New energy charging pile charging process early warning system based on artificial intelligence

Abstract

The application discloses a new energy charging pile charging process early warning system based on artificial intelligence, which comprises a dynamic charging parameter real-time acquisition and multi-dimensional analysis unit, a feature depth mining and space-time correlation mapping unit, a charging state evaluation and abnormal trend prediction unit, a multi-source abnormal information fusion and hierarchical determination unit, an intelligent early warning strategy generation and dynamic adjustment unit, and an early warning information real-time pushing and execution feedback unit. The dynamic charging parameter real-time acquisition and multi-dimensional analysis unit acquires multiple parameters, which are processed by the feature depth mining and space-time correlation mapping unit, and then the charging state evaluation and abnormal trend prediction unit evaluates the state and predicts the trend, the multi-source abnormal information fusion and hierarchical determination unit performs fusion and hierarchical determination, the intelligent early warning strategy generation and dynamic adjustment unit generates an adjustment strategy, and the early warning information real-time pushing and execution feedback unit pushes information and receives feedback. The system is suitable for various scenes and improves charging safety and stability.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of new energy charging pile charging, and particularly to a new energy charging pile charging process early warning system based on artificial intelligence. BACKGROUND

[0002] With the rapid increase in the popularity of new energy vehicles, the safety and stability of the charging process of charging piles, as key supporting facilities, have attracted increasing attention. During the charging process, voltage fluctuations, current surges, abnormal temperature rises of batteries and pile bodies, and other situations frequently occur, which can easily cause equipment failure, battery damage, and even fire risks. Traditional methods relying on manual inspection or simple sensor monitoring have been unable to cope with complex and variable charging scenarios, and cannot accurately capture potential abnormalities in real time. Under this background, integrating artificial intelligence technology into the charging process of charging piles and analyzing multi-dimensional parameters through neural network models has become an important direction for improving the efficiency and accuracy of early warning.

[0003] The prior art has obvious deficiencies in the early warning of the charging process of charging piles. On the one hand, the correlation between feature extraction and state evaluation is weak, and most systems only analyze single parameters or simple combinations of parameters, failing to deeply mine the spatio-temporal correlation between multiple parameters such as voltage, current, and temperature, resulting in one-sided evaluation of the charging state and difficulty in identifying potential complex abnormalities in advance. On the other hand, the dynamic adaptability of the abnormality early warning strategy is insufficient, and the early warning threshold and strategy are mostly fixed, without considering the differences in different charging pile models, battery types, and environmental conditions. When the charging scenario changes, early warning may be delayed or false alarms may occur, and the early warning logic cannot be adjusted in a timely manner according to real-time feedback. SUMMARY

[0004] In order to overcome the shortcomings and deficiencies of the prior art, the present application provides a new energy charging pile charging process early warning system based on artificial intelligence.

[0005] The technical scheme adopted by the application is a new energy charging pile charging process early warning system based on artificial intelligence, comprising: a dynamic charging parameter real-time acquisition and multi-dimensional analysis unit for acquiring charging pile output voltage, charging current, battery temperature, charging time, pile body temperature and power grid fluctuation frequency; a feature deep mining and space-time correlation mapping unit, the input end of which is connected with the output end of the dynamic charging parameter real-time acquisition and multi-dimensional analysis unit, and the above parameters are subjected to feature extraction and space-time correlation analysis; a charging state evaluation and abnormal trend prediction unit, the input end of which is connected with the output end of the feature deep mining and space-time correlation mapping unit, and the charging state is evaluated and the abnormal trend is predicted based on the extracted features; a multi-source abnormal information fusion and hierarchical judgment unit, the input end of which is connected with the output end of the charging state evaluation and abnormal trend prediction unit, and the abnormal information is fused and graded; an intelligent early warning strategy generation and dynamic adjustment unit, the input end of which is connected with the output end of the multi-source abnormal information fusion and hierarchical judgment unit, and the early warning strategy is generated and adjusted according to the grading result; an early warning information real-time pushing and execution feedback unit, the input end of which is connected with the output end of the intelligent early warning strategy generation and dynamic adjustment unit, and the early warning information is pushed and the execution feedback is received, and the output end of the unit is connected with the dynamic charging parameter real-time acquisition and multi-dimensional analysis unit, the feature deep mining and space-time correlation mapping unit in a closed loop.

[0006] Further, the feature deep mining and space-time correlation mapping unit adopts an improved feature extraction model, and the model formula is: wherein, is the fusion feature value at t moment, , , is the adaptive weight coefficient, is the feature extraction result of the long short-term memory neural network optimized adaptively for voltage and current at t moment, is the battery temperature at t moment, is the temperature influence function of the pile body temperature , is the charging time, is the frequency influence function of the power grid fluctuation frequency ; the charging state evaluation and abnormal trend prediction unit adopts a state evaluation model, and the formula is: wherein, is the charging state evaluation value at t moment, is an improved grey wolf optimization algorithm processing function, is a back propagation neural network function, is a weight matrix, is a bias term.

[0007] Further, the feature deep mining and space-time correlation mapping unit further comprises a parameter dynamic adaptation subunit, and an adaptive model formula adopted by the parameter dynamic adaptation subunit is as follows: wherein, is a network parameter adaptation value at t, is an inverse function of an adaptive optimization long short-term memory neural network, are partial derivatives of the fusion feature value with respect to voltage and current respectively, are change amounts of voltage and current at t respectively; the state of charge evaluation and abnormal trend prediction unit further comprises an abnormal sensitivity adjustment subunit, and a model formula of the abnormal sensitivity adjustment subunit is as follows: wherein, is an abnormal sensitivity at t, is an improved grey wolf optimization algorithm processing function, is a state of charge evaluation value at t, is a derivative function of a back propagation neural network, are temperature change amounts of a battery and a pile body respectively, is a temperature influence coefficient.

[0008] Further, the dynamic charging parameter real-time acquisition and multi-dimensional analysis unit adopts a parameter analysis model, and a formula of the parameter analysis model is as follows: wherein, is a multi-dimensional analysis result at t, is a parameter category number, is an analysis coefficient of the i-th parameter, is an acquisition value of the i-th parameter at t, is a time attenuation function of the i-th parameter; and a feature enhancement model of the feature deep mining and space-time correlation mapping unit is as follows: wherein, is an enhanced feature value, is an enhancement coefficient, is an activation function, are multi-dimensional analysis results at t and t respectively, , is a fusion feature value at t, is a feature extraction result of the adaptive optimization long short-term memory neural network at t and t respectively.

[0009] Further, an abnormal trend prediction model of the state of charge evaluation and abnormal trend prediction unit is as follows: wherein, is a predicted state value at t, is a state of charge evaluation value at t, is a predicted state value at t, ​​​Processing function for the improved gray wolf optimization algorithm, is the back propagation neural network function, is the gradient of the state evaluation value at time t, is the prediction time interval, is the coefficient of fluctuation, is the parameter change vector at time t; the fusion model of the multi-source abnormal information fusion and classification judgment unit is: ,in, is the abnormal fusion value at time t, To predict the number of steps, is the weight of the j-th step prediction, is the temperature correction function corresponding to the j-th step prediction, is the battery temperature at time t, is the temperature of the pile at time t.

[0010] Furthermore, the strategy generation model of the intelligent early warning strategy generation and dynamic adjustment unit is: ,in, is the early warning strategy at time t, Based on the abnormal fusion value and classification thresholds The policy function, is the adjustment coefficient, To adaptively optimize the long short-term memory neural network function, for The strategy adjustment amount at the moment; the feedback processing model of the real-time push of the warning information and the execution feedback unit is: ,in, is the feedback processing result at time t, For execution results The feedback function, is the feedback coefficient, First, the product of the early warning strategy Str (t) and the feedback coefficient κ is processed by the back propagation BP algorithm, and then the result is input into the improved gray wolf optimization The algorithm performs nested operations for optimization calculations.

[0011] Further, the charging state evaluation and abnormal trend prediction unit comprises a state feature matching subunit that receives feature data output from the feature deep mining and space-time correlation mapping unit, compares the feature data with a preset normal charging state feature library one by one, determines the matching degree by calculating the Euclidean distance between the feature vectors, and screens out feature combinations with a matching degree lower than a set threshold; an abnormal trend quantification subunit that performs time series analysis on the screened low-matching-degree feature combinations, calculates the slope and curvature of the features with respect to the charging time, and combines the change rates of the battery temperature and the pile body temperature to convert the abnormal trend into quantifiable numerical indicators; an optimization evaluation subunit that optimizes the evaluation results of the back propagation neural network by using the improved grey wolf optimization algorithm, adjusts the weight and bias parameters of the network, and reduces the deviation of the evaluation results from the actual charging state to the minimum range; and a result output subunit that organizes the optimized charging state evaluation results and abnormal trend prediction indicators according to a set data format and transmits them to the multi-source abnormal information fusion and hierarchical judgment unit.

[0012] Further, the multi-source abnormal information fusion and hierarchical judgment unit comprises an abnormal information receiving subunit that receives various types of abnormal information output from the charging state evaluation and abnormal trend prediction unit, including abnormal features, trend indicators and evaluation scores, and checks the integrity and format of the information to eliminate invalid information; a multi-dimensional fusion subunit that classifies abnormal information from different sources according to the time dimension and the feature dimension, uses the weighted average method to fuse the same type of abnormal information at the same time point, and generates a comprehensive abnormal vector; a hierarchical threshold determination subunit that dynamically sets abnormal thresholds of different levels according to the model of the charging pile, the type of the battery and the charging stage, and sets a warning level corresponding to each threshold; and a hierarchical result output subunit that compares the fused abnormal information with the set threshold, determines the corresponding warning level, and transmits the hierarchical result to the intelligent early warning strategy generation and dynamic adjustment unit.

[0013] Further, the intelligent early warning strategy generation and dynamic adjustment unit comprises: a basic strategy library calling subunit that retrieves a corresponding initial early warning strategy from a preset basic early warning strategy library according to the early warning level output by the multi-source abnormal information fusion and hierarchical determination unit, wherein the initial strategy includes an early warning mode, a push object and a processing suggestion; a strategy dynamic adjustment subunit that adjusts the initial strategy by using an adaptive optimization long short-term memory neural network in combination with the feedback results returned by the early warning information real-time pushing and execution feedback unit, and corrects the early warning frequency and the push content, so that the strategy is more suitable for the actual situation; a multi-scene adaptation subunit that analyzes the environmental parameters of the current charging scene, including the environmental temperature, humidity and power grid stability, and performs scene adaptation optimization on the adjusted strategy to ensure consistent early warning effect in different scenes; and a final strategy output subunit that transmits the final early warning strategy after dynamic adjustment and scene adaptation to the early warning information real-time pushing and execution feedback unit, and stores the strategy parameters in a historical database.

[0014] The new energy charging pile charging process early warning system based on artificial intelligence comprises:

[0015] Step S1: The dynamic charging parameter real-time acquisition and multi-dimensional analysis unit continuously acquires and multi-dimensionally analyzes the output voltage, charging current, battery temperature, charging time, pile body temperature and power grid fluctuation frequency of the charging pile;

[0016] Step S2: The analyzed parameters are transmitted to the feature depth mining and space-time correlation mapping unit for feature depth mining and space-time correlation mapping analysis to form a feature data set;

[0017] Step S3: The feature data set is input into the charging state evaluation and abnormal trend prediction unit to develop charging state evaluation and abnormal trend prediction, and obtain abnormal trend information;

[0018] Step S4: The abnormal trend information is sent to the multi-source abnormal information fusion and hierarchical determination unit to implement multi-source abnormal information fusion and hierarchical determination, and generate an abnormal hierarchical result;

[0019] Step S5: The abnormal hierarchical result is transmitted to the intelligent early warning strategy generation and dynamic adjustment unit to develop an intelligent early warning strategy and perform dynamic adjustment, and output an adapted early warning strategy;

[0020] Step S6: The adapted early warning strategy is delivered to the early warning information real-time pushing and execution feedback unit to perform early warning information real-time pushing and execution feedback receiving, and the feedback information is respectively fed back to the dynamic charging parameter real-time acquisition and multi-dimensional analysis unit and the feature depth mining and space-time correlation mapping unit to form a complete early warning closed-loop process.

[0021] Beneficial effects: The present invention proposes an artificial intelligence-based new energy charging pile charging process warning system. The system obtains comprehensive parameters through real-time collection of dynamic charging parameters and a multi-dimensional analysis unit. The system deeply mines the spatiotemporal correlation of multiple parameters through feature deep mining and spatiotemporal correlation mapping units. The charging state assessment and abnormal trend prediction unit then accurately assesses the state and predicts the trend. The multi-source abnormal information fusion and classification judgment unit fuse the classified abnormal information. The intelligent early warning strategy generation and dynamic adjustment unit generates and adjusts the strategy. Finally, the early warning information is pushed in real time and the execution feedback unit pushes the information and forms a closed loop, significantly improving the accuracy and timeliness of the early warning. At the same time, the system uses an adaptively optimized long short-term memory neural network to conduct in-depth mining and spatiotemporal correlation analysis of multi-dimensional parameters. The improved gray wolf optimization-back propagation neural network performs state assessment and abnormal trend prediction based on this, strengthens the correlation between the two, and can identify complex abnormalities. In response to the problem of insufficient dynamic adaptability of the early warning strategy, the intelligent early warning strategy generation and dynamic adjustment unit dynamically adjusts the strategy based on the abnormal classification results and feedback information to adapt to different scenarios and avoid early warning lags or false alarms. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 It is a diagram of the system unit composition of the present invention;

[0023] Figure 2 This is a flow chart of the system operation of the present invention. DETAILED DESCRIPTION

[0024] It should be noted that, unless there is a conflict, the embodiments in this application and the features in the embodiments can be combined with each other. The application is further described in detail below with reference to the drawings and specific embodiments.

[0025] like Figure 1 As shown in the figure, the artificial intelligence-based new energy charging pile charging process early warning system includes:

[0026] Dynamic charging parameter real-time acquisition and multi-dimensional analysis unit, used to obtain charging pile output voltage, charging current, battery temperature, charging time, pile body temperature and grid fluctuation frequency;

[0027] Specifically, the real-time collection and multi-dimensional analysis unit for dynamic charging parameters serves as the data input core of the entire early warning system. Its role is to provide comprehensive and accurate raw data support for subsequent feature mining, status assessment, and other steps. The technical parameters covered by this unit include charging pile output voltage (typically ranging from 200V to 1000V), charging current (generally 10A to 200A), battery temperature (normally between 25°C and 45°C), charging time (ranging from a few minutes to several hours), pile body temperature (typically no more than 60°C), and grid fluctuation frequency (approximately 50Hz, with an allowable fluctuation range of ±0.5Hz). These parameters directly reflect the power transmission status, equipment operating conditions, and real-time battery response during the charging process. Their accuracy and timeliness are crucial to the performance of the entire early warning system. The collection and analysis of these multi-dimensional parameters provides the system with a rich information foundation, enabling more accurate detection of potential anomaly risks.

[0028] The specific implementation process of this unit is as follows: First, various sensors deployed inside the charging pile and on the connecting lines collect electrical parameters such as output voltage and charging current in real time at sampling intervals of 10ms-50ms. Simultaneously, temperature sensors installed on the battery pack surface and the charging pile housing collect battery and pile body temperatures at sampling intervals of 1s-5s. Grid fluctuation frequency is collected by a frequency monitoring module connected to the grid at sampling intervals of 0.1s. After filtering, the collected raw data is transmitted to the analysis module. According to preset rules, the analysis module converts voltage and current data into derived parameters such as power and power, temperature data into characteristic quantities such as temperature change rate, and grid fluctuation frequency into indicators such as fluctuation amplitude. After analysis, all parameters are integrated according to timestamps to form a structured dataset. This dataset is sent to the feature deep mining and spatiotemporal correlation mapping unit via the data transmission interface and backed up to a local storage device for 30 days to meet the needs of data traceability and system optimization.

[0029] A feature deep mining and spatiotemporal correlation mapping unit, the input of which is connected to the output of the dynamic charging parameter real-time acquisition and multi-dimensional analysis unit, and performs feature extraction and spatiotemporal correlation analysis on the above parameters;

[0030] Specifically, the feature deep mining and spatio-temporal correlation mapping unit is the key link for the system to process data and extract features. Its core function is to deeply mine the multi-dimensional parameters transmitted by the real-time dynamic charging parameter collection and multi-dimensional analysis unit, extract information that can reflect the essential characteristics of the charging process, and establish the spatio-temporal correlation between these features. The adaptive optimization long short-term memory neural network used in this unit can effectively solve the gradient vanishing or gradient explosion problem of traditional neural networks when processing time series data by optimizing the gating mechanism and memory unit inside the network, and is particularly suitable for parameter analysis with time dependence in the charging process. Through feature mining and spatio-temporal correlation mapping of parameters such as voltage, current, and temperature, the original and scattered parameter data can be converted into feature vectors with clear physical meaning and strong correlation, providing high-quality input for subsequent state assessment and abnormal prediction, and significantly improving the system's ability to recognize complex abnormal patterns.

[0031] The specific implementation process of this unit is as follows: First, receive the structured data set sent by the real-time dynamic charging parameter collection and multi-dimensional analysis unit, and perform time alignment processing on the data to ensure that all parameters are consistent in the time dimension. Then, divide the processed data set into multiple data windows according to the time sequence, and set the length of each window to 10s-60s according to the stability of the charging process. Next, input each data window into the adaptive optimization long short-term memory neural network, and the network extracts features such as voltage variation trend, current fluctuation rule, and temperature rise rate through the coordinated action of input gate, forget gate, and output gate, while learning the correlation between different parameters, such as the phase relationship between voltage and current, and the dynamic response relationship between battery temperature and charging current. During network training, the network parameters are continuously adjusted through the backpropagation algorithm, so that the network can adapt to parameter features under different charging pile models, different battery types, and different environmental conditions. The extracted feature vectors and spatio-temporal correlation information are standardized to form a fixed-dimensional feature matrix, which is transmitted to the charging state assessment and abnormal trend prediction unit, and the key parameters in the feature extraction process, such as the activation function output value of the network and the memory cell state, are recorded for subsequent system debugging and performance optimization.

[0032] The charging state assessment and abnormal trend prediction unit is connected to the output end of the feature deep mining and spatio-temporal correlation mapping unit, and assesses the charging state and predicts abnormal trends based on the extracted features.

[0033] Specifically, the charging state evaluation and abnormal trend prediction unit undertakes the important responsibilities of accurate evaluation of the charging state and early prediction of abnormal trends in the system. The unit optimizes the initial weights and biases of the back propagation neural network using the improved grey wolf optimization algorithm, effectively solving the problems of slow convergence speed and easy falling into local optimum of traditional back propagation neural network, and improving the learning efficiency and generalization ability of the network. By taking the feature vectors extracted by the feature depth mining and spatio-temporal correlation mapping unit as input, the unit can quantitatively evaluate the normal state, slight abnormal state and serious abnormal state during charging, and predict the possible abnormal situation in advance 5-30min according to the historical data and current trend. This evaluation and prediction ability enables the system to issue an early warning before the abnormality occurs, which saves valuable time for taking preventive measures, thereby reducing the probability of equipment damage and safety accidents.

[0034] The specific implementation process of the unit is as follows: first, receive the feature matrix transmitted by the feature depth mining and spatio-temporal correlation mapping unit, divide it into training set and test set, where the training set accounts for 70%-80%, which is used for network training, and the test set is used to verify the network performance. Then, initialize the structure of the back propagation neural network, set the number of input layer nodes to the dimension of the feature matrix, set the number of hidden layer nodes to 20-50 according to the feature complexity, and set the number of output layer nodes to 3, corresponding to the evaluation values of normal state, slight abnormal state and serious abnormal state respectively. Next, the improved grey wolf optimization algorithm is used to optimize the initial weights and biases of the network. In the optimization process, the prediction error of the network is used as the fitness function, and the optimal parameter combination is found through the search, surround and attack behavior of the wolf group, and the iteration number is set to 50-100 times. After optimization, the network is trained using the training set, and the weights and biases are adjusted through the back propagation algorithm until the prediction error of the network is less than the preset threshold (usually 0.01-0.05). The trained network processes the test set, outputs the charging state evaluation value at each time point and the abnormal trend prediction result in the future period of time, and these results are sent to the multi-source abnormal information fusion and hierarchical judgment unit through the data interface, and the network evaluation and prediction records, including evaluation value, prediction time and error value, are stored for subsequent analysis and optimization.

[0035] Multi-source abnormal information fusion and hierarchical judgment unit, the input end of which is connected with the output end of the charging state evaluation and abnormal trend prediction unit, which fuses and classifies the abnormal information;

[0036] Specifically, the multi-source abnormal information fusion and hierarchical judgment unit is an important part of the system for integrating abnormal information and classifying risk levels. Its main function is to fuse the abnormal trend prediction results output by the charging state evaluation and abnormal trend prediction unit in multiple dimensions and multiple levels, and to classify the abnormal risks according to the fused information. Since the abnormal information during charging may come from multiple parameters and multiple evaluation results, a single information source may have errors or one-sidedness. By multi-source information fusion, various factors can be considered comprehensively to improve the accuracy and reliability of abnormal identification. The hierarchical judgment can take appropriate warning measures according to the severity of the abnormality, avoiding unnecessary panic and resource waste, while ensuring that serious abnormalities can be handled in a timely manner. Through effective fusion and classification of abnormal information, this unit provides clear basis for the generation of intelligent warning strategies, making the warning more accurate and efficient.

[0037] The specific implementation process of this unit is as follows: First, receive the abnormal trend prediction results sent by the charging state evaluation and abnormal trend prediction unit, including abnormal probability, abnormal parameter combination and abnormal development rate at different time points. At the same time, obtain the latest raw parameter data from the dynamic charging parameter real-time acquisition and multi-dimensional analysis unit as auxiliary verification information. Then, use a weighted fusion algorithm to fuse the multi-source abnormal information, with the weight of the abnormal trend prediction result set to 0.7-0.8 and the weight of the raw parameter data set to 0.2-0.3. The weights are dynamically adjusted according to the importance and reliability of the parameters. During the fusion process, consistency test is performed on information from different sources, and obviously contradictory data is excluded, and information with deviation is corrected. After fusion, according to the preset classification standard, the abnormal risk is divided into 4 levels: level 1 (slight abnormality, no need to handle immediately, only need to strengthen monitoring), level 2 (general abnormality, need to remind operation and maintenance personnel to pay attention), level 3 (more serious abnormality, need to send operation and maintenance personnel to check in time), level 4 (serious abnormality, need to stop charging immediately and take emergency measures). The threshold of the classification standard is set according to the model of the charging pile, the tolerance capacity of the battery and the safety specification, for example, when the battery temperature exceeds 45℃ and the rising rate is greater than 2℃ / min, it is judged as level 3 abnormality; when the pile temperature exceeds 60℃, it is judged as level 4 abnormality. After the classification result is generated, it is transmitted to the intelligent warning strategy generation and dynamic adjustment unit, and the detailed records of the fusion process and classification result are stored, including fusion weight, test result and classification basis, etc., for subsequent query and analysis.

[0038] The intelligent warning strategy generation and dynamic adjustment unit is connected to the output end of the multi-source abnormal information fusion and hierarchical judgment unit, generates and adjusts the warning strategy according to the classification result;

[0039] Specifically, the intelligent early warning strategy generation and dynamic adjustment unit is the core module of the system for formulating and optimizing the early warning scheme. Its main function is to generate corresponding early warning strategies according to the abnormal risk levels output by the multi-source abnormal information fusion and hierarchical judgment unit, and to dynamically adjust the strategies according to real-time feedback information. The rationality and adaptability of the early warning strategy directly affect the effect of early warning. Different levels of abnormalities require different early warning methods, push objects, and processing suggestions to ensure that early warning information can be correctly received and effectively processed. At the same time, since the charging scene and device state are constantly changing, a fixed early warning strategy may not be able to adapt to actual needs. Through dynamic adjustment, the strategy can always maintain the best early warning effect. This unit generates targeted early warning strategies and optimizes them in real time, ensuring that the early warning system can flexibly cope with various complex situations and improving the practicality and effectiveness of early warning.

[0040] The specific implementation process of this unit is as follows: First, receive the abnormal risk levels transmitted by the multi-source abnormal information fusion and hierarchical judgment unit, call the preset strategy template library, and generate initial early warning strategies according to different levels. For example, the early warning strategy for a first-level abnormality is to prompt through the local display screen of the charging pile, updated every 5 minutes; the strategy for a second-level abnormality is to send a reminder message to the mobile APP of the maintenance personnel, with abnormal parameter details attached; the strategy for a third-level abnormality is to send an emergency notification to the maintenance personnel and arrange the nearest personnel to go to the scene, while notifying the vehicle owner through a short message; the strategy for a fourth-level abnormality is to immediately trigger the emergency stop device of the charging pile, cut off the charging circuit, send an alarm signal to the maintenance center, and make a phone call to the vehicle owner for notification. Then, receive the feedback information returned by the early warning information real-time push and execution feedback unit, including the reception status of the early warning information, the response time of the maintenance personnel, and the processing results, etc. According to these feedback information, the early warning strategy is evaluated, and if the early warning information is not received or processed in a timely manner, etc., the strategy is adjusted, such as changing the push channel, increasing the push frequency, or adjusting the processing suggestions. During the adjustment process, refer to the optimal strategy in the historical data under similar scenarios, optimize the strategy parameters through adaptive algorithms, and the adjustment period is 10-30 minutes. The adjusted early warning strategy is transmitted to the early warning information real-time push and execution feedback unit, and the strategy template library is updated, saving the effective adjustment scheme as a new template for subsequent similar situations, while recording the whole process of strategy generation and adjustment, including the initial strategy content, feedback information, and adjustment details, etc., for continuous improvement of the system.

[0041] The early warning information real-time push and execution feedback unit is connected to the output end of the intelligent early warning strategy generation and dynamic adjustment unit, pushes the early warning information and receives the execution feedback, and the output end of the unit is connected in a closed loop with the dynamic charging parameter real-time collection and multi-dimensional analysis unit, and the feature depth mining and spatio-temporal correlation mapping unit.

[0042] Specifically, the early warning information real-time pushing and execution feedback unit is the key link for the system to realize early warning information transmission and closed-loop control. Its main function is to convert the early warning strategy formulated by the intelligent early warning strategy generation and dynamic adjustment unit into specific early warning information, and push it to relevant personnel and equipment in real time through various channels, while collecting the execution and processing results of the early warning information to form feedback data. This unit is the bridge for the system to interact with the outside world, and the timeliness and accuracy of its information pushing directly affect the efficiency of abnormal handling, while the feedback data provides an important basis for the self-optimization of the system. Through real-time pushing of early warning information and collection of execution feedback, it can ensure that abnormal situations are paid attention to and handled in a timely manner, while enabling the system to continuously adjust and improve according to actual conditions, forming a complete closed-loop control system.

[0043] The specific implementation process of this unit is as follows: First, receive the early warning strategy transmitted by the intelligent early warning strategy generation and dynamic adjustment unit, and parse the pushing objects, content, method and frequency parameters in the strategy. According to the parsing result, early warning information is pushed through various channels: for local prompt of charging pile, control the display screen to display abnormal information and processing suggestions, adjust the brightness to the maximum, and execute the display frequency according to the strategy setting; for mobile phone APP pushing, encapsulate the early warning information into a standard format message and send it to the APP of designated operation and maintenance personnel and vehicle owners through mobile network, ensuring that the information is delivered within 10s; for emergency notification, send SMS through SMS gateway and make a voice call to remind, until confirmation of receipt; for device control instructions such as emergency stop signal, send them to the control module of the charging pile through the internal communication bus to ensure execution within 1s. Then, monitor the pushing state of the early warning information in real time, including whether it is successfully sent and whether it is read, etc., and collect execution feedback information such as the departure time, arrival time, processing measures and results of the operation and maintenance personnel, as well as the execution state of the charging pile, etc. through interaction with the APP of the operation and maintenance personnel and the control module of the charging pile. Feedback information is sorted in chronological order to form a feedback report, which is sent to the intelligent early warning strategy generation and dynamic adjustment unit and the dynamic charging parameter real-time acquisition and multi-dimensional analysis unit through data interface, and the pushing records and feedback reports are stored in the database with a storage period of 1 year for data statistics and system evaluation.

[0044] Preferably, the feature deep mining and spatio-temporal correlation mapping unit adopts an improved feature extraction model, and the model formula is: wherein, is the fusion feature value at time t, , , is the adaptive weight coefficient, is the voltage and current The feature extraction results of is the battery temperature at time t, Pile temperature The temperature effect function, For charging time, The grid fluctuation frequency The charging state assessment and abnormal trend prediction unit adopts a state assessment model, the formula is: ,in, is the charge state evaluation value at time t, Processing function for the improved gray wolf optimization algorithm, is the back propagation neural network function, is the weight matrix, is the bias term.

[0045] Specifically, the improved feature extraction model employed by the Deep Feature Mining and Spatiotemporal Correlation Mapping unit uses adaptive weighting coefficients to comprehensively consider the neural network-processed voltage and current features, as well as the influence of battery temperature, charging pile temperature, charging duration, and grid fluctuation frequency. This model more comprehensively extracts fused feature values ​​reflecting the charging state, enhancing the relevance of these features to the essence of the charging process. In this model, the temperature and frequency influence functions are set based on the actual variations of charging pile temperature and grid fluctuation frequency, respectively, ensuring that these parameters receive appropriate weight in feature extraction. The state assessment model in the Charging State Assessment and Abnormal Trend Prediction unit utilizes an improved Grey Wolf Optimization algorithm to optimize the results of backpropagation neural network processing. By properly setting the weight matrix and bias term, the mapping relationship between features and charging state assessment values ​​is strengthened, improving assessment accuracy. During implementation, the value range and function form of each parameter are first determined. The model is then trained using extensive historical data to determine the optimal values ​​for the adaptive weighting coefficients, weight matrix, and bias term. This ensures that the model consistently outputs reliable fused feature values ​​and state assessment values, providing a solid foundation for subsequent anomaly detection.

[0046] Preferably, the feature depth mining and spatiotemporal correlation mapping unit further includes a parameter dynamic adaptation subunit, and the adaptation model formula adopted by the unit is: ,in, is the network parameter adaptation value at time t, To adaptively optimize the inverse function of the long short-term memory neural network, are the partial derivatives of the fused eigenvalue with respect to voltage and current, are the changes in voltage and current at time t respectively; the charging state assessment and abnormal trend prediction unit also includes an abnormal sensitivity adjustment subunit, and the model formula is: ,in, is the abnormal sensitivity at time t, is an improved grey wolf optimization algorithm processing function, is a state of charge evaluation value at time t, is a derivative function of a back propagation neural network, are respectively a battery and a pile body temperature change amount, is a temperature influence coefficient.

[0047] Specifically, the parameter dynamic adaptation subunit of the feature depth mining and space-time correlation mapping unit realizes real-time adjustment of the neural network parameters with the changes of voltage and current by calculating the network parameter adaptation value, so as to ensure that the network can still maintain good feature extraction performance when the parameters fluctuate. The adaptation model quantifies the influence degree of parameter change on feature extraction by means of the partial derivative of the fusion feature value with respect to voltage and current, and adjusts the network parameters combined with the parameter change amount, thereby enhancing the adaptability of the network to the dynamic charging process. The abnormal sensitivity adjustment subunit of the state of charge evaluation and abnormal trend prediction unit adjusts the sensitivity of the state evaluation to the abnormality according to the changes of the battery and pile body temperature, so as to improve the sensitivity to discover the abnormality earlier when the temperature changes rapidly, and reduce the sensitivity to reduce the misjudgment when the temperature is stable. In the implementation process, the calculation method of the partial derivative and the reasonable value of the temperature influence coefficient need to be determined through experiments, and then the model is trained using actual charging data, so that the network parameter adaptation value and the abnormal sensitivity can accurately reflect the parameter change and the temperature condition, and improve the sensing flexibility of the system to the abnormality.

[0048] Preferably, the dynamic charging parameter real-time acquisition and multi-dimensional analysis unit adopts a parameter analysis model, and the formula is: wherein, is a multi-dimensional analysis result at time t, is the number of parameter types, is an analysis coefficient of the i-th parameter, is the acquisition value of the i-th parameter at time t, is a time decay function of the i-th parameter; and the feature enhancement model of the feature depth mining and space-time correlation mapping unit is: wherein, is an enhanced feature value, is an enhancement coefficient, is an activation function, are respectively , a multi-dimensional analysis result at time t, is a fusion feature value at time t, is a long short-term memory neural network optimized by self-adaption, , a feature extraction result at time t.

[0049] Specifically, the dynamic charging parameter real-time acquisition and multi-dimensional analysis unit has a parameter analysis model, which, by setting an analysis coefficient and a time decay function, performs weighting and time decay processing on the collected values of different parameters, highlights the influence of important parameters at the current time, and weakens the interference of historical data, so that the analysis result is more in line with the real-time charging state. The analysis coefficient is determined according to the influence degree of the parameter on the charging safety, and the time decay function is set according to the timeliness of the parameter, so as to ensure that the recent data occupies a higher weight in the analysis. The feature depth mining and space-time correlation mapping unit has a feature enhancement model, which, by introducing an enhancement coefficient and an activation function, uses the analysis results of the previous two time points to enhance the current fusion feature value, strengthens the historical trend information contained in the feature, and makes the feature more representative. When implemented, the specific forms of the analysis coefficient, the time decay function, the enhancement coefficient and the activation function need to be determined through data analysis, and after multiple iterations and optimization, the analysis result can accurately reflect the real-time state of the parameter, and the enhanced feature can more effectively support the subsequent state evaluation.

[0050] Preferably, the charging state evaluation and abnormal trend prediction unit has an abnormal trend prediction model: wherein, is the prediction state value at the time t, is the charging state evaluation value at the time t, is an improved grey wolf optimization algorithm processing function, is a back propagation neural network function, is the gradient of the state evaluation value at the time t, is a prediction time interval, is a fluctuation coefficient, is a parameter change vector at the time t; and the multi-source abnormal information fusion and hierarchical determination unit has a fusion model: wherein, is the abnormal fusion value at the time t, is a prediction step number, is a weight of the jth prediction step, is a temperature correction function corresponding to the jth prediction step, is the battery temperature at the time t, is the pile body temperature at the time t.

[0051] ​Specifically, the abnormal trend prediction model of the charging state assessment and abnormal trend prediction unit predicts the charging state at future moments by incorporating the prediction interval, the gradient of the state assessment value, the fluctuation coefficient, and the parameter change vector. The gradient reflects the changing trend of the current state, the parameter change vector reflects the impact of parameter fluctuations on the future state, and the fluctuation coefficient adjusts the degree of this impact, ensuring that the prediction results can proactively reflect potential abnormalities. The fusion model of the multi-source abnormal information fusion and classification determination unit performs a weighted fusion of multi-step predicted state values ​​and incorporates a temperature correction function to comprehensively consider the reliability of different prediction steps and the impact of temperature on anomalies. This ensures that the fused anomaly value comprehensively reflects abnormal conditions across multiple time periods. In implementation, the prediction interval, fluctuation coefficient, number of prediction steps, and weights for each step are determined based on the dynamic characteristics of the charging process. The temperature correction function is then trained on a large amount of sample data to determine its form. This ensures that the predicted state values ​​and fused anomaly values ​​accurately reflect future abnormal trends, providing a reliable basis for classification determination.

[0052] Preferably, the strategy generation model of the intelligent early warning strategy generation and dynamic adjustment unit is: ,in, is the early warning strategy at time t, Based on the abnormal fusion value and classification thresholds The policy function, is the adjustment coefficient, To adaptively optimize the long short-term memory neural network function, for The strategy adjustment amount at the moment; the feedback processing model of the real-time push of the warning information and the execution feedback unit is: ,in, is the feedback processing result at time t, For execution results The feedback function, is the feedback coefficient, First, the product of the early warning strategy Str (t) and the feedback coefficient κ is processed by the back propagation BP algorithm, and then the result is input into the improved gray wolf optimization The algorithm performs nested operations for optimization calculations.

[0053] Specifically, the strategy generation model of the intelligent early warning strategy generation and dynamic adjustment unit determines the basic early warning strategy in combination with the abnormal fusion value and the grading threshold value, and introduces the adjustment coefficient and the neural network processing result of the previous time strategy adjustment amount, so that the strategy can be optimized according to the historical adjustment experience and the pertinence of the strategy to different abnormal degrees is enhanced. The model converts the abnormal fusion value and the grading threshold value into the initial strategy through a function, and then learns the historical adjustment amount by using a neural network, so that the strategy remains applicable in dynamic changes. The feedback processing model of the early warning information real-time pushing and execution feedback unit associates the early warning strategy with the execution effect by processing the feedback function and optimizing the neural network, so that the feedback result can accurately reflect the effectiveness of the strategy. During implementation, the reasonable range of the grading threshold value and the adjustment coefficient needs to be set, the form of the strategy function and the feedback function is determined through actual cases, and then the neural network is trained by using the historical strategy adjustment data and the execution feedback data, so that the generated early warning strategy can be dynamically optimized according to the abnormal situation and the feedback information, and the continuous improvement of the early warning effect is ensured.

[0054] Preferably, the state of charge evaluation and abnormal trend prediction unit comprises: a state feature matching subunit that receives feature data output from the feature deep mining and space-time correlation mapping unit, compares the feature data with a preset normal charging state feature library one by one, determines the matching degree by calculating the Euclidean distance between the feature vectors, and screens out feature combinations with a matching degree lower than a set threshold; an abnormal trend quantification subunit that performs time series analysis on the screened low-matching-degree feature combinations, calculates the slope and curvature of the features with the charging time, and converts the abnormal trend into a quantifiable numerical index in combination with the change rates of the battery temperature and the pile body temperature; an optimization evaluation subunit that optimizes the evaluation result of the back propagation neural network by using an improved grey wolf optimization algorithm, and reduces the deviation of the evaluation result from the actual charging state to the minimum range by adjusting the weight and bias parameters of the network; and a result output subunit that organizes the optimized charging state evaluation result and the abnormal trend prediction index according to a set data format, and transmits them to the multi-source abnormal information fusion and grading determination unit.

[0055] Preferably, the multi-source abnormal information fusion and grading judgment unit includes: an abnormal information receiving subunit, which receives various types of abnormal information output from the charging state evaluation and abnormal trend prediction unit, including abnormal characteristics, trend indicators and evaluation scores, and verifies the integrity and format of the information to eliminate invalid information; a multi-dimensional fusion subunit, which classifies abnormal information from different sources according to the time dimension and feature dimension, and uses the weighted average method to fuse similar abnormal information at the same time point to generate a comprehensive abnormal vector; a grading threshold determination subunit, which dynamically sets different levels of abnormal thresholds according to the model of the charging pile, the battery type and the charging stage, and each threshold corresponds to a set warning level; a grading result output subunit, which compares the fused abnormal information with the set threshold, determines the corresponding warning level, and transmits the grading result to the intelligent warning strategy generation and dynamic adjustment unit.

[0056] Preferably, the intelligent early warning strategy generation and dynamic adjustment unit includes: a basic strategy library calling subunit, which calls the corresponding initial early warning strategy from the preset basic early warning strategy library according to the early warning level output by the multi-source abnormal information fusion and hierarchical judgment unit, and the initial strategy includes the early warning method, push object and processing suggestion; a strategy dynamic adjustment subunit, which combines the early warning information real-time push and the feedback result returned by the execution feedback unit, and uses the adaptive optimization long short-term memory neural network to adjust the initial strategy, correct the early warning frequency and push content, so that the strategy is more in line with the actual situation; a multi-scenario adaptation subunit, which analyzes the environmental parameters of the current charging scene, including ambient temperature, humidity and grid stability, and performs scene adaptation optimization on the adjusted strategy to ensure consistent early warning effects in different scenarios; a final strategy output subunit, which transmits the final early warning strategy after dynamic adjustment and scene adaptation to the early warning information real-time push and execution feedback unit, and stores the strategy parameters in the historical database.

[0057] like Figure 2 As shown, the new energy charging pile charging process early warning system based on artificial intelligence includes:

[0058] Step S1: The dynamic charging parameter real-time acquisition and multi-dimensional analysis unit continuously acquires and performs multi-dimensional analysis on the charging pile output voltage, charging current, battery temperature, charging time, pile body temperature, and grid fluctuation frequency;

[0059] Step S2: transmitting the parsed parameters to the feature deep mining and spatiotemporal correlation mapping unit to perform feature deep mining and spatiotemporal correlation mapping analysis to form a feature data set;

[0060] Step S3: Input the characteristic data set into the charging state assessment and abnormal trend prediction unit to perform charging state assessment and abnormal trend prediction to obtain abnormal trend information;

[0061] Step S4, send the abnormal trend information to the multi-source abnormal information fusion and hierarchical judgment unit, implement multi-source abnormal information fusion and hierarchical judgment, and generate abnormal hierarchical results;

[0062] Step S5, transmit the abnormal hierarchical results to the intelligent early warning strategy generation and dynamic adjustment unit, formulate the intelligent early warning strategy and dynamically adjust it, and output the adaptive early warning strategy;

[0063] Step S6, deliver the adaptive early warning strategy to the early warning information real-time pushing and execution feedback unit, perform early warning information real-time pushing and execution feedback receiving, and return the feedback information to the dynamic charging parameter real-time acquisition and multi-dimensional analysis unit, the feature deep mining and spatio-temporal correlation mapping unit, respectively, to form a complete early warning closed-loop process.

[0064] The new energy charging pile charging process early warning system based on artificial intelligence has strong multi-parameter collaborative analysis capability. The dynamic charging parameter real-time acquisition and multi-dimensional analysis unit can comprehensively capture information such as charging pile output voltage, current, battery and pile body temperature, charging time, power grid fluctuation frequency, etc. After processing by the feature deep mining and spatio-temporal correlation mapping unit, the spatio-temporal correlation between parameters can be deeply mined to provide accurate feature support for subsequent evaluation. Based on these features, the charging state evaluation and abnormal trend prediction unit can achieve detailed evaluation of the charging state and accurate prediction of the abnormal trend, greatly improving the identification ability of abnormalities in complex charging scenarios.

[0065] The system can effectively overcome the weak correlation between feature extraction and state evaluation in the prior art. Traditional technologies mostly analyze single parameters or simple combinations of parameters, ignoring the correlation between parameters, resulting in one-sided evaluation. The system deeply mines and analyzes the spatio-temporal correlation of multiple parameters through adaptive optimization of long short-term memory neural networks, converts scattered parameters into features with strong correlation, and improves the gray wolf optimization-back propagation neural network to perform state evaluation and abnormal trend prediction based on these correlated features, so that feature extraction and state evaluation form a close linkage, and complex problems caused by multiple parameter abnormalities can be accurately identified.

[0066] At the same time, the system solves the problem of insufficient dynamic adaptability of the prior art early warning strategy. Traditional early warning strategies are mostly fixed settings and are difficult to adapt to different charging pile models, battery types and environmental conditions. The intelligent early warning strategy generation and dynamic adjustment unit in the system adjusts the early warning logic based on the results of the multi-source abnormal information fusion and hierarchical judgment unit and the feedback of the early warning information real-time pushing and execution feedback unit. Through this dynamic adjustment mechanism, the strategy can flexibly adapt to various scenarios, avoiding early warning lag or false positives caused by changes in the scene, and ensuring efficient early warning effect under different conditions.

[0067] In the description of the application, it should be noted that unless otherwise explicitly specified and limited, the terms "arranged", "mounted", "connected", "linked", "fixed" should be understood broadly, for example, can be fixedly connected, can be detachably connected, or integrally connected; can be mechanically connected, can be electrically connected; can be directly connected, can be indirectly connected through an intermediate medium, or can be internal communication of two elements. For those of ordinary skill in the art, the specific meanings of the above terms in the present application can be understood according to the specific circumstances.

[0068] Although embodiments of the present application have been shown and described, it would be appreciated by those of ordinary skill in the art that various equivalents, modifications, replacements and variations of these embodiments can be made without departing from the principles and spirit of the present application, and the scope of the present application is defined by the appended claims and their equivalent scope.

Claims

1. The artificial intelligence-based new energy charging pile charging process early warning system is characterized by: include: The dynamic charging parameter real-time acquisition and multi-dimensional analysis unit is used to obtain the charging pile output voltage, charging current, battery temperature, charging time, pile body temperature and grid fluctuation frequency; the feature deep mining and spatiotemporal correlation mapping unit, the input end of which is connected to the output end of the dynamic charging parameter real-time acquisition and multi-dimensional analysis unit, and performs feature extraction and spatiotemporal correlation analysis on the parameters; the charging state evaluation and abnormal trend prediction unit, the input end of which is connected to the output end of the feature deep mining and spatiotemporal correlation mapping unit, and evaluates the charging state and predicts abnormal trends based on the extracted features; the multi-source abnormal information fusion and hierarchical judgment unit, the output end of which is connected to the output end of the feature deep mining and spatiotemporal correlation mapping unit, and evaluates the charging state and predicts abnormal trends based on the extracted features; The input end is connected to the output end of the charging state assessment and abnormal trend prediction unit to fuse and classify abnormal information; the intelligent early warning strategy generation and dynamic adjustment unit, the input end of which is connected to the output end of the multi-source abnormal information fusion and classification judgment unit, and generates and adjusts the early warning strategy based on the classification result; the early warning information real-time push and execution feedback unit, the input end of which is connected to the output end of the intelligent early warning strategy generation and dynamic adjustment unit, pushes early warning information and receives execution feedback, and the output end of this unit forms a closed-loop connection with the dynamic charging parameter real-time acquisition and multi-dimensional analysis unit and the feature deep mining and spatiotemporal correlation mapping unit respectively; The charging state evaluation and abnormal trend prediction unit includes: a state feature matching subunit, which receives feature data output from the feature depth mining and spatiotemporal correlation mapping unit, compares the feature data with a preset normal charging state feature library one by one, determines the matching degree by calculating the Euclidean distance between feature vectors, and screens out feature combinations with a matching degree lower than a set threshold; an abnormal trend quantification subunit, which performs time series analysis on the screened low-matching feature combinations, calculates the slope and curvature of the feature as the charging time changes, and converts the abnormal trend into a quantifiable numerical indicator in combination with the rate of change of the battery temperature and the pile body temperature; an optimization evaluation subunit, which optimizes the evaluation results of the back propagation neural network using an improved gray wolf optimization algorithm, and reduces the deviation between the evaluation results and the actual charging state to a minimum range by adjusting the weights and bias parameters of the network; a result output subunit, which organizes the optimized charging state evaluation results and abnormal trend prediction indicators according to the set data format, and transmits them to the multi-source abnormal information fusion and classification judgment unit; The multi-source abnormal information fusion and classification determination unit includes: an abnormal information receiving subunit, which receives various types of abnormal information output by the charging state evaluation and abnormal trend prediction unit, including abnormal characteristics, trend indicators and evaluation scores, and verifies the integrity and format of the information and eliminates invalid information; a multi-dimensional fusion subunit, which classifies abnormal information from different sources according to the time dimension and feature dimension, and adopts the weighted average method to fuse similar abnormal information at the same time point to generate a comprehensive abnormal vector; a classification threshold determination subunit, which dynamically sets different levels of abnormal thresholds according to the model of the charging pile, the battery type and the charging stage, and each threshold corresponds to a set warning level; a classification result output subunit, which compares the fused abnormal information with the set threshold, determines the corresponding warning level, and transmits the classification result to the intelligent warning strategy generation and dynamic adjustment unit; The intelligent early warning strategy generation and dynamic adjustment unit includes: a basic strategy library calling subunit, which calls the corresponding initial early warning strategy from the preset basic early warning strategy library according to the early warning level output by the multi-source abnormal information fusion and hierarchical judgment unit. The initial early warning strategy includes the early warning method, push object and processing suggestions; a strategy dynamic adjustment subunit, which combines the feedback results returned by the early warning information real-time push and execution feedback unit, and uses the adaptive optimization long short-term memory neural network to adjust the initial early warning strategy, correct the early warning frequency and push content, and make the strategy more in line with the actual situation; a multi-scenario adaptation subunit, which analyzes the environmental parameters of the current charging scenario, including ambient temperature, humidity and power grid stability, and performs scenario adaptation optimization on the adjusted strategy to ensure consistent early warning effects in different scenarios; a final strategy output subunit, which transmits the final early warning strategy after dynamic adjustment and scenario adaptation to the early warning information real-time push and execution feedback unit, and stores the strategy parameters in the historical database.

2. The artificial intelligence-based new energy charging pile charging process early warning system according to claim 1 is characterized in that: The feature depth mining and spatiotemporal correlation mapping unit adopts an improved feature extraction model, and the model formula is: ,in, is the fusion eigenvalue at time t, is the adaptive weight coefficient, The voltage at time t is the adaptively optimized long short-term memory neural network and current The feature extraction results of is the battery temperature at time t, Pile temperature The temperature effect function, For charging time, The grid fluctuation frequency The charging state assessment and abnormal trend prediction unit adopts a state assessment model, the formula is: ,in, is the charge state evaluation value at time t, Processing function for the improved gray wolf optimization algorithm, is the back propagation neural network function, is the weight matrix, is the bias term.

3. The artificial intelligence-based new energy charging pile charging process early warning system according to claim 2 is characterized in that: The feature depth mining and spatiotemporal correlation mapping unit also includes a parameter dynamic adaptation subunit, which adopts the adaptation model formula: ,in, is the network parameter adaptation value at time t, To adaptively optimize the inverse function of the long short-term memory neural network, are the partial derivatives of the fused eigenvalue with respect to voltage and current, are the changes in voltage and current at time t respectively; the charging state assessment and abnormal trend prediction unit also includes an abnormal sensitivity adjustment subunit, and the model formula is: ,in, is the abnormal sensitivity at time t, Processing function for the improved gray wolf optimization algorithm, is the charge state evaluation value at time t, is the derivative function of the back-propagation neural network, are the temperature changes of the battery and pile respectively, is the temperature influence coefficient.

4. The artificial intelligence-based new energy charging pile charging process early warning system according to claim 2 is characterized in that: The dynamic charging parameter real-time acquisition and multi-dimensional analysis unit adopts a parameter analysis model, and the formula is: ,in, is the multi-dimensional analysis result at time t, is the number of parameter types, is the analytical coefficient of the i-th parameter, is the collected value of the i-th parameter at time t, is the time decay function of the i-th parameter; the feature enhancement model of the feature depth mining and spatiotemporal correlation mapping unit is: ,in, is the enhanced eigenvalue, is the enhancement coefficient, is the activation function, They are 、 The multi-dimensional analysis results of the moment, is the fusion eigenvalue at time t, The adaptively optimized long short-term memory neural network 、 Feature extraction results at that moment.

5. The artificial intelligence-based new energy charging pile charging process early warning system according to claim 4 is characterized in that: The abnormal trend prediction model of the charging state evaluation and abnormal trend prediction unit is: ,in, for The predicted state value at time t, is the charge state evaluation value at time t, Processing function for the improved gray wolf optimization algorithm, is the back propagation neural network function, is the gradient of the state evaluation value at time t, is the prediction time interval, is the coefficient of fluctuation, is the parameter change vector at time t; the fusion model of the multi-source abnormal information fusion and classification judgment unit is: ,in, is the abnormal fusion value at time t, To predict the number of steps, is the weight of the j-th step prediction, is the temperature correction function corresponding to the j-th step prediction, is the battery temperature at time t, is the temperature of the pile at time t.

6. The artificial intelligence-based new energy charging pile charging process early warning system according to claim 5 is characterized in that: The strategy generation model of the intelligent early warning strategy generation and dynamic adjustment unit is: ,in, is the early warning strategy at time t, Based on the abnormal fusion value and classification thresholds The policy function, is the adjustment coefficient, To adaptively optimize the long short-term memory neural network function, for The strategy adjustment amount at the moment; the feedback processing model of the real-time push of the warning information and the execution feedback unit is: ,in, is the feedback processing result at time t, For execution results The feedback function, is the feedback coefficient, First, the product of the early warning strategy Str (t) and the feedback coefficient κ is processed by the back propagation BP algorithm, and then the result is input into the improved gray wolf optimization GWOᵢ mp The algorithm performs nested operations for optimization calculations.

7. The artificial intelligence-based new energy charging pile charging process warning system according to any one of claims 1 to 6, characterized in that: The system operation includes: Step S1: The dynamic charging parameter real-time acquisition and multi-dimensional analysis unit continuously acquires and performs multi-dimensional analysis on the charging pile output voltage, charging current, battery temperature, charging time, pile body temperature, and grid fluctuation frequency; Step S2: transmitting the parsed parameters to the feature deep mining and spatiotemporal correlation mapping unit to perform feature deep mining and spatiotemporal correlation mapping analysis to form a feature data set; Step S3: Input the characteristic data set into the charging state assessment and abnormal trend prediction unit to perform charging state assessment and abnormal trend prediction to obtain abnormal trend information; Step S4: Send the abnormal trend information to the multi-source abnormal information fusion and classification determination unit, implement multi-source abnormal information fusion and classification determination, and generate an abnormal classification result; Step S5: Transmit the abnormality classification results to the intelligent early warning strategy generation and dynamic adjustment unit, formulate an intelligent early warning strategy and perform dynamic adjustment to produce an adaptive early warning strategy; Step S6: The adapted warning strategy is delivered to the warning information real-time push and execution feedback unit to push the warning information in real time and receive execution feedback, and the feedback information is respectively transmitted back to the dynamic charging parameter real-time acquisition and multi-dimensional analysis unit and the feature deep mining and spatiotemporal correlation mapping unit to form a complete warning closed-loop process.

Images