Intelligent Charging Pile Operation Safety Control System

Abstract

The present application discloses an intelligent charging pile operation safety control system, belonging to the technical field of charging pile safety control, including: a plurality of networked charging piles; a charging pile monitoring module for monitoring the operation data of each of the charging piles, wherein the operation data includes electrical data, environmental data, and mechanical data; a safety status control model for obtaining the safety status control model of the charging pile according to the operation data of the charging pile through a neural network, simulating the stable state and safety change trend of the charging pile under different working conditions, and realizing the prediction of the operation state of the charging pile; at the same time, setting evaluation indexes for the operation data, setting target thresholds for the evaluation indexes, and presetting safety strategies for the evaluation indexes; thereby judging the safety attributes of the charging pile and matching the corresponding safety strategies for the operation data. The present invention improves the safety and stability of the charging pile.

Description

Technical Field

[0001] The present application relates to an intelligent charging pile operation safety control system, belonging to the technical field of charging pile safety control. Background Art

[0002] The charging pile safety control system is a comprehensive management system integrating intelligent, networked, and information technologies, aiming to ensure the safe operation of charging piles, improve charging efficiency, and optimize the user experience. With the rapid development of the new energy vehicle industry and the continuous growth of the number of charging piles, the intelligent charging pile operation safety control system meets and realizes the comprehensive control of the operation status of charging piles through real-time monitoring, data analysis and prediction, and the execution of safety strategies. Therefore, it is the general trend to optimize the operation and decision-making of the system through technologies such as artificial intelligence and big data to improve efficiency and the user experience. Summary of the Invention

[0003] According to one aspect of the present application, an intelligent charging pile operation safety control system is provided, which comprehensively controls the operation status of charging piles.

[0004] The intelligent charging pile operation safety control system is characterized by including:

[0005] Multiple networked charging piles;

[0006] A charging pile monitoring module for monitoring the operation data of each charging pile, where the operation data includes electrical data, environmental data, and mechanical data;

[0007] A safety status control model for obtaining the safety status control model of the charging pile according to the operation data of the charging pile through a neural network, simulating the stable state and safety change trend of the charging pile under different working conditions, and realizing the prediction of the operation status of the charging pile; at the same time, setting evaluation indicators for the operation data, setting target thresholds for the evaluation indicators, and presetting safety strategies for the evaluation indicators; thereby judging the safety attributes of the charging pile and matching the corresponding safety strategies for the operation data;

[0008] The electrical data is used to ensure the electrical insulation performance of the charging pile, the environmental data is used to ensure that the charging pile meets different environmental standards, and the mechanical data is used to ensure the mechanical performance of the charging pile.

[0009] Further, the evaluation indicators of the operation data include:

[0010] Electrical performance indicators, and the electrical performance indicators include an operating temperature indicator and a current indicator;

[0011] Environmental performance indicators, and the environmental temperature indicators include an environmental temperature indicator and an environmental humidity indicator;

[0012] Mechanical performance indicators, and the mechanical performance indicators include vibration indicators and displacement indicators.

[0013] Furthermore, the method for constructing the safety state control model includes:

[0014] Construct three model data sets based on electrical data, environmental data, and mechanical data respectively;

[0015] Divide the three model data sets into three types of base models;

[0016] Perform cross-modal feature fusion on the outputs of the three types of base models through a hierarchical attention mechanism to obtain a safety state control model.

[0017] Furthermore, performing cross-modal feature fusion on the outputs of the three types of base models through a hierarchical attention mechanism includes:

[0018] Align the time series of electrical data, environmental data, and mechanical data based on dynamic time warping;

[0019] Allocate cross-modal attention weights according to the physical space correlation between electrical data and mechanical data;

[0020] Among them, when calculating the cross-modal attention weights, it satisfies:

[0021] ;

[0022] In the formula, is the query matrix, is the key matrix, is the feature dimension, and the softmax function is used to convert the weight matrix into a probability distribution form;

[0023] Generate a gating coefficient using environmental data to dynamically adjust the cross-modal attention weights;

[0024] Fuse the time series aligned features with the weighted cross-modal features to generate a safety state representation.

[0025] Furthermore, setting a target threshold for the evaluation index includes:

[0026] Input the charging pile feature data in the operation data into the safety state control model to obtain the target prediction value corresponding to the operation data;

[0027] Determine the first prediction value and the second prediction value of the target prediction value, and use the average value of the first prediction value and the second prediction value as the reference threshold;

[0028] If the change amount of the reference threshold is less than the first preset threshold, use the reference threshold as the target threshold;

[0029] Among them, the first predicted value is the maximum value among the target predicted values;

[0030] The second predicted value is the minimum value among the target predicted values;

[0031] The first preset threshold is used to represent the tolerance of the stability of the reference threshold.

[0032] Furthermore, a safety policy is preset for the evaluation index, including:

[0033] Compare the electrical performance index with the corresponding target threshold. When the electrical performance index is between 5% and 10% of the target threshold, generate first adjustment information. When the electrical performance index exceeds 10% of the target threshold, stop the charging authorization action of the charging pile;

[0034] Compare the environmental performance index with the corresponding target threshold. When the environmental performance index is between 10% and 20% of the target threshold, generate second adjustment information. When the environmental performance index exceeds 20% of the target threshold, stop the charging authorization action of the charging pile;

[0035] Compare the mechanical performance index with the corresponding target threshold. When the mechanical performance index is between 3% and 5% of the target threshold, generate third adjustment information. When the mechanical performance index exceeds 5% of the target threshold, stop the charging authorization action of the charging pile.

[0036] Furthermore, the first adjustment information includes:

[0037] Determine the current operating temperature or charging current gear of the charging pile;

[0038] Adjust the operating temperature or charging current gear of the charging pile at the next moment;

[0039] Or, determine whether the current operating temperature exceeds 5% - 10% of the target threshold;

[0040] If the current operating temperature does not exceed 5% - 10% of the target threshold, then reduce or increase the current charging current gear by at least one charging current gear;

[0041] Determine the charging current gear of the charging pile at the next moment.

[0042] Furthermore, the second adjustment information includes:

[0043] Cool or heat the environment of the charging pile;

[0044] Or, dehumidify or humidify the charging pile.

[0045] Further, the third adjustment information includes:

[0046] Adding a shock absorption design to the charging pile or reducing the external load on the charging pile.

[0047] Further, it further includes:

[0048] A resource allocation module that determines whether the operation data of a certain charging pile meets the first condition. If it meets, the charging pile is disabled during the corresponding period, and other charging piles are allocated for users to use;

[0049] The first condition includes that the usage times of the charging pile exceed the first threshold and the call times exceed the second threshold.

[0050] The beneficial effects that this application can produce include:

[0051] The intelligent charging pile operation safety control system provided by this application realizes the comprehensive control of the operation state of the charging pile through real-time monitoring, data analysis and prediction, and the execution of safety strategies. This system not only improves the safety and stability of the charging pile, but also reduces the management cost and maintenance difficulty, providing a strong guarantee for the development of new energy vehicles. Description of the Drawings

[0052] Figure 1 It is a structural diagram of the intelligent charging pile operation safety control system in an embodiment of this application. Detailed Embodiment

[0053] The following describes this application in detail with reference to the embodiments, but this application is not limited to these embodiments.

[0054] See Figure 1 , an intelligent charging pile operation safety control system, characterized by including:

[0055] Multiple networked charging piles;

[0056] A charging pile monitoring module for monitoring the operation data of each charging pile, where the operation data includes electrical data, environmental data, and mechanical data;

[0057] A safety state control model that, based on the operation data of the charging pile, obtains the safety state control model of the charging pile through a neural network, simulates the stable state and safety change trend of the charging pile under different working conditions, and realizes the prediction of the operation state of the charging pile; at the same time, sets evaluation indicators for the operation data, sets target thresholds for the evaluation indicators, and presets safety strategies for the evaluation indicators; thereby judging the safety attributes of the charging pile and matching the corresponding safety strategies for the operation data;

[0058] The electrical data is used to ensure the electrical insulation performance of the charging pile, the environmental data is used to ensure that the charging pile meets different environmental standards, and the mechanical data is used to ensure the mechanical performance of the charging pile.

[0059] Specifically, the system includes multiple networked charging piles that can transmit and communicate data through the network to achieve remote monitoring and management. The charging pile monitoring module can monitor the operating data of each charging pile in real time, including electrical data, environmental data, and mechanical data. The safety status control model uses a neural network to process and analyze the operating data of the charging pile, simulates the stable state and safety change trend of the charging pile under different working conditions, and realizes the prediction of the operating state of the charging pile. The system sets evaluation indicators and target thresholds for the operating data, and presets safety strategies according to these evaluation indicators. When the operating data of the charging pile reaches or exceeds the target threshold, the system will automatically match the corresponding safety strategy to ensure the safe operation of the charging pile.

[0060] Among them, multiple networked charging piles provide charging services, transmit electric energy to electric vehicles, have network communication functions, and can upload operating data to the monitoring module in real time; the charging pile monitoring module monitors the operating data of each charging pile in real time. Electrical data such as current and power is used to ensure the electrical insulation performance and charging efficiency of the charging pile. Environmental data such as temperature and humidity is used to ensure the stability and safety of the charging pile under different environmental conditions. Mechanical data such as vibration and noise is used to evaluate the mechanical performance and durability of the charging pile. The safety status control model simulates the stable state and safety change trend of the charging pile through a neural network based on the operating data of the charging pile. Its working principle is specifically to receive the operating data uploaded by the charging pile monitoring module; use a neural network to process and analyze the data, establish a safety status control model for the charging pile; predict the operating state and safety change trend of the charging pile under different working conditions according to the model; set evaluation indicators and target thresholds, and match the corresponding safety strategy according to the operating data.

[0061] Through real-time monitoring and intelligent prediction, the system can timely detect potential safety hazards of the charging pile, adopt corresponding safety strategies to ensure the safe operation of the charging pile, and the system can automatically match safety strategies to improve the use efficiency and operation efficiency of the charging pile.

[0062] The evaluation indicators of the operating data include:

[0063] Electrical performance indicators, which include operating temperature indicators and current indicators;

[0064] Environmental performance indicators, which include environmental temperature indicators and environmental humidity indicators;

[0065] Mechanical performance indicators, and the mechanical performance indicators include vibration indicators and displacement indicators.

[0066] Specifically, among the electrical performance indicators, temperature is one of the key factors affecting the performance and lifespan of the charging pile equipment. Excessive or too low temperature may lead to a decline in equipment performance or damage. The upper temperature limit specified in the charging pile national standard is 50°C, and the lower limit may vary according to specific standards (such as -30°C). During actual operation, the working temperature of the charging pile should be ensured within its designed range. Current is a basic indicator for the operation of the charging pile equipment, directly affecting the charging speed and safety. The current should be set according to the battery capacity and charging time, generally set between 0.5C - 1C (C is the battery capacity, unit is ampere-hour Ah). For example, for a 60Ah battery, the charging current is generally set between 30A - 60A.

[0067] Among the environmental performance indicators, the ambient temperature directly affects the working performance and lifespan of the charging pile. The charging pile should be able to work normally within a certain range of ambient temperature, such as -30°C to 50°C (the specific range may vary according to the standard). Too high or too low humidity may cause damage to the circuit board and components of the charging pile. The relative humidity requirement for the charging pile is usually between 5% - 95%, and the ideal working environment humidity is 30% - 50%.

[0068] Among the mechanical performance indicators, vibration affects the structural stability of the charging pile and the connection reliability of components. The charging pile should be able to withstand vibrations of a certain frequency and amplitude without affecting its normal operation. Displacement reflects the stability problem of the charging pile during installation or use. After the charging pile is installed, it should remain stable and should not show obvious displacement or shaking.

[0069] The construction method of the safety state control model includes:

[0070] Construct three model data sets according to electrical data, environmental data, and mechanical data respectively;

[0071] Divide the three model data sets into three types of basic models;

[0072] Among them, the three basic models are the electrical data basic model, the environmental data basic model, and the mechanical data basic model; the electrical data basic model can extract local features (such as harmonic distortion) of the current / voltage waveform through CNN, and GRU captures the time series dependence relationship. The input features are the normalized real-time current, voltage curve, and temperature gradient.

[0073] Specifically, the electrical data base model includes a data preprocessing module, a CNN (Convolutional Neural Network) feature extraction module, a GRU (Gated Recurrent Unit) time series modeling module, and a fully connected decision module. Among them, the data preprocessing module inputs real-time current, voltage, and temperature gradient time series data, independently performs Z-score (standard score) or Min-Max normalization on each channel of data, cuts the continuous data stream into fixed-length time windows (such as 100 sampling points), adds Gaussian noise or time warping to enhance the robustness of the model, and thus outputs standardized time series data with a shape of (batch_size, T, 3); the CNN feature extraction module inputs the preprocessed time series window (T, 3), captures local features such as harmonic distortion through multi-scale 1D convolutional kernels (kernel_size = 3, 5), performs max pooling downsampling (pool_size = 2) to retain the main features, and selects residual blocks to alleviate the problem of gradient disappearance at the same time, and thus outputs a compressed feature sequence (T', C), where T' is the number of time steps after compression and C is the number of feature channels; the GRU time series modeling module inputs the feature sequence (T', C) output by the CNN, captures the temporal dependencies before and after at the same time (hidden_units = 64 - 128), applies attention weights to the output of the GRU to strengthen the key time steps, and thus outputs a time series feature vector, configured to return the state at the last moment or the states at all moments; the fully connected decision module inputs the feature vector output by the GRU, and can use a multi-layer perceptron (MLP) for non-linear feature transformation and a linear activation output layer to output the prediction result.

[0074] The environmental data base model can predict environmental adaptability through the coupling relationship between temperature, humidity, and altitude (such as the insulation performance attenuation curve under high humidity), and the input feature is the sliding window mean of environmental sensor data.

[0075] Specifically, the environmental data base model inputs the multi-sensor sliding window mean sequence, with the shape of (batch_size, T, 3), where batch_size represents the number of samples selected for one training, T is the time window length (such as the 7-day mean sequence), and 3 corresponds to the means of temperature, humidity, and altitude. It predicts the parameters of the insulation performance decay curve, such as the decay coefficient and half-life, judges the environmental level, such as normal / warning / danger, and finally outputs the environmental adaptability index. Among them, the environmental data base model includes a data preprocessing module, a coupling relationship modeling module, and a non-linear decision module. The data preprocessing module calculates the sliding window mean of the original sensor data, and statistical quantities such as variance and maximum value can be superimposed. Second-order / third-order interaction terms of temperature, humidity, and altitude are generated through polynomial features. When the humidity > 80%, the altitude threshold adjustment is introduced, and Z-score normalization is performed according to the feature dimension to retain the distribution information. The problem of inconsistent multi-sensor sampling rates is solved through interpolation processing; the coupling relationship modeling module captures the dynamic evolution patterns of temperature, humidity-altitude, and processes long-range dependencies in parallel, meeting the real-time requirements of industrial scenarios. The two-way influences of temperature, humidity → altitude, altitude → temperature, and humidity are explicitly modeled to assist in explaining the change of coupling strength over time; the non-linear decision module adopts a residual network structure (ResNet) to alleviate the disappearance of deep gradients. The Squeeze-and-Excitation (SE) module is introduced to adaptively adjust the channel importance; multi-task output is performed to predict the parameters of the insulation performance decay curve and simultaneously predict the change rates of temperature and humidity and the abnormal probability of altitude.

[0076] The mechanical data base model can model the long-term trend of vibration signals through LSTM, and peak detection captures sudden mechanical shocks. The input features are vibration spectrum features, specifically the frequency domain energy distribution after FFT transformation.

[0077] Specifically, the continuous vibration signal is segmented into short-time frames of a fixed length, and overlaps are set between frames to avoid information loss. After frame segmentation, each time-domain frame is windowed to reduce spectral leakage. The FFT transform is performed on each frame of the signal to obtain the power spectral density (PSD). According to the frequency distribution of mechanical component fault characteristics, frequency bands are divided, and the energy of each frequency band is statistically calculated as a feature vector. The sliding window method is used to detect sudden changes in the signal amplitude, and the adaptive threshold is calculated, such as the mean + 3 times the standard deviation. Peak points that exceed the threshold and meet the minimum interval are marked, and the number of peaks within a unit time is statistically calculated as the impact intensity index. The proportion of high-frequency band energy is calculated, and a dynamic threshold is set to mark energy sudden increase events as potential fault precursors. Among them, a double-layer LSTM (Long Short-Term Memory) network is constructed. The input layer receives the frequency-domain feature sequence. The first layer of LSTM (64 units) captures short-term frequency-domain patterns, and the second layer of LSTM (32 units) extracts long-term trend features. The Dropout layer (0.2 ratio) prevents overfitting. The peak features are concatenated with the LSTM output and fused through a fully connected layer. The peak features are normalized and then input into the dense layer and concatenated with the LSTM output, and then output the health probability through the Sigmoid activation.

[0078] The cross-modal feature fusion of the outputs of the three types of base models is performed through a hierarchical attention mechanism to obtain a safety state control model.

[0079] Specifically, after obtaining the three model datasets, they are respectively input into the corresponding base models. The base models extract key features from their respective datasets and conduct a preliminary evaluation of the safety state based on these features. Among them, cross-modal feature fusion is a key step in the construction of the safety state control model. Since electrical, environmental, and mechanical data come from different data sources, they have different feature representations and semantic information. To make full use of these data, an effective fusion method is required. The hierarchical attention mechanism is an effective cross-modal feature fusion method. By applying the attention mechanism at different levels, it can adaptively select important feature information and suppress irrelevant or redundant information. Specifically, the hierarchical attention mechanism can first perform feature selection on the outputs of each base model, and then fuse these features at a higher level to obtain a more comprehensive and accurate safety state evaluation result.

[0080] The safety state control model constructed by the above method can comprehensively consider data in multiple aspects such as electricity, environment, and machinery, and conduct a comprehensive and accurate evaluation of the safety state of the system. It not only helps to detect and warn potential safety hazards in a timely manner, but also provides strong support for the maintenance and management of equipment. In addition, the model can be extended and optimized according to actual needs to adapt to the safety state evaluation requirements in different scenarios.

[0081] Furthermore, cross-modal feature fusion is performed on the outputs of the three types of base models through a hierarchical attention mechanism, including:

[0082] Align the time series of electrical data, environmental data, and mechanical data based on dynamic time warping;

[0083] Allocate cross-modal attention weights according to the physical space correlation between electrical data and mechanical data;

[0084] Among them, when calculating the cross-modal attention weights, the following is satisfied:

[0085] ;

[0086] In the formula, is the query matrix, is the key matrix, is the feature dimension, and the softmax function is used to convert the weight matrix into a probability distribution form;

[0087] Generate a gating coefficient using environmental data to dynamically adjust the cross-modal attention weights;

[0088] Fuse the time series aligned features with the weighted cross-modal features to generate a safety state representation.

[0089] Specifically, different data all have time series attributes. It is necessary to analyze that the dynamic laws of data with respect to charging cycles, environmental changes, or mechanical wear are all related to the overall operating state of the charging pile, and outliers may reflect potential safety hazards. Since electrical data, environmental data, and mechanical data have different sampling frequencies and time series characteristics, it is first necessary to align them based on dynamic time warping (DTW) or other time series alignment methods. DTW is a method for measuring the similarity between two time series, which can handle the non-linear distortion of time series on the time axis. Through alignment, it can be ensured that different data sources have comparable feature values at the same time point. In cross-modal feature fusion, the attention mechanism is used to dynamically allocate weights between different data sources to emphasize the features that are more important for safety state assessment. According to the physical space correlation between electrical data and mechanical data, a cross-modal attention mechanism can be designed to allocate weights.

[0090] Among them, calculate the spatial correlation weight between electrical data and mechanical data. During the calculation process, electrical is used as the source modality and mechanical data is used as the target modality, which is expressed as:

[0091] ;

[0092] In the formula, is the query matrix after linear transformation of the electrical data features; The key matrix after the linear transformation of mechanical data features, with a dimension of [d×d], is extracted from electrical features through linear transformation; d is the feature dimension; is the scaling factor, which is used to prevent the softmax gradient from vanishing due to an overly large dot product result;

[0093] The mechanical data is used as the source modality, and the electrical data is used as the target modality, which is expressed as:

[0094] ;

[0095] In the formula, is the query matrix after the linear transformation of mechanical data features; is the key matrix after the linear transformation of electrical data features, with a dimension of [d×d], which is extracted from mechanical features through linear transformation; d is the feature dimension; is the scaling factor, which is used to prevent the softmax gradient from vanishing due to an overly large dot product result;

[0096] Calculate the fused spatial correlation: ;

[0097] In the formula, represents the spatial attention matrix, with a dimension of [N×N], where N is the number of sensor nodes, represents element-wise multiplication.

[0098] Environmental data (such as temperature, humidity, etc.) may have an important impact on the operation of electrical and mechanical systems. Therefore, environmental data can be used to generate a gating coefficient for dynamically adjusting the cross-modal attention weights. The gating coefficient can be adjusted according to the real-time changes of environmental data, thereby achieving dynamic adjustment of the weights between different data sources.

[0099] Among them, environmental data can suppress unreliable attention weights. By generating a gating coefficient from environmental data, the fusion ratio of electrical and mechanical features can be controlled;

[0100] Gating coefficient: ;

[0101] In the formula, represents the gating network weight matrix, with a dimension of [2d×d], which maps the concatenated features to the gating space; represents the vector concatenation operation; represents the gating network bias vector, with a dimension of [d×1]; represents the environmental feature vector, with a dimension of [d×1], represents the preliminary fused feature, with a dimension of [d×1].

[0102] Dynamic weight adjustment:

[0103] ;

[0104] ;

[0105] In the formula, is the temperature coefficient, which controls the sharpness of the attention distribution; represents the gated attention weight matrix; represents the final attention weight matrix.

[0106] Finally, the time-aligned features are fused with the weighted cross-modal features. This can be achieved through simple concatenation, weighted summation, or other feature fusion methods. The fused feature vector contains comprehensive information of electrical, environmental, and mechanical data and can be used to generate an accurate representation of the safety state.

[0107] Among them, cross-modal weighted fusion is implemented to obtain the feature fusion weight: ;

[0108] In the formula, represents matrix multiplication; and respectively represent the electrical and mechanical feature vectors, with dimensions of [d×1];

[0109] The feature fusion weight is obtained, and finally the safety state representation is generated.

[0110] The safety state representation can be used in subsequent decision support, fault diagnosis, or warning systems to help users identify and handle potential safety hazards in a timely manner.

[0111] Set a target threshold for the evaluation index, including:

[0112] Input the charging pile feature data in the operation data into the safety state control model to obtain the target prediction value corresponding to the operation data;

[0113] Determine the first prediction value and the second prediction value of the target prediction value, and use the average value of the first prediction value and the second prediction value as the reference threshold;

[0114] If the change amount of the reference threshold is less than the first preset threshold, use the reference threshold as the target threshold;

[0115] Among them, the first prediction value is the maximum value in the target prediction value;

[0116] The second prediction value is the minimum value in the target prediction value;

[0117] The first preset threshold is used to represent the tolerance of the stability of the reference threshold.

[0118] Specifically, real-time operation data is obtained from the charging pile monitoring module. This data includes electrical data, environmental data, mechanical data, etc., which together constitute the characteristic data set of the charging pile. The collected characteristic data of the charging pile is input into a pre-trained safety status control model. This model is based on a neural network and can simulate the stable state and safety change trend of the charging pile under different working conditions. The safety status control model outputs corresponding predicted values according to the input characteristic data. This predicted value reflects the possible operating state of the charging pile in the future for a period of time.

[0119] Among them, in the obtained target predicted values, the maximum value is identified as the first predicted value, and the minimum value is identified as the second predicted value. These two predicted values respectively represent the highest and lowest operating states that the charging pile may reach. The average value of the first predicted value and the second predicted value is used as the reference threshold. This reference threshold represents an intermediate level that the charging pile may reach under normal operating conditions. If the change amount of the reference threshold is less than the first preset threshold, it indicates that the reference threshold is relatively stable and can be used as the target threshold. This means that when the operating state of the charging pile fluctuates near the target threshold, it can be considered safe.

[0120] It should be noted that the reference threshold is calculated continuously multiple times (for example, multiple predictions within a set time window), and its change amount is observed. The change amount reflects the stability of the reference threshold. The calculated change amount of the reference threshold is compared with the preset first preset threshold. The first preset threshold is a tolerance index used to measure the acceptable range of the change of the reference threshold.

[0121] Furthermore, the setting of the first preset threshold needs to be determined according to the actual operating conditions and safety requirements of the charging pile. Under certain working conditions, it should be small enough to ensure the stability of the target threshold; at the same time, under certain working conditions, it should be large enough to accommodate normal operating fluctuations. In actual applications, it is necessary to continuously monitor the operating data of the charging pile and adjust the target threshold according to the actual situation. This can ensure that the target threshold is always synchronized with the actual operating state of the charging pile. Through the above steps, a reasonable target threshold can be set for the evaluation index of the operating data of the charging pile, thereby realizing precise monitoring and management of the operating state of the charging pile.

[0122] A safety policy is preset for the evaluation index, including:

[0123] Compare the electrical performance index with the corresponding target threshold. When the electrical performance index is between 5 - 10% of the target threshold, generate the first adjustment information. When the electrical performance index exceeds 10% of the target threshold, stop the charging authorization action of the charging pile;

[0124] Compare the environmental performance index with the corresponding target threshold. When the environmental performance index is within 10 - 20% of the target threshold, generate second adjustment information. When the environmental performance index exceeds 20% of the target threshold, stop the authorized charging operation of the charging pile;

[0125] Compare the mechanical performance index with the corresponding target threshold. When the mechanical performance index is within 3 - 5% of the target threshold, generate third adjustment information. When the mechanical performance index exceeds 5% of the target threshold, stop the authorized charging operation of the charging pile.

[0126] Specifically, monitor the electrical performance indices of the charging pile in real time, such as the operating temperature index and the current index. Compare these indices with the pre - set target thresholds. When the electrical performance index is within the range of 5% - 10% of the target threshold, it indicates that the electrical performance of the charging pile starts to show slight anomalies but has not reached a dangerous level. At this time, the system generates first adjustment information to prompt the operation and maintenance personnel to pay attention and check the electrical system of the charging pile, and may take preliminary adjustment measures, such as adjusting the charging current or the heat dissipation strategy. When the electrical performance index exceeds 10% of the target threshold, it indicates that the electrical performance of the charging pile has seriously deviated from the normal range and there are potential safety hazards. At this time, the system immediately stops the authorized charging operation of the charging pile to prevent further safety risks.

[0127] Monitor the environmental temperature index and the environmental humidity index of the environment where the charging pile is located in real time. Compare these indices with the pre - set target thresholds. When the environmental performance index is within the range of 10% - 20% of the target threshold, it indicates that the environment where the charging pile is located begins to become unfavorable for its normal operation. At this time, the system generates second adjustment information to prompt the operation and maintenance personnel to pay attention to the environmental changes and may take environmental control measures, such as turning on a dehumidifier, etc. When the environmental performance index exceeds 20% of the target threshold, it indicates that the environment where the charging pile is located has seriously deteriorated and cannot guarantee its safe operation. At this time, the system also immediately stops the authorized charging operation of the charging pile.

[0128] Monitor the mechanical performance indices of the charging pile in real time, such as the vibration index and the displacement index. Compare these indices with the pre - set target thresholds. When the mechanical performance index is within the range of 3% - 5% of the target threshold, it indicates that the mechanical performance of the charging pile starts to show slight fluctuations or anomalies. At this time, the system generates third adjustment information to prompt the operation and maintenance personnel to check the mechanical structure of the charging pile and may take reinforcement or adjustment measures. When the mechanical performance index exceeds 5% of the target threshold, it indicates that the mechanical performance of the charging pile has been seriously damaged and there are potential safety hazards. At this time, the system also immediately stops the authorized charging operation of the charging pile.

[0129] Among them, in practical applications, the specific parameters of the safety policy (such as 5%-10%, 10%-20%, 3%-5%, etc.) may need to be adjusted according to the specific model of the charging pile, the usage environment, and the operation and maintenance experience. The system should be able to continuously monitor the operation data of the charging pile and generate adjustment information or stop the charging authorization in a timely manner according to the safety policy. At the same time, the operation and maintenance personnel should be able to quickly respond to the prompts and alarms of the system to ensure the safe operation of the charging pile. By implementing the above safety policy, accurate monitoring and management of the operation status of the charging pile can be achieved, effectively preventing potential safety risks.

[0130] The first adjustment information includes:

[0131] Determine the current operating temperature or charging current level of the charging pile;

[0132] Adjust the operating temperature or charging current level of the charging pile at the next moment;

[0133] Or, determine whether the current operating temperature exceeds 5-10% of the target threshold;

[0134] If the current operating temperature does not exceed 5-10% of the target threshold, then reduce or increase the current charging current level by at least one charging current level;

[0135] Determine the charging current level of the charging pile at the next moment.

[0136] Specifically, in one implementation, the system continuously monitors the current operating temperature or charging current level of the charging pile. According to the maximum / minimum operating temperature limit, maximum / minimum charging current limit, etc., adjust the operating temperature or charging current level of the charging pile at the next moment. When adjusting the temperature, according to the actual situation, the system adjusts the rotation speed of the cooling fan, enables or disables additional cooling equipment, etc. to adjust the operating temperature.

[0137] In another implementation, compare these real-time data with the pre-set target threshold. At the current operating temperature, if the current charging current level does not match the current range related to the target threshold (for example, the current is too large or too small), the system can determine whether it is necessary to adjust the charging current level. The system adjusts the current controller inside the charging pile or sends an instruction to the battery management system (BMS) of the electric vehicle to adjust the charging current level.

[0138] Among them, the system first determines whether the current operating temperature exceeds 5-10% of the target threshold. If the current operating temperature does not exceed 5-10% of the target threshold, but the charging current gear is relatively high (which may lead to the risk of overheating), the system considers reducing the current charging current gear by at least one gear. On the contrary, if the current operating temperature is within the safe range and the charging current gear is relatively low (which may lead to low charging efficiency), the system considers increasing the current charging current gear by at least one gear. After completing the adjustment of the charging current gear, the system determines the charging current gear of the charging pile at the next moment and sends a corresponding instruction to the charging pile or the BMS of the electric vehicle.

[0139] The second adjustment information includes:

[0140] Cool or heat-insulate the environment of the charging pile;

[0141] Or, dehumidify or humidify the charging pile.

[0142] Specifically, the second adjustment information makes necessary adjustments to the environment where the charging pile is located according to the environmental performance indicators to ensure its operation under suitable environmental conditions. The system monitors the temperature of the environment where the charging pile is located in real time and compares it with the preset target threshold. If the environmental temperature is higher than a certain range of the target threshold (such as exceeding 10%-20% of the target threshold), the system determines that cooling treatment is required. The cooling measures include starting the air-conditioning equipment, fans or spray systems near the charging pile to lower the environmental temperature. In some cases, the system may also assist in cooling by adjusting the heat dissipation strategy of the charging pile itself. If the environmental temperature is lower than a certain range of the target threshold, the system determines that heat-insulating treatment is required. The heat-insulating measures include closing the ventilation equipment near the charging pile, enabling heating equipment or taking other heat-insulating measures.

[0143] The system monitors the humidity of the environment where the charging pile is located in real time and compares it with the preset target threshold. If the environmental humidity is higher than a certain range of the target threshold, the system determines that dehumidification treatment is required. The dehumidification measures include starting a dehumidifier, increasing the ventilation volume or using other dehumidification means. If the environmental humidity is lower than a certain range of the target threshold, the system determines that humidification treatment is required. The humidification measures may include using a humidifier, reducing the ventilation volume or taking other humidification means.

[0144] The system selects an appropriate adjustment strategy and executes it according to the real-time monitoring results of the environmental temperature and humidity. The execution of the adjustment strategy involves the control system of the charging pile itself and also requires linkage with other environmental control systems (such as air conditioners, dehumidifiers, etc.).

[0145] The third adjustment information includes:

[0146] Add shock-absorbing design to the charging pile or reduce the external load on the charging pile.

[0147] Specifically, the third adjustment information enhances the structural stability and safety of the charging pile, mainly by adding shock absorption design or reducing the external load on the charging pile. First, identify the key parts of the charging pile that are vulnerable to vibration or impact, such as connectors, support structures, etc. Introduce shock absorption materials at the key parts, such as rubber pads, spring shock absorbers, etc., to absorb and disperse vibration energy. At the same time, consider using shock absorption methods such as dynamic shock absorption systems to improve the shock absorption effect. Optimize the overall structure design of the charging pile, increase redundant supports, improve the stiffness and stability of the structure, ensure that the shock absorption design is coordinated with the overall structure of the charging pile, and avoid introducing new safety hazards. Second, conduct a comprehensive assessment of the external loads borne by the charging pile, including wind loads, human factors, etc. According to the assessment results, determine which loads can be reduced through design or management measures. Reduce the external load on the charging pile through optimized design. For example, adopt a streamlined design to reduce wind resistance, or increase snow protection measures to reduce snow load. Develop management measures to limit the external load on the charging pile to ensure that it does not cause additional load on the charging pile.

[0148] It also includes:

[0149] A resource allocation module that determines whether the operating data of a certain charging pile meets the first condition. If it meets, the charging pile is disabled during the corresponding period, and other charging piles are allocated for users to use;

[0150] The first condition includes that the usage times of the charging pile exceed the first threshold and the call times exceed the second threshold.

[0151] Specifically, the resource allocation module understands the current operating status, usage frequency, and potential problems of the charging pile based on the operating data of each charging pile, including key indicators such as usage times, call times, charging power, and charging duration. According to the preset first condition, the resource allocation module will determine whether a certain charging pile meets the disabling standard. The first condition includes two key elements: that the usage times exceed the first threshold indicates that the charging pile has been frequently used in a short period, and it may have reached the limit of its design life or is close to the state requiring maintenance. That the call times exceed the second threshold indicates that although the charging pile has been frequently requested for use, due to various reasons (such as faults, maintenance, occupancy, etc.), it has not been able to successfully meet all charging demands. When these two conditions are met simultaneously, the resource allocation module will consider that the charging pile has a relatively high safety risk or maintenance requirement, so it needs to be considered disabled during the corresponding period until the first condition is lifted.

[0152] It should be noted that once it is determined that a certain charging pile meets the disabling conditions, the resource allocation module will immediately mark it as unavailable and announce or notify it in the system. At the same time, the module will intelligently reallocate other available charging piles for users according to factors such as the current distribution of charging piles, user needs, and charging efficiency. Ensure that users can quickly find available charging piles when they need to charge, avoiding situations such as charging interruptions or overly long waiting times. After disabling the charging pile, the resource allocation module will continuously monitor its status and the usage of the charging piles reallocated to users; if it is found that the status of the disabled charging pile has improved (such as returning to normal after repair), or the reallocated charging piles cannot meet user needs (such as overly long queuing times, low charging efficiency, etc.), the module will adjust the strategy in a timely manner to ensure the reasonable allocation and efficient utilization of resources.

[0153] The above are only several embodiments of the present application, and do not impose any form of limitation on the present application. Although the present application is disclosed above with preferred embodiments, it is not intended to limit the present application. Any person skilled in the art, without departing from the scope of the technical solution of the present application, makes some changes or modifications using the technical content disclosed above, which are equivalent to equivalent implementation cases and all fall within the scope of the technical solution.

Claims

1. Intelligent charging pile operation safety control system, characterized in that, Including: Multiple connected charging piles; A charging pile monitoring module for monitoring the operation data of each charging pile, where the operation data includes electrical data, environmental data, and mechanical data; A safety status control model for obtaining the safety status control model of the charging pile based on the operation data of the charging pile through a neural network, simulating the stable state and safety change trend of the charging pile under different working conditions, and realizing the prediction of the operation state of the charging pile; at the same time, setting evaluation indicators for the operation data, setting target thresholds for the evaluation indicators, and presetting safety strategies for the evaluation indicators; thereby judging the safety attributes of the charging pile and matching the corresponding safety strategies for the operation data; The electrical data is used to ensure the electrical insulation performance of the charging pile, the environmental data is used to ensure that the charging pile meets different environmental standards, and the mechanical data is used to ensure the mechanical performance of the charging pile; The construction method of the safety status control model includes: Constructing three model data sets according to electrical data, environmental data, and mechanical data respectively; Dividing the three model data sets into three types of base models; Performing cross-modal feature fusion on the outputs of the three types of base models through a hierarchical attention mechanism to obtain a safety status control model; Performing cross-modal feature fusion on the outputs of the three types of base models through a hierarchical attention mechanism, including: Aligning the time series of electrical data, environmental data, and mechanical data based on dynamic time warping; Allocating cross-modal attention weights according to the physical space correlation between electrical data and mechanical data; Among them, when calculating the cross-modal attention weights, it satisfies: In the formula, Q is the query matrix, K is the key matrix, d is the feature dimension, and the softmax function is used to convert the weight matrix into a probability distribution form; Generating a gating coefficient using environmental data to dynamically adjust the cross-modal attention weights; Fusing the time series aligned features with the weighted cross-modal features to generate a safety status representation.

2. The intelligent charging pile operation safety control system according to claim 1, wherein The evaluation indicators of the operation data include: Electrical performance indicators, where the electrical performance indicators include operating temperature indicators and current indicators; Environmental performance indicators, where the environmental performance indicators include environmental temperature indicators and environmental humidity indicators; Mechanical performance indicators, where the mechanical performance indicators include vibration indicators and displacement indicators.

3. The intelligent charging pile operation safety control system according to claim 2, wherein, Setting target thresholds for the evaluation indicators, including: Inputting the charging pile feature data in the operation data into the safety status control model to obtain the target prediction value corresponding to the operation data; Determining the first prediction value and the second prediction value of the target prediction value, and taking the average of the first prediction value and the second prediction value as the reference threshold; If the change amount of the reference threshold is less than the first preset threshold, taking the reference threshold as the target threshold; Among them, the first prediction value is the maximum value in the target prediction value; The second prediction value is the minimum value in the target prediction value; The first preset threshold is used to represent the tolerance of the stability of the reference threshold.

4. The intelligent charging pile operation safety control system according to claim 3, wherein, Presetting safety strategies for the evaluation indicators, including: Compare the electrical performance index with the corresponding target threshold. When the electrical performance index is within 5-10% of the target threshold, generate first adjustment information. When the electrical performance index exceeds 10% of the target threshold, stop the charging authorization action of the charging pile; Compare the environmental performance index with the corresponding target threshold. When the environmental performance index is within 10-20% of the target threshold, generate second adjustment information. When the environmental performance index exceeds 20% of the target threshold, stop the charging authorization action of the charging pile; Compare the mechanical performance index with the corresponding target threshold. When the mechanical performance index is within 3-5% of the target threshold, generate third adjustment information. When the mechanical performance index exceeds 5% of the target threshold, stop the charging authorization action of the charging pile.

5. The intelligent charging pile operation safety control system according to claim 4, wherein The first adjustment information includes: Determine the current operating temperature or charging current level of the charging pile; Adjust the operating temperature or charging current level of the charging pile at the next moment; Or, judge whether the current operating temperature exceeds 5-10% of the target threshold; If the current operating temperature does not exceed 5-10% of the target threshold, reduce or increase the current charging current level by at least one charging current level; Determine the charging current level of the charging pile at the next moment.

6. The intelligent charging pile operation safety control system according to claim 4, wherein The second adjustment information includes: Cool or heat-insulate the environment of the charging pile; Or, dehumidify or humidify the charging pile.

7. The intelligent charging pile operation safety control system according to claim 4, wherein The third adjustment information includes: Add shock absorption design to the charging pile or reduce the external load on the charging pile.

8. The intelligent charging pile operation safety control system according to claim 1, wherein, It further includes: A resource allocation module that judges whether the operation data of a certain charging pile meets the first condition. If it meets, disable the charging pile during the corresponding period and allocate other charging piles for users to use; The first condition includes that the usage times of the charging pile exceed the first threshold and the call times exceed the second threshold.

Images