figure 3 The present invention is a block diagram of an AC charging pile emergency stop processing judgment module in an intelligent control system of an AC charging pile for electric vehicles according to an embodiment of the present application.

[0019] Figure 4 The figure is a flow chart of an intelligent control method for an AC charging pile of an electric vehicle according to an embodiment of the present application.

[0020] Figure 5 is a block diagram of an electronic device according to an embodiment of the present application. DETAILED DESCRIPTION

[0021] Below, the exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application, and it should be understood that the present application is not limited to the exemplary embodiments described here.

[0022] Figure 1 FIG. 1 is a block diagram of an intelligent control system for an electric vehicle AC charging pile according to an embodiment of the present application. Figure 1 As shown, according to the embodiment of the present application, the intelligent control system 100 of the AC charging pile of an electric vehicle includes: an AC charging pile data acquisition module 110, which is used to obtain the relay image collected by the camera, the relay detection chip output detection signal collected by the sensor, and the AC charging pile relay voltage value at multiple predetermined time points collected by the voltage sensor; an AC charging pile data extraction module 120, which is used to extract the relay appearance channel attention feature vector and the relay detection information association feature vector from the relay image collected by the camera, the relay detection chip output detection signal collected by the sensor, and the AC charging pile relay voltage value at multiple predetermined time points collected by the voltage sensor; an AC charging pile emergency stop processing judgment module 130, which is used to judge whether the electric vehicle AC charging pile needs to be emergency stopped based on the relay appearance channel attention feature vector and the relay detection information association feature vector.

[0023] In the above-mentioned intelligent control system 100 of the AC charging pile of electric vehicles, the AC charging pile data acquisition module 110 is used to obtain the relay image collected by the camera, the relay detection chip output detection signal collected by the sensor, and the AC charging pile relay voltage value at multiple predetermined time points collected by the voltage sensor. It should be understood that electricity is the main energy supply mode for new energy vehicles. The charging pile is similar to the gas pump at a gas station, fixedly installed on the ground or wall, and distributed in public buildings (such as shopping malls, public parking lots) and parking lots or charging stations in residential areas. They can provide charging services according to different models and voltage requirements of electric vehicles. The input end of the charging pile is directly connected to the AC power grid, and the output end is equipped with a charging plug for charging electric vehicles. When using an AC charging pile, there is a potential problem: under high current working conditions, the AC power relay of the charging pile may have contact adhesion, resulting in the inability to disconnect the power supply in time. This situation may hinder the normal power control function of the charging pile, and maintenance is required to restore operation. More seriously, if such faults occur at the same time as other problems, such as loss of control of operation or risk of electric shock to the operator, the possibility of safety accidents will increase. Therefore, in the technical solution of the present application, by acquiring the relay image captured by the camera, the relay detection chip output detection signal collected by the sensor, and the AC charging pile relay voltage values ​​at multiple predetermined time points collected by the voltage sensor, and combining with deep learning technology, timely detection and judgment can be made as to whether the electric vehicle AC charging pile needs to be emergency stopped. This can effectively detect relay anomalies or faults, thereby preventing potential safety hazards and equipment damage, and further improving the operational safety and stability of the charging pile.

[0024] Specifically, the camera provides visual information of the relay, including its appearance and status, such as whether it is damaged or abnormal. The output signal of the detection chip obtained by the sensor reflects the internal working status and circuit information of the relay, such as switch status, current load, etc. The voltage value recorded by the voltage sensor is used to monitor the power supply of the charging pile, including the stability and change trend of the voltage, to ensure the safety and effectiveness of the charging process. These data are extracted, correlated and analyzed through the system's data processing and analysis module, so as to realize real-time monitoring and intelligent control of the operating status of the charging pile, and ensure the reliability of electric vehicle charging facilities and user safety.

[0025] In the above-mentioned intelligent control system 100 for AC charging pile of electric vehicle, the AC charging pile data extraction module 120 is used to extract the relay appearance channel attention feature vector and the relay detection information association feature vector from the relay image collected by the camera, the relay detection chip output detection signal collected by the sensor, and the AC charging pile relay voltage value at multiple predetermined time points collected by the voltage sensor. It should be understood that the purpose of extracting the relay appearance channel attention feature vector and the relay detection information association feature vector is to achieve comprehensive monitoring and real-time management of the operating status of the AC charging pile of electric vehicle through comprehensive analysis and intelligent judgment, and to ensure the safety and convenience of the user's charging experience.

[0026] Figure 2 FIG. 1 is a block diagram of an AC charging pile data extraction module in an intelligent control system for an electric vehicle AC charging pile according to an embodiment of the present application. Figure 2 As shown, in a specific embodiment of the present application, the AC charging pile data extraction module 120 includes: a relay image feature extraction unit 121, which is used to perform feature extraction on the relay image collected by the camera to obtain the relay appearance channel attention feature vector; a relay detection chip output detection feature extraction unit 122, which is used to perform feature extraction on the relay detection chip output detection signal collected by the sensor to obtain a global feature vector of relay detection chip output detection information; a relay voltage value feature extraction unit 123, which is used to perform feature extraction on the AC charging pile relay voltage values ​​at multiple predetermined time points collected by the voltage sensor to obtain an AC charging pile relay voltage feature vector; a relay detection information feature association unit 124, which is used to associate the relay detection chip output detection information global feature vector with the AC charging pile relay voltage feature vector to obtain the relay detection information association feature vector.

[0027] It should be understood that image feature extraction uses advanced algorithms such as convolutional neural networks to convert relay images into mathematically tractable feature representations. These feature representations not only include the appearance details of the relay, but also capture the structure and key information of different parts in the image. For example, CNN can learn the outline, edges and possible worn or damaged areas of the relay from the image, which are crucial for detecting abnormal working conditions of the relay. The extraction of the relay appearance channel attention feature vector involves further abstracting and encoding the high-dimensional features extracted from CNN. These vectors can describe the overall characteristics of the relay appearance, such as the geometric characteristics of the shape, the statistical information of the color distribution, and the marking of possible damaged or contaminated areas. Through effective feature encoding and abstraction, the system can accurately distinguish different types of relays and identify any potential operating problems or safety hazards. The extraction of the relay appearance channel attention feature vector not only helps to timely discover and solve the appearance problems of the relay, but also improves the system's monitoring accuracy and response speed of the charging pile equipment. This technical means not only ensures the safety and reliability of charging facilities, but also improves the user experience and service quality of electric vehicle users. In summary, image feature extraction has important practical significance and application prospects for realizing the intelligent management and maintenance of electric vehicle AC charging piles.

[0028] Furthermore, feature extraction of the detection signal output by the relay detection chip collected by the sensor is to achieve comprehensive monitoring and accurate analysis of the operating status of the relay in the AC charging pile of the electric vehicle. The sensor outputs the corresponding detection signal by monitoring the working parameters of the relay, such as the switch state, current load, working temperature, etc. These signals contain key information about the internal working state of the relay, but directly using the original signal may be too complex and lengthy, so feature extraction is needed to extract the most representative and distinguishing features. The feature extraction process converts the timing signal output by the sensor into a feature vector through signal processing technology and machine learning algorithms. These feature vectors not only include the static and dynamic characteristics of the relay, such as the frequency of the switch state, the shape, peak value and fluctuation of the current waveform, but also reflect the stability and reliability of the relay under different working conditions. The generation of the global feature vector of the detection information output by the relay detection chip involves the integration and comprehensive analysis of the feature vectors at multiple time points or under multiple working states. By comprehensively considering the feature changes and working state transitions in different time periods, the system can generate a global feature vector to describe the overall working characteristics and performance of the relay. These global feature vectors have the ability to describe the operating status of the relay more comprehensively and accurately, which helps the system to evaluate the operating health of the relay as a whole and predict possible failure risks.

[0029] Furthermore, feature extraction is performed on the voltage values ​​of the AC charging pile relay at multiple predetermined time points collected by the voltage sensor, mainly to monitor and evaluate the stability and quality of the power supply of the charging pile. Voltage is an important parameter required for the normal operation of the charging pile, which directly affects the safety and efficiency of the charging process. The feature extraction process obtains the voltage value sequence of the charging pile relay from multiple time points by sampling and analyzing the voltage signal. These voltage values ​​reflect the power input status of the charging pile under different working conditions, including the stability of the voltage, the degree of fluctuation, and possible abnormal changes. The generation of the AC charging pile relay voltage feature vector involves feature extraction and description of the collected voltage sequence. These feature vectors may include the average value, peak-to-peak value, frequency distribution of the waveform, harmonic components, etc. Through these feature vectors, the system can analyze and evaluate the overall characteristics and stability of the charging pile voltage, and identify the causes of voltage fluctuations or anomalies, such as instability of the power supply network or failure of the equipment.

[0030] In particular, the global feature vector of the detection information output by the relay detection chip and the voltage feature vector of the AC charging pile relay are associated, aiming to comprehensively analyze and evaluate the operating status and power supply quality of the charging pile relay, so as to achieve more comprehensive and accurate equipment monitoring and management. First, the global feature vector of the detection information output by the relay detection chip provides a comprehensive perspective on the internal working status of the relay by analyzing various working parameters and states output by the sensor, such as switch state, current load, etc. These feature vectors can reflect the operating stability, anomaly detection and fault prediction capabilities of the relay, providing a basis for real-time monitoring and remote diagnosis of the system. Secondly, the voltage feature vector of the AC charging pile relay reveals the quality and stability of the charging pile power supply through the timing analysis and feature extraction of the voltage signal. These feature vectors include information such as the average value, fluctuation amplitude, and frequency distribution of the voltage, which can effectively evaluate the reliability of the power supply and the voltage problems that may exist during the operation of the charging pile. The potential correlation and mutual influence between them can be found by correlating these two types of feature vectors. For example, when an abnormal state of the relay detection chip output is detected, it may be related to the voltage fluctuation or anomaly appearing in the voltage feature vector. Through correlation analysis, the system can more accurately identify the root causes of relay abnormalities, such as unstable power supply, excessive grid load or equipment aging.

[0031] In a specific embodiment of the present application, the relay image feature extraction unit 121 includes: passing the relay image captured by the camera through a relay appearance feature encoder based on a convolutional neural network to obtain an initial convolution feature map of the relay appearance; passing the initial convolution feature map of the relay appearance through a relay appearance channel attention encoder to obtain the relay appearance channel attention feature vector.

[0032] It should be understood that the initial convolutional feature map of the relay appearance is generated after each layer of convolutional operation of the convolutional neural network, and each layer of feature maps represents the representation of the image at different abstract levels. For example, the low-level feature map can capture the edges and simple shapes of the relay, while the high-level feature map can extract more complex semantic information, such as the overall appearance of the relay and possible damage areas. Through the initial convolutional feature map of the relay appearance, the system can achieve semantic understanding and automatic classification of relay images. These feature maps not only provide a highly abstract representation of the details of the relay appearance, but also can identify possible abnormalities or damage conditions, such as wear, cracks or contamination, without human intervention, so as to timely discover and deal with equipment problems. The relay appearance feature encoder based on the convolutional neural network can effectively convert the relay image captured by the camera into a semantically rich initial convolutional feature map of the relay appearance. This technology not only improves the level of intelligence of equipment monitoring and maintenance, but also provides important guarantees for the safety, reliability and service quality of electric vehicle charging piles. Specifically, each layer of the relay appearance feature encoder based on the convolutional neural network is used to perform convolution processing, mean pooling processing based on the local feature matrix, and nonlinear activation processing on the input data in the forward transfer of the layer, so that the last layer of the relay appearance feature encoder based on the convolutional neural network outputs the initial convolution feature map of the relay appearance, wherein the input of the relay appearance feature encoder based on the convolutional neural network is the relay image captured by the camera.

[0033] Furthermore, the relay appearance initial convolutional feature map is converted into a relay appearance channel attention feature vector through the relay appearance channel attention encoder in order to further refine the analysis and understanding of the importance and contribution of different parts in the relay appearance image. This process improves the effect and accuracy of the relay appearance image in automated monitoring and classification tasks by strengthening the representation of key feature areas. The relay appearance channel attention encoder uses the attention mechanism to weight and integrate the initial convolutional feature map of the relay appearance. This attention mechanism can dynamically assign the importance weights of different channels (feature maps) and adjust them according to their contribution in interpreting the overall semantic information and local details of the image. For example, for the relay appearance image, the attention mechanism can give priority to the key components of the relay, such as switches, connectors, or logos, so as to enhance the feature representation of these areas. The relay appearance channel attention feature vector is generated by the attention encoder, which reflects the importance and contribution of each channel in the relay appearance image in the overall image representation. Through the generation of the relay appearance channel attention feature vector, the system can achieve a deeper and more accurate understanding of the relay appearance image. This refined feature representation not only helps to improve the real-time perception capability of the equipment monitoring system on the relay status, but also can provide maintenance personnel with more targeted guidance and suggestions, such as detecting possible faulty or damaged relay components. Specifically, each layer of the relay appearance channel attention encoder performs the following operations on the input data in the forward pass of the layer: convolution processing on the input data based on a two-dimensional convolution kernel to generate a convolution feature map; pooling processing on the convolution feature map to generate a pooling feature map; activation processing on the pooling feature map to generate an activation feature map; calculating the quotient of the eigenvalue mean of the feature matrix corresponding to each channel in the activation feature map and the sum of the eigenvalue means of the feature matrices corresponding to all channels as the weighting coefficient of the feature matrix corresponding to each channel; weighting the feature matrix of each channel with the weighting coefficient of each channel in the activation feature map to generate a channel attention feature map; wherein the output of the last layer of the relay appearance channel attention encoder is the relay appearance channel attention feature vector.

[0034] In a specific embodiment of the present application, the relay detection chip output detection feature extraction unit 122 includes: performing word segmentation on the relay detection chip output detection signal collected by the sensor to obtain multiple relay detection chip output detection information data items; passing the multiple relay detection chip output detection information data items through a relay detection chip output detection information multi-scale feature extractor to obtain the relay detection chip output detection information global feature vector.

[0035] It should be understood that the signal output by the relay detection chip is usually a continuous waveform, which may contain information about multiple physical quantities, such as current, voltage, temperature, etc. This information changes continuously in time, so it needs to be divided into multiple discrete data items by word segmentation, each of which corresponds to a specific detection parameter or state. The word segmentation process involves signal processing and data analysis techniques, including filtering, sampling, and time window division. These techniques can divide the continuous detection signal into a series of discrete data points or data segments, and each data item reflects the specific physical quantity value or state information detected within a specific time period. The multiple relay detection chip output detection information data items obtained by word segmentation can provide a basis for subsequent data analysis and algorithm processing. For example, statistical analysis, trend analysis, or anomaly detection can be performed on each data item, so as to timely discover and diagnose possible operating problems or potential faults of the relay.

[0036] Furthermore, multiple relay detection chip output detection information data items are converted into relay detection chip output detection information global feature vectors through relay detection chip output detection information multi-scale feature extractors, aiming to describe and understand the overall operation status and performance characteristics of the relay more comprehensively and accurately by combining feature representations at multiple scales and levels. The multiple data items output by the relay detection chip may cover a variety of physical quantity information in different time scales and frequency ranges, such as transient response, long-term steady state and periodic changes. The multi-scale feature extractor can perform signal analysis and feature extraction at different scales for these data items through a variety of feature extraction methods (such as wavelet transform, spectrum analysis, etc.). The generation of the global feature vector of the relay detection chip output detection information is achieved by integrating feature information of different scales and levels into a unified feature vector. These global feature vectors integrate the key features of the relay in different time periods and frequencies during operation, such as maximum value, average value, change trend, etc., thereby forming an abstract expression of the overall operation status of the relay. Through the global feature vector of the relay detection chip output detection information, high-level representation and overall grasp of the relay state can be achieved. This feature vector can not only reduce the dimension and complexity of the data, but also improve the efficiency and accuracy of data analysis and machine learning algorithms, such as for applications such as anomaly detection, fault prediction, and equipment performance optimization. Specifically, the relay detection chip outputs a multi-scale feature extractor for detection information, including: a first convolutional layer, a second convolutional layer parallel to the first convolutional layer, and a cascade layer connected to the first convolutional layer and the second convolutional layer, wherein the first convolutional layer uses a one-dimensional convolutional kernel with a first scale, and the second convolutional layer uses a one-dimensional convolutional kernel with a second scale.

[0037] In a specific embodiment of the present application, the relay voltage value feature extraction unit 123 includes: constructing the AC charging pile relay voltage values ​​at multiple predetermined time points collected by the voltage sensor into an AC charging pile relay voltage input vector; passing the AC charging pile relay voltage input vector through an AC charging pile relay voltage timing encoder to obtain the AC charging pile relay voltage feature vector.

[0038] It should be understood that the voltage sensor continuously collects the voltage data of the relay at a predetermined time point. These data are usually periodic waveforms of alternating current, and their values ​​change periodically over time. By constructing a voltage input vector, these complex periodic signals can be converted into a vector with fixed dimensions, each dimension corresponding to a voltage value at a specific time point. The construction of the voltage input vector utilizes the discretization of time points, which helps to simplify the data processing and analysis process. For example, several time points within each cycle can be selected for sampling, or voltage data can be extracted regularly according to predetermined monitoring requirements. This discrete representation not only reduces the storage and transmission costs of data, but also helps to improve the efficiency and real-time performance of data processing. Through the voltage input vector, the voltage of the AC charging pile relay can be quickly monitored and analyzed. The monitoring system can perform real-time detection of anomalies or trend analysis based on the data in the vector to determine whether the operating status of the charging pile relay is normal. For example, fault diagnosis can be performed by comparing the voltage value in the vector with a preset threshold, or the frequency and amplitude changes of the voltage waveform can be analyzed to evaluate the stability and performance of the equipment.

[0039] Furthermore, the AC charging pile relay voltage input vector is converted into an AC charging pile relay voltage feature vector through an AC charging pile relay voltage timing encoder, aiming to perform deep processing and feature extraction of time series data through the timing encoder, thereby capturing important patterns and periodic changes in the voltage signal, and providing accurate and meaningful information for higher-level data analysis and system decision-making. The AC charging pile relay voltage timing encoder is capable of effectively modeling and encoding the time series data in the input vector. This includes adopting appropriate neural network structures (such as recurrent neural networks or transformer models) to handle complex periodicity and timing dependencies in voltage signals. Through these techniques, the timing encoder is able to extract important features hidden in the time dimension from the input vector. The AC charging pile relay voltage feature vector is a high-level abstract representation generated by the timing encoder, which integrates the dynamic changes and frequency characteristics of the voltage signal in time. These feature vectors can include frequency components, amplitude changes, phase information, etc. of the voltage waveform, reflecting the overall characteristics of the voltage state of the charging pile relay in different time periods. Higher-level data analysis and applications can be achieved through the AC charging pile relay voltage feature vector. For example, these feature vectors can be used for voltage load prediction, charging efficiency optimization, or equipment health status monitoring. The generation of feature vectors also provides more informative input for algorithms based on machine learning or deep learning, thereby improving the system's capabilities in fault diagnosis, anomaly detection, and real-time feedback. Specifically, the fully connected layer of the AC charging pile relay voltage timing encoder is used to fully connect the AC charging pile relay voltage input vector to extract high-dimensional implicit features of the feature values ​​at each position in the AC charging pile relay voltage input vector; and the one-dimensional convolutional layer of the AC charging pile relay voltage timing encoder is used to one-dimensionally encode the AC charging pile relay voltage input vector to extract high-dimensional implicit correlation features of the correlation between the feature values ​​at each position in the AC charging pile relay voltage input vector.

[0040] In a specific embodiment of the present application, the AC charging pile relay voltage input vector is passed through an AC charging pile relay voltage timing encoder to obtain the AC charging pile relay voltage feature vector, including: using the fully connected layer of the AC charging pile relay voltage timing encoder to fully connect the AC charging pile relay voltage input vector to extract high-dimensional implicit features of the eigenvalues ​​at each position in the AC charging pile relay voltage input vector; using the one-dimensional convolutional layer of the AC charging pile relay voltage timing encoder to one-dimensionally encode the AC charging pile relay voltage input vector to extract high-dimensional implicit correlation features of the correlation between the eigenvalues ​​at each position in the AC charging pile relay voltage input vector.

[0041] It should be understood that the use of the fully connected layer and one-dimensional convolutional layer of the AC charging pile relay voltage time series encoder to encode the AC charging pile relay voltage input vector is intended to extract rich and meaningful high-dimensional implicit features from the input vector through different neural network layers. These features reflect the complex changes and associations of the voltage signal in time and space. The role of the fully connected layer in the time series encoder is to extract global high-dimensional implicit features of the voltage input vector. The fully connected layer linearly combines and transforms the eigenvalues ​​of each position in the input vector through the weight matrix and the bias vector, thereby obtaining a highly abstract feature representation of each position. These features include statistical information such as the mean, variance, and maximum value of the voltage data, which can reflect the overall voltage change trend and fluctuation characteristics. The one-dimensional convolutional layer is used for local feature extraction and correlation analysis. The one-dimensional convolutional layer can effectively capture the local feature patterns between adjacent positions in the voltage input vector, such as short-term changes and frequency components. By sliding the input vector through convolution kernels of different sizes, the one-dimensional convolution layer can identify and enhance the characteristic patterns within a specific time range, and then extract high-dimensional implicit correlation features between positions, such as correlation or periodic changes between adjacent time points. By combining the fully connected layer and the one-dimensional convolution layer, the temporal encoder can convert the AC charging pile relay voltage input vector into a more abstract and informative feature expression. These features not only help to understand the overall trend and local fluctuations of the voltage signal, but also provide richer input data for subsequent deep learning models or machine learning algorithms. For example, based on these features, voltage status prediction, anomaly detection, or optimization adjustment of charging pile operation can be performed, thereby improving the stability and efficiency of charging facilities.

[0042] In the above-mentioned intelligent control system 100 for AC charging piles of electric vehicles, the emergency stop processing judgment module 130 of the AC charging pile is used to judge whether it is necessary to perform emergency stop processing on the AC charging pile of electric vehicles based on the attention feature vector of the relay appearance channel and the associated feature vector of the relay detection information. It should be understood that the attention feature vector of the relay appearance channel reflects the visual characteristics and appearance status of the relay device. These features may include damage to the relay housing, abnormal heat, light indication status, etc. Through the attention feature extraction of the visual channel, the system can identify appearance anomalies, such as obvious physical damage or abnormal heating, which may indicate that the equipment has potential safety risks and requires emergency stop processing to prevent further accidents. The relay detection information associated feature vector combines a variety of key operating data collected from sensors and monitoring equipment. These data may cover multiple parameters such as voltage, current, temperature, etc., which are used to evaluate the actual working state and performance of the relay. Through the analysis of the associated feature vector, the system can detect abnormal voltage, current fluctuation, high temperature and other indicators. These anomalies may indicate that the equipment is faulty or unstable, and emergency stop processing is required in time to avoid further damage to the equipment or cause safety hazards. Taking into account the appearance channel attention features and the detection information association features, the system can achieve a comprehensive assessment and monitoring of the status of the charging pile equipment. If either or both of the two feature vectors show abnormal signals, the system will determine that the equipment may have a serious risk of failure or safety issues, so triggering the emergency stop processing mechanism is a necessary measure. This may include remotely shutting down the power, disabling the charging pile function, or sending an alarm to notify the operation and maintenance personnel for emergency repairs. The decision to perform emergency stop processing based on the relay appearance channel attention feature vector and the relay detection information association feature vector not only improves the safety and stability of the charging pile equipment, but also ensures the overall safety of the user and the environment around the equipment. This intelligent safety monitoring and emergency handling mechanism is one of the important technical innovations and implementation plans in the management of modern electric vehicle charging infrastructure.

[0043] Figure 3 FIG. 1 is a block diagram of an AC charging pile emergency stop processing judgment module in an AC charging pile intelligent control system for electric vehicles according to an embodiment of the present application. Figure 3As shown, in a specific embodiment of the present application, the AC charging pile emergency stop processing judgment module 130 includes: an AC charging pile feature fusion unit 131, which is used to fuse the relay appearance channel attention feature vector and the relay detection information association feature vector to obtain an AC charging pile emergency stop judgment classification feature vector; an AC charging pile feature optimization unit 132, which is used to perform compatibility-aware feature manifold modulation on the AC charging pile emergency stop judgment classification feature vector to obtain an optimized AC charging pile emergency stop judgment classification feature vector; an electric vehicle AC charging pile emergency stop classification judgment unit 133, which is used to pass the optimized AC charging pile emergency stop judgment classification feature vector through a classifier to obtain a classification result, and the classification result is used to determine whether the electric vehicle AC charging pile needs to be emergency stopped.

[0044] It should be understood that the fusion of the relay appearance channel attention feature vector and the relay detection information association feature vector to obtain the AC charging pile emergency stop judgment classification feature vector is intended to improve the accurate assessment of the safety status of the charging pile equipment and the emergency stop judgment ability through the comprehensive analysis of multi-dimensional data, thereby effectively preventing potential safety risks and equipment failures. Taking into account the fusion of the feature vectors of the appearance channel and the detection information, the AC charging pile emergency stop judgment classification feature vector can comprehensively reflect the multi-faceted information of the equipment safety status. This fusion not only improves the comprehensive monitoring capability of the equipment health status, but also improves the accuracy and reliability of judgment and decision-making through the synergy of multi-source data. When the emergency stop judgment classification feature vector shows obvious abnormal patterns or warning signals, the system will automatically trigger the emergency stop mechanism, such as remotely stopping charging, turning off the power, or sending an emergency notification to the operation and maintenance personnel to prevent potential accidents or equipment damage.

[0045] Furthermore, in the technical solution of the present application, considering that the operating state of the AC charging pile involves information in multiple dimensions such as the appearance characteristics of the relay, the signal output by the detection chip, and the voltage value, by utilizing the prior information of the neural network model to modulate the characteristic manifold of the classification feature vector of the emergency stop judgment of the AC charging pile, the performance and robustness of the model in processing classification tasks can be optimized. Among them, the neural network can capture the complex associations and feature expressions in the data by learning the patterns and structures of a large amount of data. In this way, the model can more effectively distinguish different situations during classification, improve the accuracy and generalization ability of the classification task, thereby effectively improving the performance and reliability of the emergency stop judgment system of the AC charging pile, so that it can make emergency stop decisions quickly and accurately in a real-time environment, and ensure the safe operation of electric vehicle charging facilities.

[0046] Among them, the AC charging pile emergency stop judgment classification feature vector is modulated with a feature manifold based on compatibility perception to obtain an optimized AC charging pile emergency stop judgment classification feature vector, including: calculating the mean and variance of the AC charging pile emergency stop judgment classification feature vector; calculating the difference between the eigenvalue of the predetermined position of the AC charging pile emergency stop judgment classification feature vector and the mean of the AC charging pile emergency stop judgment classification feature vector to obtain a first intermediate calculation value; multiplying the network parameter matrix by the AC charging pile emergency stop judgment classification feature vector and then subtracting it from the AC charging pile emergency stop judgment classification feature vector by position to obtain a first intermediate vector, and calculating the Frobenius norm of the first intermediate vector to obtain a first norm value; adding the first norm value to the The square root of the variance of the AC charging pile emergency stop judgment classification feature vector is taken to obtain a second intermediate calculation value; the first intermediate calculation value is divided by the second intermediate calculation value to obtain a third intermediate calculation value; the squares of each position in the AC charging pile emergency stop judgment classification feature vector are calculated and then accumulated to obtain a first total value; the square root of the first total value is taken and then divided by the scale of the network parameter matrix to calculate the natural exponential function value to obtain a fourth intermediate calculation value, wherein the scale of the network parameter matrix is ​​equal to the product of the width value of the network parameter matrix and the height value thereof; the fourth intermediate calculation value is multiplied by the weighted hyperparameter and then added to the third intermediate calculation value to obtain the eigenvalue of the predetermined position of the optimized AC charging pile emergency stop judgment classification feature vector.

[0047] Among them, the AC charging pile emergency stop judgment classification feature vector is modulated based on the compatibility perception feature manifold to obtain the optimized AC charging pile emergency stop judgment classification feature vector, including:

[0048]

[0049] Wherein, V is the classification feature vector of the emergency stop judgment of the AC charging pile, M is the network parameter matrix, β represents the weighted hyperparameter, and v i is the eigenvalue of the i-th feature vector of the AC charging pile emergency stop judgment classification, N is the scale of the network parameter matrix, the scale of the network parameter matrix is ​​equal to the product of the width value and the height value of the network parameter matrix, L is the length of the AC charging pile emergency stop judgment classification feature vector, ||·|| F The Frobenius norm of the classification feature vector representing the emergency stop judgment of the AC charging pile, represents matrix multiplication, μ represents the mean of the feature set composed of all eigenvalues ​​of the classification feature vector for the emergency stop judgment of the AC charging pile, σ 2 represents the variance of the feature set composed of all eigenvalues ​​of the classification feature vector for the emergency stop judgment of the AC charging pile, v' iIt is the eigenvalue of the i-th feature in the optimized AC charging pile emergency stop judgment classification feature vector.

[0050] That is, in the technical solution of the present application, considering that when the neural network model performs feature mining on the input data based on the network parameter matrix, if the inherent implicit association between the network output feature and the network parameter matrix can be further considered, the prior information of the neural network model can be further used to modulate the feature manifold of the AC charging pile emergency stop judgment classification feature vector to reduce the possibility of special instance disturbances of its feature values. Based on this, in a preferred embodiment, the AC charging pile emergency stop judgment classification feature vector is modulated based on compatibility perception, which first maps the AC charging pile emergency stop judgment classification feature vector to the high-dimensional feature space where the network parameter matrix is ​​located, and uses the compatibility perception factor between the mapping result and the original input as the inherent implicit association descriptor between the network output feature and the network parameter matrix. Then, the statistical characteristics of the feature group of the AC charging pile emergency stop judgment classification feature vector are used as the inherent modal information of the AC charging pile emergency stop judgment classification feature vector, and the inherent implicit association descriptor is used as the instance disturbance differential adjustment variable, and the attribute values ​​of each position of the AC charging pile emergency stop judgment classification feature vector are dynamically standardized. At the same time, the class-standardized modulus data of the AC charging pile emergency stop judgment classification feature vector is used as a bias item to constrain the acceptable range of the group distribution of the AC charging pile emergency stop judgment classification feature vector. In this way, the attribute influence of the interference distribution of abnormal samples is suppressed as much as possible, thereby improving the expression accuracy of the AC charging pile emergency stop judgment classification feature vector.

[0051] Furthermore, the optimized emergency stop judgment classification feature vector integrates multi-dimensional data from the attention features of the relay appearance channel and the associated features of the relay detection information, which reflect the appearance status and key operating parameters of the equipment. Through the extraction and combination of feature vectors, a more comprehensive and in-depth understanding of the safety and health status of the equipment can be obtained. The classification results obtained by the classifier directly reflect whether the current charging pile equipment is in a state that requires emergency stop processing. When the classification results show that there is a safety risk or abnormality in the equipment, the system will automatically trigger the corresponding emergency stop measures, such as remotely stopping the charging process, disconnecting the power supply, or issuing an emergency notification to the operation and maintenance personnel. This automated emergency stop processing mechanism effectively improves the safety and reliability of charging facilities and avoids safety accidents or further damage that may be caused by equipment failure or abnormality.

[0052] In summary, by inputting the emergency stop judgment classification feature vector into the classifier, real-time monitoring and response to the status of the electric vehicle AC charging pile can be achieved. This data-driven safety management method not only improves the intelligence level of equipment operation, but also ensures the safety of users and equipment during the charging process, providing important support and guarantee for the sustainable operation of modern charging infrastructure.

[0053] In summary, the embodiment of the present application first obtains the relay image captured by the camera, the relay detection chip output detection signal collected by the sensor, and the AC charging pile relay voltage value at multiple predetermined time points collected by the voltage sensor, and then uses deep learning technology to perform feature extraction and correlation analysis on the three. Finally, a classifier is used to determine whether the electric vehicle AC charging pile needs to be emergency stopped, so as to promptly detect relay abnormalities or faults, avoid safety risks and equipment damage, and thereby improve the safety and reliability of the charging pile.

[0054] As described above, the electric vehicle AC charging pile intelligent control system 100 according to the embodiment of the present application can be implemented in various terminal devices. In one example, the electric vehicle AC charging pile intelligent control system 100 can be integrated into the terminal device as a software module and / or a hardware module. For example, the electric vehicle AC charging pile intelligent control system 100 can be a software module in the operating system of the terminal device, or can be an application developed for the terminal device; of course, the electric vehicle AC charging pile intelligent control system 100 can also be one of the many hardware modules of the terminal device.

[0055] Alternatively, in another example, the electric vehicle AC charging pile intelligent control system 100 and the terminal device may also be separate devices, and the electric vehicle AC charging pile intelligent control system 100 may be connected to the terminal device via a wired and / or wireless network, and transmit interactive information in accordance with an agreed data format.

[0056] Figure 4 Flow chart of the intelligent control method of the AC charging pile of an electric vehicle according to an embodiment of the present application. Figure 4As shown, according to the embodiment of the present application, the intelligent control method of the AC charging pile of an electric vehicle includes: S110, acquiring a relay image captured by a camera, a detection signal output by a relay detection chip captured by a sensor, and a voltage value of the AC charging pile relay at multiple predetermined time points captured by a voltage sensor; S120, extracting a relay appearance channel attention feature vector and a relay detection information association feature vector from the relay image captured by the camera, the detection signal output by the relay detection chip captured by the sensor, and the voltage value of the AC charging pile relay at multiple predetermined time points captured by the voltage sensor; S130, judging whether it is necessary to perform emergency stop processing on the AC charging pile of the electric vehicle based on the relay appearance channel attention feature vector and the relay detection information association feature vector.

[0057] Here, those skilled in the art can understand that the specific operations of each step in the above-mentioned electric vehicle AC charging pile intelligent control method have been referred to above. Figures 1 to 3 The description of the intelligent control system of the electric vehicle AC charging pile has been introduced in detail, and therefore, its repeated description will be omitted.

[0058] Below, reference Figure 5 To describe an electronic device according to an embodiment of the present application.

[0059] like Figure 5 As shown, the electronic device 10 includes: a processor 11 and a memory 13. The processor 11 and the memory 13 are connected, such as through a bus 12. Optionally, the electronic device 10 may also include a transceiver 14. It should be noted that in actual applications, the transceiver 14 is not limited to one, and the structure of the electronic device 10 does not constitute a limitation on the embodiments of the present invention.

[0060] The processor 11 may be a CPU (Central Processing Unit), a general-purpose processor, a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. It may implement or execute various exemplary logic blocks, modules, and circuits described in conjunction with the disclosure of the present invention. The processor 11 may also be a combination that implements computing functions, such as a combination of one or more microprocessors, a combination of a DSP and a microprocessor, and the like.

[0061] The bus 12 may include a path for transmitting information between the above components. The bus 12 may be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus. The bus 12 may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 5 Only one thick line is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0062] The memory 13 is used to store a computer program corresponding to the intelligent control method for an AC charging pile of an electric vehicle according to the above embodiment of the present invention, and the computer program is controlled and executed by the processor 11. The processor 11 is used to execute the computer program stored in the memory 13 to implement the contents shown in the above method embodiment.

[0063] Among them, the electronic device 10 includes but is not limited to: mobile terminals such as mobile phones, laptops, digital broadcast receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), vehicle-mounted terminals (such as vehicle-mounted navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. Figure 5 The electronic device 10 shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present invention.

[0064] It should be noted that the logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be specifically implemented in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in combination with these instruction execution systems, devices or apparatuses. For the purposes of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in combination with these instruction execution systems, devices or apparatuses. More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection portion with one or more wirings (electronic device), a portable computer disk box (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium and then editing, interpreting or processing in other suitable ways if necessary, and then stored in a computer memory.

[0065] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, a plurality of steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0066] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "examples", "specific examples", or "some examples" means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner.

[0067] In the description of the present invention, it is to be understood that the terms “center”, “longitudinal”, “lateral”, “length”, “width”, “thickness”, “up”, “down”, “front”, “back”, “left”, “right”, “vertical”, “horizontal”, “top”, “bottom”, “inside”, “outside”, “clockwise”, “counterclockwise”, “axial”, “radial”, “circumferential”, etc., indicating orientations or positional relationships based on the orientations or positional relationships shown in the accompanying drawings, are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the referred device or element must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as limiting the present invention.

[0068] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features. In the description of the present invention, the meaning of "plurality" is at least two, such as two, three, etc., unless otherwise clearly and specifically defined.

[0069] In the present invention, unless otherwise clearly specified and limited, the terms "installed", "connected", "connected", "fixed" and the like should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium, it can be the internal connection of two elements or the interaction relationship between two elements, unless otherwise clearly defined. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0070] In the present invention, unless otherwise clearly specified and limited, a first feature being "above" or "below" a second feature may mean that the first and second features are in direct contact, or the first and second features are in indirect contact through an intermediate medium. Moreover, a first feature being "above", "above" or "above" a second feature may mean that the first feature is directly above or obliquely above the second feature, or simply means that the first feature is higher in level than the second feature. A first feature being "below", "below" or "below" a second feature may mean that the first feature is directly below or obliquely below the second feature, or simply means that the first feature is lower in level than the second feature.

[0071] Although the embodiments of the present invention have been shown and described above, it is to be understood that the above embodiments are exemplary and are not to be construed as limitations of the present invention. A person skilled in the art may change, modify, replace and vary the above embodiments within the scope of the present invention.

Claims

1. An intelligent control system for an electric vehicle AC charging pile, characterized in that: include: The AC charging pile data acquisition module is used to acquire the relay image collected by the camera, the relay detection chip output detection signal collected by the sensor, and the AC charging pile relay voltage values ​​at multiple predetermined time points collected by the voltage sensor; An AC charging pile data extraction module, used to extract a relay appearance channel attention feature vector and a relay detection information association feature vector from the relay image collected by the camera, the relay detection chip output detection signal collected by the sensor, and the AC charging pile relay voltage values ​​at multiple predetermined time points collected by the voltage sensor; The AC charging pile emergency stop processing judgment module is used to judge whether the electric vehicle AC charging pile needs to be emergency stopped based on the relay appearance channel attention feature vector and the relay detection information association feature vector.

2. The intelligent control system for electric vehicle AC charging pile according to claim 1 is characterized in that: The AC charging pile data extraction module includes: A relay image feature extraction unit, used for performing feature extraction on the relay image collected by the camera to obtain the relay appearance channel attention feature vector; A relay detection chip output detection feature extraction unit, used for performing feature extraction on the relay detection chip output detection signal collected by the sensor to obtain a global feature vector of the relay detection chip output detection information; A relay voltage value feature extraction unit, used for extracting features from the AC charging pile relay voltage values ​​at a plurality of predetermined time points collected by the voltage sensor to obtain an AC charging pile relay voltage feature vector; The relay detection information feature association unit is used to associate the global feature vector of the detection information output by the relay detection chip with the voltage feature vector of the AC charging pile relay to obtain the relay detection information association feature vector.

3. The intelligent control system for electric vehicle AC charging pile according to claim 2 is characterized in that: The relay image feature extraction unit comprises: Passing the relay image captured by the camera through a relay appearance feature encoder based on a convolutional neural network to obtain an initial convolution feature map of the relay appearance; The relay appearance initial convolutional feature map is passed through a relay appearance channel attention encoder to obtain the relay appearance channel attention feature vector.

4. The intelligent control system for electric vehicle AC charging pile according to claim 3 is characterized in that: The relay detection chip output detection feature extraction unit comprises: Segmenting the relay detection chip output detection signal collected by the sensor to obtain a plurality of relay detection chip output detection information data items; The plurality of relay detection chip output detection information data items are passed through a relay detection chip output detection information multi-scale feature extractor to obtain a global feature vector of the relay detection chip output detection information.

5. The intelligent control system for electric vehicle AC charging pile according to claim 4 is characterized in that: The relay voltage value feature extraction unit comprises: Constructing the AC charging pile relay voltage values ​​at a plurality of predetermined time points collected by the voltage sensor as an AC charging pile relay voltage input vector; The AC charging pile relay voltage input vector is passed through an AC charging pile relay voltage timing encoder to obtain the AC charging pile relay voltage characteristic vector.

6. The intelligent control system for electric vehicle AC charging pile according to claim 5 is characterized in that: Passing the AC charging pile relay voltage input vector through an AC charging pile relay voltage timing encoder to obtain the AC charging pile relay voltage characteristic vector includes: Using the fully connected layer of the AC charging pile relay voltage timing encoder to fully connect encode the AC charging pile relay voltage input vector to extract high-dimensional implicit features of the feature values ​​at each position in the AC charging pile relay voltage input vector; and The one-dimensional convolution layer of the AC charging pile relay voltage timing encoder is used to one-dimensionally encode the AC charging pile relay voltage input vector to extract high-dimensional implicit correlation features of the correlation between the feature values ​​at each position in the AC charging pile relay voltage input vector.

7. The intelligent control system for electric vehicle AC charging pile according to claim 6 is characterized in that: The AC charging pile emergency stop processing judgment module includes: An AC charging pile feature fusion unit, used to fuse the relay appearance channel attention feature vector and the relay detection information association feature vector to obtain an AC charging pile emergency stop judgment classification feature vector; An AC charging pile feature optimization unit, used for performing feature manifold modulation based on compatibility perception on the AC charging pile emergency stop judgment classification feature vector to obtain an optimized AC charging pile emergency stop judgment classification feature vector; The electric vehicle AC charging pile emergency stop classification judgment unit is used to pass the optimized AC charging pile emergency stop judgment classification feature vector through a classifier to obtain a classification result, and the classification result is used to judge whether the electric vehicle AC charging pile needs to be emergency stopped.

8. The intelligent control system for electric vehicle AC charging pile according to claim 7 is characterized in that: The AC charging pile feature optimization unit includes: Calculate the mean and variance of the AC charging pile emergency stop judgment classification feature vector; Calculating the difference between the feature value of the predetermined position of the AC charging pile emergency stop judgment classification feature vector and the mean value of the AC charging pile emergency stop judgment classification feature vector to obtain a first intermediate calculation value; Multiplying the network parameter matrix by the AC charging pile emergency stop judgment classification feature vector and then subtracting the matrix from the AC charging pile emergency stop judgment classification feature vector by position to obtain a first intermediate vector, and calculating the Frobenius norm of the first intermediate vector to obtain a first norm value; The first norm value is added to the variance of the AC charging pile emergency stop judgment classification feature vector and then the square root is taken to obtain a second intermediate calculation value; dividing the first intermediate calculated value by the second intermediate calculated value to obtain a third intermediate calculated value; Calculating the square of each position in the AC charging pile emergency stop judgment classification feature vector and then accumulating and summing them to obtain a first total value; Calculate a natural exponential function value after taking the square root of the first sum value and dividing it by the scale of the network parameter matrix to obtain a fourth intermediate calculation value, wherein the scale of the network parameter matrix is ​​equal to the product of the width value of the network parameter matrix and the height value thereof; The fourth intermediate calculated value is multiplied by the weighted hyperparameter and then added to the third intermediate calculated value to obtain the feature value of the predetermined position of the optimized AC charging pile emergency stop judgment classification feature vector.

9. An intelligent control method for an electric vehicle AC charging pile, characterized in that: include: Acquire a relay image acquired by a camera, a relay detection chip output detection signal acquired by a sensor, and an AC charging pile relay voltage value at a plurality of predetermined time points acquired by a voltage sensor; Extracting a relay appearance channel attention feature vector and a relay detection information association feature vector from the relay image collected by the camera, the relay detection chip output detection signal collected by the sensor, and the AC charging pile relay voltage values ​​at multiple predetermined time points collected by the voltage sensor; Based on the relay appearance channel attention feature vector and the relay detection information association feature vector, it is determined whether an emergency stop is required for the AC charging pile of the electric vehicle.

10. The intelligent control method for an AC charging pile of an electric vehicle according to claim 9, characterized in that: Extracting a relay appearance channel attention feature vector and a relay detection information association feature vector from the relay image collected by the camera, the relay detection chip output detection signal collected by the sensor, and the AC charging pile relay voltage values ​​at multiple predetermined time points collected by the voltage sensor, including: Performing feature extraction on the relay image captured by the camera to obtain the relay appearance channel attention feature vector; Performing feature extraction on the relay detection chip output detection signal collected by the sensor to obtain a global feature vector of the relay detection chip output detection information; Performing feature extraction on the AC charging pile relay voltage values ​​at a plurality of predetermined time points collected by the voltage sensor to obtain an AC charging pile relay voltage feature vector; The global feature vector of the detection information output by the relay detection chip is associated with the voltage feature vector of the AC charging pile relay to obtain the relay detection information associated feature vector.

Images