Charging pile fault protection method, system, equipment and medium

Abstract

The invention relates to the technical field of charging pile fault protection, in particular to a charging pile fault protection method, system and device and a medium, and the method comprises the steps: continuously collecting the multi-modal data of a charging pile in real time, and carrying out the preprocessing; inputting the preprocessed multi-modal data into a pre-trained charging pile fault probability prediction model, outputting a fault probability value, comparing the fault probability value with a set threshold value, judging whether a fault risk exists or not, and if yes, giving out fault early warning; when a fault occurs, classifying and positioning the fault according to the multi-modal data, judging the fault level, executing hierarchical protection based on the fault level, and updating the charging pile fault probability prediction model by using the multi-modal data when the fault occurs; and after grading protection is completed, corresponding recovery measures are executed according to fault classification and positioning. According to the invention, early accurate early warning, flexible protection and rapid recovery of the fault of the charging pile can be realized, and the safety and operation and maintenance efficiency of the charging pile are comprehensively improved.

Description

Background Art

[0002] With the rapid popularization of electric vehicles, the scale of charging piles, as the core infrastructure, continues to expand. Their safety and reliability are directly related to user trust and public safety. Charging piles are long-term exposed to complex environments, and potential hazards such as electrical aging, abnormal temperature rise, and mechanical wear have gradually become the main causes of failures. However, traditional protection mechanisms often struggle to achieve risk pre-control in the face of dynamically changing operating conditions, and the industry urgently needs more intelligent solutions.

[0003] In the prior art, the safety protection of charging piles mainly relies on monitoring electrical parameters in a single dimension. For example, current and voltage sensors are used to detect overload or short circuit, and fuses or circuit breakers are triggered for passive protection. Some solutions introduce temperature sensors to monitor the temperature rise of key components and combine thermistors to achieve current limiting; other solutions reduce the impact of failures on the continuous operation of the system through regular manual inspections or redundant circuit designs.

[0004] However, the prior art still has limitations in multi-dimensional state perception, dynamic fault prediction, and hierarchical response. Single-sensor data is difficult to cover all the risks in the operation of charging piles, resulting in delayed warnings or misjudgments; the fixed threshold mechanism cannot adapt to environmental interference and load fluctuations, and the protection strategy lacks flexibility; fault recovery relies on manual experience, and the process efficiency from fault occurrence to repair is low, and the operation and maintenance costs are difficult to effectively control. Summary of the Invention

[0005] Aiming at the technical problems that the existing charging pile fault protection methods rely on single-sensor data, fixed threshold judgment, and manual recovery processes, resulting in delayed fault warnings, rigid protection strategies, and low operation and maintenance efficiency for charging piles, this application provides a charging pile fault protection method, system, device, and medium. Through multi-modal data fusion, deep learning dynamic prediction, hierarchical protection execution, and recovery mechanisms, it realizes early and accurate fault warnings, flexible protection, and rapid recovery, comprehensively improving the safety and operation and maintenance efficiency of charging piles.

[0006] In the first aspect, this application provides a charging pile fault protection method, including the following steps: S1. Continuously collect multi-modal data of the charging pile in real time. The multi-modal data includes electrical data, temperature data, and sound signal data of the charging pile, and preprocess the multi-modal data at the current moment; S2. Input the preprocessed multi-modal data into a pre-trained charging pile fault probability prediction model, and output a fault probability value; S3. Compare the fault probability value with a set threshold to determine whether there is a fault risk. If there is a fault risk, issue a fault warning; S4. When a failure occurs, classify and locate the failure based on multimodal data, determine the failure level, perform hierarchical protection based on the failure level, and update the charging pile failure probability prediction model using the multimodal data at the time of failure; S5. After the hierarchical protection is completed, perform corresponding recovery measures according to the failure classification and location.

[0007] Further, it should be noted that in step S1, the electrical data includes the charging current and the charging voltage; The temperature data includes the temperatures of the key heat - generating parts inside the charging pile, and the key heat - generating parts include the power module, the transformer, and the charging interface; The sound signal data includes the system operation sound signals of the key parts with abnormal sounds inside the charging pile, and the key parts with abnormal sounds include the relay and the contactor.

[0008] Further, it should be noted that in step S1, the pre - processing includes: Perform Fourier transform on the sound signal data for a fixed duration before the current moment to convert it into a sound frequency - domain signal. The formula is:

[0009] In the formula, is the sound signal data, is the imaginary unit, is the frequency, is the time; Perform data cleaning and denoising processing on the electrical data, temperature data, and sound frequency - domain signal.

[0010] Further, it should be noted that in step S2, the charging pile failure probability prediction model is constructed based on a deep - learning algorithm model. The deep - learning algorithm model includes one or a combination of an LSTM network model, a CNN network model, and a GRU network model. The output layer of the charging pile failure probability prediction model uses the Sigmoid function to generate the failure probability value.

[0011] Further, it should be noted that the training steps of the charging pile failure probability prediction model include: S201. Collect historical data of multimodal data as samples. After performing the same pre - processing as in step S1, add real labels to the pre - processed multimodal data according to the actual failure occurrence situation, and then divide it into a training set and a validation set; S202. Construct a charging pile failure probability prediction model, input the training set for training, and output the failure probability prediction values of each sample; S203. Calculate the loss function based on the difference between the failure probability prediction value and the real label, and update the parameters of the charging pile failure probability prediction model through the backpropagation algorithm; S204. Verify the prediction accuracy of the charging pile fault probability prediction model using the validation set. Stop training when the prediction accuracy reaches the predetermined standard to obtain a pre-trained charging pile fault probability prediction model.

[0012] It should be further noted that in step S4, when the multi-modal data meets at least one fault condition, it is determined that a fault has occurred. The fault conditions include: The ratio of the equivalent resistance to the reference value exceeds the preset threshold, and the equivalent resistance is calculated based on the current and voltage in the electrical data; The temperature change rate of any key heat-generating part exceeds the preset threshold, and the temperature change rate is calculated based on the temperature at the current moment and the temperature during a fixed period of time before the current moment; The amplitude ratio of the sound frequency domain signal to the background noise in a specific frequency band exceeds the preset threshold.

[0013] It should be further noted that the determination criteria for the fault level are: Calculate the ratios between the ratio of the equivalent resistance to the reference value, the temperature change rate, and the amplitude ratio of the sound frequency domain signal to the background noise in a specific frequency band and the preset threshold respectively. When the highest value among the three is ∈[1, 2), it is determined as a minor fault. When the highest value among the three is ∈[2, 5), it is determined as a moderate fault. When the highest value among the three is ≥5, it is determined as a severe fault.

[0014] It should be further noted that the hierarchical protection in step S4 includes: Set a PTC-liquid metal composite switch in the main charging circuit of the charging pile. The PTC-liquid metal composite switch is provided with a PTC resistance part and a liquid metal part; In case of a minor fault, the PTC resistance part dynamically increases the resistance value based on the real-time temperature increase amount to limit the current growth; In case of a moderate fault, the PTC resistance part and the liquid metal part work together to further increase the resistance; In case of a severe fault, trigger the fuse to operate and cut off the main circuit.

[0015] It should be further noted that in step S5, the fault categories include software faults and hardware faults.

[0016] It should be further noted that the fault classification is achieved by analyzing the correlation between the mutation of the electrical data and the sound frequency domain signal; the fault location is completed by comparing the temperature gradient distribution of each key heat-generating part before and after the fault and the time stamp of the energy mutation of the sound signal data.

[0017] It should be further noted that in step S5, the recovery measures include: For software program faults, call the backup program and data to restore the system operation; For hardware failures, if the faulty component is an electronic component that can be automatically reset, perform an automatic reset; If the faulty component is an electronic component that cannot be automatically reset or an electronic component that fails after an automatic reset, disconnect the connection between the electronic component and the system inside the charging pile and generate a fault report. The content of the fault report includes the name, model, fault cause analysis, and maintenance suggestions of the faulty component.

[0018] In a second aspect, the present application provides a charging pile fault protection system for implementing the above-mentioned charging pile fault protection method for an integrated energy system, including: A data acquisition and processing module for continuously collecting multi-modal data in real time and preprocessing the multi-modal data at the current moment; A fault probability prediction module for inputting the preprocessed multi-modal data into a pre-trained charging pile fault probability prediction model and outputting a fault probability value; A fault judgment and early warning module for comparing the fault probability value with a set threshold to determine whether there is a fault risk, and issuing a fault early warning if there is a fault risk; A fault classification and protection execution module for classifying and locating the fault according to the multi-modal data when a fault occurs, determining the fault level, and performing hierarchical protection based on the fault level; A model update module for updating the parameters of the charging pile fault probability prediction model using the multi-modal data when a fault occurs; A recovery execution module for performing corresponding recovery measures according to the fault classification and location after the hierarchical protection is completed.

[0019] In a third aspect, the present application provides an electronic device, including a memory, a processor, and a computer program stored on the memory and executable on the processor. The processor is used to implement the steps of the above-mentioned charging pile fault protection method when executing the computer program.

[0020] In a fourth aspect, the present application provides a storage medium with a computer program stored thereon. The computer program is used to implement the steps of the above-mentioned charging pile fault protection method when executed by a processor.

[0021] From the above technical solutions, it can be seen that the present application has the following advantages: 1. By collecting the electrical data, temperature data, and sound signal data of the charging pile in real time, and performing synchronous preprocessing and fusion analysis on the multi-modal data, the present application solves the problem of insufficient monitoring dimensions of traditional single sensors, can comprehensively perceive potential hazards such as electrical abnormalities, local temperature rise, and mechanical wear in the operation state of the charging pile, realizes the early identification of potential faults such as poor contact and arc discharge, and enables the warning signal to change from passive response to active prediction, thus winning time for early intervention before the fault occurs.

[0022] 2. This application predicts the fault probability of multimodal data through a pre-trained deep learning model, replacing the static judgment logic based on fixed thresholds, and solving the problem of high false alarm rate under complex environmental interference. The model dynamically learns the data characteristics of different loads and working conditions, autonomously differentiates normal fluctuations from real faults, and shifts the warning accuracy from experience-driven to data-driven, significantly reducing the risks of missed detection and misjudgment.

[0023] 3. By implementing the fault classification and protection strategy, this application overcomes the problem of overprotection caused by the indiscriminate circuit cutting of traditional fuses, can dynamically adjust the protection intensity according to the severity of the fault, reduces unnecessary downtime while ensuring equipment safety, and avoids frequent impacts on core components.

[0024] 4. By implementing corresponding recovery measures after a fault, this application changes the situation of low efficiency in manual troubleshooting, quickly restores the system operation or generates maintenance guidance according to the fault type, and shifts the operation and maintenance process from passive response to active closed-loop management, shortening the equipment recovery cycle. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] In order to more clearly illustrate the technical solutions of this application, the drawings required for description will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of this application. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0026] Figure 1 is a flowchart of a charging pile fault protection method in an embodiment of this application.

[0027] Figure 2 is a schematic block diagram of a charging pile fault protection system in an embodiment of this application.

[0028] Figure 3 is a schematic hardware structure diagram of an electronic device in an embodiment of this application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0029] In order to make the application purpose, features, and advantages of this application more obvious and understandable, the technical solutions protected by this application will be clearly and completely described below by using specific embodiments and the accompanying drawings. Obviously, the embodiments described below are only some embodiments of this application, rather than all embodiments. Based on the embodiments in this application, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of this application.

[0030] The charging pile fault protection method involved in this application mainly targets the technical field of charging pile fault protection. By collecting electrical data, temperature data, and sound signal data of the charging pile in real time, and performing synchronous preprocessing and fusion analysis on multi-modal data, it solves the problem of insufficient monitoring dimensions of traditional single sensors. It can comprehensively perceive potential hazards such as electrical abnormalities, local temperature rise, and mechanical wear in the operation state of the charging pile, realize the early identification of potential faults such as poor contact and arc discharge, and shift the warning signal from passive response to active prediction, striving for time for early intervention before the fault; By using a pre-trained deep learning model to predict the fault probability of multi-modal data, it replaces the static judgment logic based on fixed thresholds, solves the problem of high false alarm rate under the interference of complex environments. The model autonomously distinguishes normal fluctuations from real faults by dynamically learning the data characteristics of different loads and working conditions, shifting the warning accuracy rate from experience-driven to data-driven, significantly reducing the risks of missed detection and misjudgment; By implementing the fault grading protection strategy, it overcomes the problem of overprotection caused by the indiscriminate circuit cutting of traditional fuses, can dynamically adjust the protection intensity according to the severity of the fault, reduce unnecessary downtime while ensuring equipment safety, and avoid frequent impacts on core components; By executing corresponding recovery measures after the fault, it changes the situation of low efficiency of manual troubleshooting, quickly restores the system operation or generates maintenance guidance according to the fault type, and shifts the operation and maintenance process from passive response to active closed-loop management, shortening the equipment recovery cycle.

[0031] The charging pile fault protection method involved in this application mainly has the technical problems of lagging charging pile fault warning, rigid protection strategy, and low operation and maintenance efficiency due to relying on single sensor data, fixed threshold judgment, and manual recovery process.

[0032] The charging pile fault protection method involved in this application will be described in detail below. For the purpose of illustration rather than limitation, specific details such as specific system structures and technologies are proposed to thoroughly understand the embodiments of this application. However, those skilled in the art should clearly understand that this application can also be implemented in other embodiments without these specific details.

[0033] In the charging pile fault protection method involved in this application, the term "including" indicates the existence of the described features, wholes, steps, operations, elements, and / or components, but does not exclude the existence or addition of one or more other features, wholes, steps, operations, elements, components, and / or their combinations. The terms "including", "comprising", "having" and their variants all mean "including but not limited to", unless otherwise specifically emphasized in other ways.

[0034] For the convenience of clearly describing the technical solutions of this application, terms such as "first" and "second" are used to distinguish identical or similar items with basically the same functions and effects. Those skilled in the art can understand that terms such as "first" and "second" do not limit the quantity and execution order, and terms such as "first" and "second" do not necessarily mean different.

[0035] Statements such as "an embodiment" or "some embodiments" described in this application mean that specific features, structures, or characteristics described in the embodiment are included in one or more embodiments of this application. Thus, statements such as "in one embodiment", "in some embodiments", "in other some embodiments", "in still other embodiments", etc. that appear in different parts of this application do not necessarily refer to the same embodiment, but mean "one or more but not all embodiments", unless otherwise specifically emphasized in other ways.

[0036] Next, the technical solutions in the embodiments of this application will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of this application. Obviously, the described embodiments are only a part of the embodiments of this application, rather than all the embodiments. Based on the embodiments in this application, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of this application.

[0037] The charging pile fault protection method provided by the embodiments of this application is executed by a computer device. Correspondingly, the charging pile fault protection system runs in the computer device.

[0038] The following are some noun explanations in this solution to facilitate a better understanding of this solution: LSTM network model: That is, the long short-term memory network model, which is a special recurrent neural network (RNN). It can learn long-term dependence information and controls the flow of information by introducing memory units and gating mechanisms. The memory units can store and update long-term information, while the input gate, output gate, and forget gate determine when to add new information to the memory units and when to read information from the memory units. This structure makes LSTM perform well in processing sequence data, such as natural language processing, speech recognition, etc., and can effectively capture long-term semantic and syntactic information in the sequence, avoiding the gradient disappearance and long-term dependence problems existing in traditional RNNs.

[0039] CNN network model: That is, the convolutional neural network model, is a deep learning model designed specifically for processing data with grid structures and is commonly used in fields such as image recognition and computer vision. It automatically extracts features of data through components such as convolutional layers, pooling layers, and fully connected layers. The convolutional layer uses convolutional kernels to slide on the data for convolutional operations to extract local features and share weights, greatly reducing the number of model parameters and computational volume. The pooling layer then downsamples the output of the convolutional layer, compresses the data dimension, and reduces the risk of overfitting. Finally, the fully connected layer performs tasks such as classification or regression on the extracted features. CNN can effectively utilize the spatial locality and hierarchical structure of data, automatically learn highly abstract feature representations, and has achieved great success in image and video processing, etc.

[0040] GRU network model, that is, the gated recurrent unit network model, is a variant of LSTM that controls the flow of information through reset gates and update gates. The reset gate determines how to combine new input information with past memories, and the update gate controls the degree to which past memories are retained at the current moment. GRU simplifies and combines the input gate, forget gate, and output gate in LSTM, making the model structure simpler and the computational efficiency higher. In many sequence data processing tasks, GRU can achieve performance similar to LSTM. At the same time, due to its relatively small number of parameters and faster training speed, it has been widely used in some scenarios with high real-time requirements. PTC-liquid metal composite switch: The PTC-liquid metal composite switch is a new type of switching device that combines a positive temperature coefficient (PTC) thermistor with liquid metal. The PTC resistor has the characteristic that its resistance increases rapidly with the increase of temperature within a specific temperature range, while liquid metal has good electrical conductivity and fluidity. When the fault current is small, the temperature inside the charging pile is relatively stable, and the PTC resistor slowly increases, playing a preliminary current-limiting role. However, as the device is used and the current increases, the temperature rises rapidly, the PTC resistor increases sharply, and at the same time, the liquid metal undergoes a morphological change under the action of heat, further increasing the resistance of the circuit to achieve hierarchical flexible current limiting. Let the fault current be, the resistance value of the PTC resistor and the temperature The relationship is ( is the initial resistance, is the temperature coefficient, is the initial temperature), the change of the liquid metal resistance can establish a corresponding model according to its physical characteristics. Through the collaborative work of the two, the fault current can be limited within a safe range.

[0041] Figure 1 is the flowchart of the charging pile fault protection method in an embodiment of this application. Among them, Figure 1The executing entity can be a charging pile fault protection system. According to different requirements, the order of steps in this flowchart can be changed, and some can be omitted.

[0042] As Figure 1 shown, the charging pile fault protection method includes: Step S1, continuously collect multi-modal data of the charging pile in real time. The multi-modal data includes electrical data, temperature data, and sound signal data of the charging pile, and preprocess the multi-modal data at the current moment.

[0043] By continuously collecting electrical data, temperature data, and sound signal data of the charging pile in real time and preprocessing them, multi-dimensional dynamic monitoring of the charging pile operation status can be realized, providing a comprehensive and high-quality data basis for subsequent fault prediction and protection, effectively avoiding misjudgment or missed detection problems caused by the limitations of a single data source, and improving the timeliness and accuracy of fault identification.

[0044] In some specific embodiments, the electrical data includes charging current and charging voltage; The temperature data includes the temperature of key heat-generating parts inside the charging pile. The key heat-generating parts include power modules, transformers, and charging interfaces; The sound signal data includes the system operation sound signals of key parts with abnormal sounds inside the charging pile. The key parts with abnormal sounds include relays and contactors.

[0045] By defining the specific collection parts of electrical data, temperature data, and sound signals, key heat-generating components such as power modules and transformers, as well as parts prone to mechanical failures such as relays and contactors, can be accurately covered, and characteristic data that can best reflect the equipment health status can be obtained specifically, reducing the interference of redundant data on model training.

[0046] In some specific embodiments, the preprocessing includes: Perform Fourier transform on the sound signal data for a fixed period of time before the current moment, and convert it into a sound frequency domain signal. The formula is:

[0047] In the formula, is the sound signal data, is the imaginary unit, is the frequency, is the time; Perform data cleaning and denoising processing on the electrical data, temperature data, and sound frequency domain signals.

[0048] By performing Fourier transform on the sound signal and combining multi-modal data cleaning, the time-domain sound characteristics can be converted into effective signals required for frequency-domain analysis, and at the same time, the influence of environmental noise and sensor errors on data quality can be eliminated, providing input features with high signal-to-noise ratio for the model.

[0049] In some specific embodiments, the specific operation of data cleaning is as follows: Electrical data cleaning: Perform moving window mean filtering on current and voltage signals to eliminate high-frequency interference, detect abnormal fluctuation points using the quartile method, fill in missing values through linear interpolation of valid data before and after, synchronize and align voltage and current signals with different sampling rates, and finally perform maximum-minimum normalization to ensure data scale consistency; Temperature data cleaning: Use the moving average algorithm to eliminate sensor jitter noise, set a temperature change rate threshold to filter out mutation interference points, fill in abnormal breakpoint data with the mean value of adjacent time periods, perform spatial interpolation on multi-temperature measurement point data to compensate for temperature values in missing areas, and enhance data continuity through exponential smoothing processing; Sound frequency domain signal cleaning: Perform band-stop filtering on the spectrum after Fourier transform to eliminate power frequency harmonic interference, use independent component analysis to separate ambient background noise, perform threshold truncation processing on specific frequency band sudden amplitude increase points, fill in missing frequency band data with weighted interpolation of adjacent frequency components, and finally standardize and balance the spectrum energy distribution.

[0050] The specific operation of denoising processing is as follows: For electrical data, use Kalman filtering to dynamically track the trend components of current and voltage. For temperature data, decompose it using wavelet transform and then reconstruct the low-frequency effective components. For sound signals, suppress random pulse noise through an adaptive filter, and synchronously apply a time-domain alignment window function to multi-modal data to ensure feature time synchronization.

[0051] Step S2: Input the preprocessed multi-modal data into a pre-trained charging pile fault probability prediction model, and output a fault probability value.

[0052] By inputting the preprocessed multi-modal data into a pre-trained deep learning model to generate a fault probability value, the joint learning ability of the model for temporal and spatial features can be utilized to capture the correlation patterns between abnormal electrical parameters, temperature mutations, and sound signal distortions, realize the quantitative assessment of early fault risks, and provide a reliable basis for subsequent early warning decisions.

[0053] In some specific embodiments, the charging pile fault probability prediction model is constructed based on a deep learning algorithm model. The deep learning algorithm model includes one or a combination of LSTM network models, CNN network models, and GRU network models. The output layer of the charging pile fault probability prediction model uses the Sigmoid function to generate a fault probability value.

[0054] By adopting combinations of deep learning models such as LSTM, CNN, and GRU and a Sigmoid output layer, the temporal dependence relationship and spatial local features of multi-modal data can be simultaneously extracted, and the correlation model between complex working conditions and fault probability can be established using the non-linear mapping ability, significantly improving the processing effect of high-dimensional heterogeneous data.

[0055] In some specific embodiments, the training steps of the charging pile fault probability prediction model include: S201. Collect historical data of multi-modal data as samples. After performing the same preprocessing as in step S1, add real labels to the preprocessed multi-modal data according to the actual fault occurrence situation, and then divide it into a training set and a validation set; S202. Construct a charging pile fault probability prediction model, input the training set for training, and output the fault probability prediction values of each sample; S203. Calculate the loss function based on the difference between the fault probability prediction value and the real label, and update the parameters of the charging pile fault probability prediction model through the backpropagation algorithm; S204. Use the validation set to verify the prediction accuracy of the charging pile fault probability prediction model. When the prediction accuracy reaches the predetermined standard, stop training to obtain a pre-trained charging pile fault probability prediction model.

[0056] Through historical data annotation and iterative training of the validation set, the generalization ability of the model under different working conditions can be ensured, and the feature weight allocation can be dynamically optimized using the backpropagation algorithm, solving the problem that traditional threshold setting methods are difficult to adapt to dynamic factors such as equipment aging and environmental changes.

[0057] In some specific embodiments, the charging pile fault probability prediction model is based on a combined structure of an LSTM network model and a CNN network model. The input layer is configured with three parallel channels: the time series data of electrical data is input into a double-layer LSTM network to extract time series features, the temperature data at each position is extracted by the first CNN network to obtain the thermal field distribution features, and the audio frequency domain signal data is extracted by the second CNN network to obtain the spectral features; The middle layer uses a feature splicing method to fuse the outputs of the three parallel channels. Specifically, the time series features of electrical data are tensor-spliced with the thermal field distribution features of temperature data to obtain the main feature stream, and the spectral features of the audio frequency domain signal are fused with the main feature stream through attention weighting to obtain the integrated features; The output layer inputs the integrated features into a fully connected network, and the Sigmoid function is used at the end to generate probability values.

[0058] In some specific embodiments, the loss function uses the binary cross-entropy loss function to measure the difference between the fault probability prediction value output by the Sigmoid function and the real fault label (0 represents normal, 1 represents fault), and effectively quantifies the discriminative ability of the model for positive and negative samples by calculating the cross-entropy between the prediction distribution and the real distribution.

[0059] In step S3, compare the fault probability value with the set threshold to determine whether there is a fault risk. If there is a fault risk, issue a fault warning.

[0060] By setting a threshold to compare the failure probability value and trigger a warning, a dynamic risk identification mechanism based on a probability threshold can be established, intervening in protective measures in advance before a failure occurs, avoiding the problem of protective lag caused by the setting of a fixed threshold in traditional methods, and significantly reducing the probability of equipment damage and safety hazards.

[0061] In some specific embodiments, the set threshold is 92%.

[0062] Step S4, when a failure occurs, classify and locate the failure according to multimodal data, determine the failure level, perform hierarchical protection based on the failure level, and update the parameters of the charging pile failure probability prediction model using the multimodal data at the time of failure.

[0063] Through the failure classification, location and hierarchical protection mechanism based on multimodal data fusion, different differential disposal strategies can be matched according to different failure types and severity levels, achieving a balance between safety and equipment integrity during the failure handling process, and continuously optimizing the failure prediction accuracy through online model updates.

[0064] In some specific embodiments, when the multimodal data meets at least one failure condition, it is determined that a failure has occurred. The failure conditions include: The ratio of the equivalent resistance to the reference value exceeds the preset threshold, and the equivalent resistance is calculated based on the current and voltage in the electrical data; The temperature change rate of any key heat - generating part exceeds the preset threshold, and the temperature change rate is calculated based on the temperature at the current moment and the temperature during a fixed period of time before the current moment; The amplitude ratio of the sound frequency - domain signal to the background noise in a specific frequency band exceeds the preset threshold; Among them, the equivalent resistance is calculated according to Ohm's law calculate, is the charging current, is the charging voltage; In addition, according to the heat transfer formula calculate the heat transfer efficiency Q to assist in analyzing the abnormal temperature situation. In the formula, is the material thermal conductivity, is the heat transfer area, is the temperature difference, is the heat transfer distance; By setting three types of failure determination conditions, namely the equivalent resistance ratio, temperature change rate, and sound amplitude ratio, a multi - dimensional failure identification system based on electrical characteristics, thermodynamic characteristics, and mechanical characteristics can be established, effectively distinguishing different failure types such as overload, short - circuit, and poor contact, and improving the completeness of failure diagnosis.

[0065] In some specific embodiments, the determination criteria for the failure level are: Calculate the ratios of the equivalent resistance to the reference value, the temperature change rate, and the ratio of the amplitude of the audio frequency domain signal to the background noise in a specific frequency band to the preset threshold respectively. When the highest value among the three is ∈[1, 2), it is determined as a mild fault; when the highest value among the three is ∈[2, 5), it is determined as a moderate fault; when the highest value among the three is ≥5, it is determined as a severe fault.

[0066] By dividing the ratios of the three types of characteristic parameters into mild, moderate, and severe fault levels, a quantitative grading standard for the severity of faults can be established, providing an accurate trigger basis for the hierarchical protection strategy and avoiding the problems of overprotection or underprotection in the traditional single protection mechanism.

[0067] In some specific embodiments, the hierarchical protection includes: A PTC-liquid metal composite switch is provided in the main charging circuit of the charging pile, and a PTC resistance part and a liquid metal part are provided in the PTC-liquid metal composite switch; In case of a mild fault, the PTC resistance part dynamically increases the resistance value based on the real-time temperature increase amount to limit the current growth; In case of a moderate fault, the PTC resistance part and the liquid metal part work together to further increase the resistance; In case of a severe fault, the fuse is triggered to cut off the main circuit.

[0068] Through the hierarchical response design of the PTC-liquid metal composite switch, flexible current suppression can be achieved in case of a mild fault, current limiting protection can be strengthened in case of a moderate fault, and rapid fusing can be achieved in case of a severe fault, forming a gradually strengthened protection system, making the switch operation more stable during the normal and fault switching process, reducing the impact on the protection device, effectively extending the life of the protection device, and reducing the operation and maintenance cost.

[0069] Step S5, after the hierarchical protection is completed, corresponding recovery measures are executed according to the fault classification and location.

[0070] By executing recovery measures matching the fault type, the software system function can be quickly restored or the hardware fault component can be isolated, avoiding the problem of too long service interruption time caused by the traditional overall power-off restart, and significantly improving the charging pile fault recovery efficiency and system availability.

[0071] In some specific embodiments, the fault categories include software faults and hardware faults.

[0072] By distinguishing the recovery methods for software faults and hardware faults, rapid program recovery can be adopted for software anomalies, and component isolation and precise repair can be implemented for hardware faults, avoiding the risk of fault expansion caused by blind restart in traditional fault handling.

[0073] In some specific embodiments, fault classification is achieved by analyzing the correlation between the mutation of electrical data and the sound frequency domain signal; fault location is completed by comparing the temperature gradient distribution of each key heat - generating part before and after the fault and the time stamp of the energy mutation of the sound signal data.

[0074] In some specific embodiments, the recovery measures include: For software program faults, call the backup program and data to restore the system operation; For hardware faults, if the faulty part is an electronic component that can be automatically reset, perform automatic reset; If the faulty part is an electronic component that cannot be automatically reset or an electronic component that fails after automatic reset, disconnect the connection between the electronic component and the system inside the charging pile, and generate a fault report. The content of the fault report includes the name, model, fault cause analysis, and maintenance suggestions of the faulty component.

[0075] Through the automatic reset and fault report generation mechanism for hardware faults, while ensuring the basic operation ability of the system, it can automatically record the detailed information of the faulty components and generate a maintenance plan, greatly shortening the fault troubleshooting time and improving the maintenance efficiency.

[0076] In a specific embodiment, the charging pile fault protection method includes: Step S1, continuously and real - time collect multi - modal data of the charging pile. The multi - modal data includes the electrical data, temperature data, and sound signal data of the charging pile, where: The electrical data includes the charging current and charging voltage. The acquisition method is to connect current sensors and voltage sensors to the charging circuit to obtain the charging current in real - time and charging voltage , and according to Ohm's law continuously calculate the real - time value of the equivalent resistance of the charging circuit , record a set every 0.1 s , and the calculated value; The temperature data includes the temperature of the key heat - generating parts inside the charging pile. The key heat - generating parts include the power module, transformer, and charging interface. The acquisition method is to arrange temperature sensors at the key heat - generating parts and continuously record the temperature at the location every 0.5 s , and according to the heat transfer formula , combined with the thermal conductivity , heat transfer area and the heat transfer distance between components of each part of the charging pile, analyze the temperature change trend and abnormal heat transfer conditions; The sound signal data includes the system operation sound signals of the key parts with abnormal sounds inside the charging pile. The key parts with abnormal sounds include relays and contactors. The acquisition method is to install highly sensitive sound sensors near the key parts with abnormal sounds and collect the sound signals during operation in real time. ; Preprocess the multi-modal data at the current moment, including: Perform Fourier transform on the sound signal data in the 10 s before the current moment to convert it into a sound frequency domain signal. The formula is:

[0077] In the formula, is the sound signal data, is the imaginary unit, is the frequency, is the time; Perform data cleaning and denoising on the electrical data, temperature data, and sound frequency domain signal. The specific operations for data cleaning are: Electrical data cleaning: Perform sliding window mean filtering on the current and voltage signals to eliminate high-frequency interference, detect abnormal fluctuation points using the quartile method, fill in missing values by linear interpolation of the front and back valid data, synchronize and align the voltage and current signals with different sampling rates, and finally perform maximum-minimum normalization to ensure data scale consistency; Temperature data cleaning: Use the moving average algorithm to eliminate sensor jitter noise, set the temperature change rate threshold to filter out sudden change interference points, fill in abnormal breakpoint data with the mean value of adjacent time periods, perform spatial interpolation on multi-temperature measurement point data to compensate for the temperature values in the missing area, and enhance data continuity through exponential smoothing processing; Sound frequency domain signal cleaning: Perform band-stop filtering on the spectrum after Fourier transform to eliminate power frequency harmonic interference, use independent component analysis to separate environmental background noise, perform threshold truncation processing on the sudden amplitude increase points in specific frequency bands, fill in the missing frequency band data with weighted interpolation of adjacent frequency components, and finally standardize and balance the spectrum energy distribution.

[0078] The specific operations for denoising are: Use Kalman filtering for the electrical data to dynamically track the trend components of the current and voltage, use wavelet transform to decompose and then reconstruct the low-frequency effective components for the temperature data, suppress random pulse noise for the sound signal through an adaptive filter, and synchronously apply a time domain alignment window function to the multi-modal data to ensure feature time synchronization; Step S2, unify the format of the preprocessed multi-modal data into an input vector X = [I, V, T, S(f)], input it into the pre-trained charging pile fault probability prediction model, and output the fault probability value P; The charging pile failure probability prediction model is constructed based on the LSTM network model and the CNN network model. The output layer of the charging pile failure probability prediction model uses the Sigmoid function to generate the failure probability value; The training steps of the charging pile failure probability prediction model include: S201. Collect historical data of multi-modal data as samples. After performing the same preprocessing as in step S1, add real labels to the preprocessed multi-modal data according to the actual failure occurrence situation, and then divide it into a training set and a validation set; S202. Construct a charging pile failure probability prediction model, input the training set for training, and output the failure probability prediction values of each sample; Among them, the charging pile failure probability prediction model is based on the combined structure of the LSTM network model and the CNN network model. The input layer is configured with three parallel channels: the time series data of electrical data is input into a double-layer LSTM network to extract time series features, the temperature data at each position is extracted by the first CNN network to obtain the thermal field distribution features, and the sound frequency domain signal data is extracted by the second CNN network to obtain the spectrum features; The middle layer uses the feature splicing method to fuse the outputs of the three parallel channels. Specifically, the time series features of electrical data are tensor-spliced with the thermal field distribution features of temperature data to obtain the main feature stream, and the spectrum features of the sound frequency domain signal are fused with the main feature stream through attention weighting to obtain the integrated features; The output layer inputs the integrated features into a fully connected network, and the Sigmoid function is used at the end to generate the probability value; The loss function of the charging pile failure probability prediction model uses the cross-entropy loss function; S203. Calculate the loss function based on the difference between the failure probability prediction value and the real label, and update the parameters of the charging pile failure probability prediction model through the backpropagation algorithm; S204. Use the validation set to verify the prediction accuracy of the charging pile failure probability prediction model. When the prediction accuracy reaches the predetermined standard, stop training to obtain the pre-trained charging pile failure probability prediction model; In step S3, compare the failure probability value with the set threshold to determine whether there is a failure risk. The set threshold is 92%. When the failure probability value is greater than the set threshold, it is determined that there is a failure risk. If there is a failure risk, a failure warning is issued; In step S4, when a failure occurs, classify and locate the failure based on the multi-modal data, determine the failure level, perform hierarchical protection based on the failure level, and update the charging pile failure probability prediction model using the multi-modal data at the time of failure; Among them, when the multi-modal data meets at least one failure condition, it is determined that a failure has occurred. The failure conditions include: The ratio of the equivalent resistance to the reference value exceeds a preset threshold, and the equivalent resistance is calculated based on the current and voltage in the electrical data; The temperature change rate of any key heat - generating part exceeds the preset threshold, and the temperature change rate is calculated based on the temperature at the current moment and the temperature during a fixed period before the current moment; The amplitude ratio of the sound frequency - domain signal to the background noise in a specific frequency band exceeds the preset threshold; Among them, the equivalent resistance is calculated according to Ohm's law calculate is the charging current is the charging voltage; In addition, according to the heat transfer formula calculate the heat transfer efficiency Q to assist in analyzing abnormal temperature conditions. In the formula, is the thermal conductivity of the material is the heat transfer area is the temperature difference is the heat transfer distance; The determination criteria for the fault level are as follows: Calculate the ratios between the ratio of the equivalent resistance to the reference value, the temperature change rate, and the amplitude ratio of the sound frequency - domain signal to the background noise in a specific frequency band and the preset threshold respectively. When the highest value among the three is ∈[1, 2), it is determined as a minor fault. When the highest value among the three is ∈[2, 5), it is determined as a medium - level fault. When the highest value among the three is ≥5, it is determined as a severe fault; The hierarchical protection includes: Set a PTC - liquid metal composite switch in the main charging circuit of the charging pile. The PTC - liquid metal composite switch is provided with a PTC resistance part and a liquid metal part; In case of a minor fault, the PTC resistance part dynamically increases the resistance value based on the real - time temperature increase amount to limit the current growth; In case of a medium - level fault, the PTC resistance part and the liquid metal part work together to further increase the resistance; In case of a severe fault, trigger the fuse to operate and cut off the main circuit; Step S5, after the hierarchical protection is completed, execute corresponding recovery measures according to the fault classification and location; The fault categories include software faults and hardware faults; The fault classification is achieved by analyzing the correlation between the mutation of electrical data and the sound frequency - domain signal; the fault location is completed by comparing the temperature gradient distribution of each key heat - generating part before and after the fault and the time stamps of the energy mutation of the sound signal data; The recovery measures include: For software program faults, call the backup program and data to restore the system operation, and the recovery time is related to the backup data volume and the recovery speed related to, that is , shorten as much as possible , after the restoration is completed, conduct a comprehensive functional test on the system to ensure the normal operation of the software; For hardware failures, if the faulty part is an electronic component that can be automatically reset, after determining that it can perform self-recovery behavior, the system automatically executes the repair behavior and real-time monitors the performance indicators of the component after recovery; If the faulty part is an electronic component that cannot be automatically reset or an electronic component that fails after automatic reset, disconnect the connection between the electronic component and the system in the charging pile, ensure that the operation of the entire system is not affected as much as possible, and generate a fault report. The content of the fault report includes the name, model, fault cause analysis, and maintenance suggestions of the faulty component, guiding the operation and maintenance personnel to perform quick repairs.

[0079] The following are embodiments of the charging pile fault protection system provided by the present disclosure. The charging pile fault protection system and the charging pile fault protection methods of the above embodiments belong to the same inventive concept. Details not described in detail in the embodiments of the charging pile fault protection system can refer to the embodiments of the above charging pile fault protection methods.

[0080] Now, the mobile terminals implementing the various embodiments of the present application will be described with reference to the drawings. In the following description, suffixes such as "module", "component", or "unit" used to represent elements are only for the convenience of description of the embodiments of the present application, and they do not have a specific meaning by themselves. Therefore, "module" and "component" can be used interchangeably.

[0081] As Figure 2 shown, the charging pile fault protection system includes: A data acquisition and processing module, configured to continuously collect multi-modal data in real time and preprocess the multi-modal data at the current moment; A fault probability prediction module, configured to input the preprocessed multi-modal data into a pre-trained charging pile fault probability prediction model and output a fault probability value; A fault judgment and warning module, configured to compare the fault probability value with a set threshold to determine whether there is a fault risk, and issue a fault warning if there is a fault risk; A fault classification and protection execution module, configured to classify and locate the fault according to the multi-modal data when a fault occurs, determine the fault level, and perform hierarchical protection based on the fault level; A model update module, configured to update the parameters of the charging pile fault probability prediction model using the multi-modal data when a fault occurs; A recovery execution module, configured to execute corresponding recovery measures according to the fault classification and location after the hierarchical protection is completed.

[0082] The charging pile fault protection system of the integrated energy system in this embodiment is used to implement the charging pile fault protection method, and the steps include: S1. Continuously collect multi-modal data of the charging pile in real time. The multi-modal data includes electrical data, temperature data, and sound signal data of the charging pile, and preprocess the multi-modal data at the current moment; S2. Input the preprocessed multi-modal data into a pre-trained charging pile fault probability prediction model, and output a fault probability value; S3. Compare the fault probability value with a set threshold to determine whether there is a fault risk. If there is a fault risk, issue a fault warning; S4. When a fault occurs, classify and locate the fault based on the multi-modal data, determine the fault level, perform hierarchical protection based on the fault level, and update the charging pile fault probability prediction model using the multi-modal data when the fault occurs; S5. After the hierarchical protection is completed, execute corresponding recovery measures according to the fault classification and location.

[0083] This application also provides an electronic device for implementing each embodiment of this application. The electronic device includes a memory, a processor, and a computer program stored on the memory and executable on the processor.

[0084] Those skilled in the art can understand that the structure of the electronic device involved in the embodiments of this application does not constitute a limitation on the electronic device. The electronic device may include more or fewer components than shown in the figure, or combine some components, or have different component arrangements.

[0085] Figure 3 Schematic diagram of the hardware structure of an electronic device for implementing each embodiment of this application.

[0086] The electronic device includes, but is not limited to, components such as a processor and a memory. Those skilled in the art can understand that the structure of the electronic device involved in the embodiments of this application does not constitute a limitation on the electronic device. The electronic device may include more or fewer components than shown in the figure, or combine some components, or have different component arrangements.

[0087] In the embodiments of this application, the electronic device includes, but is not limited to, a laptop computer, a desktop computer, a workbench, a personal digital assistant, a server, a blade server, a mainframe computer, and other suitable computers. The electronic device may also represent various forms of mobile devices and other similar computing devices. The components shown in the figure, their connections and relationships, and their functions are only examples and are not intended to limit the implementation of the embodiments of this application described and / or required in this article.

[0088] In the embodiments of the present application, the processor may be implemented by using at least one of an application specific integrated circuit (ASIC), a digital signal processor (DSP), a digital signal processing device (DSPD), a processor, a controller, a microcontroller, a microprocessor, and an electronic unit designed to execute the functions described herein. In some cases, such an implementation may be implemented in the controller. For a software implementation, an implementation of a process or function may be implemented with a separate software module that allows execution of at least one function or operation. The software code may be implemented by a software application (or program) written in any appropriate programming language. The software code may be stored in a memory and executed by the controller.

[0089] In addition, the electronic device includes some functional modules not shown here, which will not be elaborated further.

[0090] Those skilled in the art can understand that various aspects of the electronic device provided in the present application can be implemented as a system, a method, or a program product. Therefore, various aspects of the present disclosure can be specifically implemented in the following forms, namely: a complete hardware implementation, a complete software implementation (including firmware, microcode, etc.), or an implementation combining hardware and software aspects, which can be collectively referred to as "circuit", "module", or "system" here.

[0091] The present application also provides a storage medium in which a program product capable of implementing the charging pile fault protection method is stored. In some possible implementation manners, various aspects of the present disclosure can also be implemented in the form of a program product, which includes program code. When the program product runs on a terminal device, the program code is used to cause the terminal device to execute the steps according to various exemplary embodiments of the present disclosure described in the above "Exemplary Method" section of this specification.

[0092] The storage medium may adopt any combination of one or more readable media. The readable media may be a readable signal medium or a readable storage medium. The readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the above. More specific examples (non-exhaustive list) of the readable storage medium include: an electrical connection with one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0093] The foregoing description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Thus, the present application is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for protecting against charging pile failures, characterized in that, Including: S1. Continuously and real-timely collect multimodal data of the charging pile. The multimodal data includes electrical data, temperature data, and sound signal data of the charging pile, and preprocess the multimodal data at the current moment; S2. Input the preprocessed multimodal data into a pre-trained charging pile fault probability prediction model, and output a fault probability value; S3. Compare the fault probability value with a set threshold to determine whether there is a fault risk. If there is a fault risk, issue a fault warning; S4. When a fault occurs, classify and locate the fault based on the multimodal data, determine the fault level, perform hierarchical protection based on the fault level, and update the parameters of the charging pile fault probability prediction model using the multimodal data at the time of the fault; S5. After the hierarchical protection is completed, perform corresponding recovery measures according to the fault classification and location.

2. The charging pile fault protection method according to claim 1, characterized in that, In step S1, the electrical data includes charging current and charging voltage; The temperature data includes the temperatures of key heat-generating parts inside the charging pile. The key heat-generating parts include power modules, transformers, and charging interfaces; The sound signal data includes the system operation sound signals of key parts with abnormal sounds inside the charging pile. The key parts with abnormal sounds include relays and contactors; The preprocessing includes: Perform Fourier transform on the sound signal data for a fixed duration before the current moment to convert it into a sound frequency domain signal; Perform data cleaning and denoising processing on the electrical data, temperature data, and sound frequency domain signal.

3. The charging pile fault protection method according to claim 1, wherein, In step S2, the charging pile fault probability prediction model is constructed based on a deep learning algorithm model. The deep learning algorithm model includes one or a combination of LSTM network models, CNN network models, and GRU network models. The output layer of the charging pile fault probability prediction model uses the Sigmoid function to generate a fault probability value.

4. The charging pile fault protection method according to claim 1, characterized in that, The training steps of the charging pile fault probability prediction model include: S201. Collect historical data of multimodal data as samples. After performing the same preprocessing as in step S1, add real labels to the preprocessed multimodal data according to the actual fault occurrence situation, and then divide it into a training set and a validation set; S202. Construct a charging pile fault probability prediction model, input the training set for training, and output the fault probability prediction values of each sample; S203. Calculate the loss function based on the difference between the fault probability prediction value and the real label, and update the parameters of the charging pile fault probability prediction model through the backpropagation algorithm; S204. Use the validation set to verify the prediction accuracy of the charging pile fault probability prediction model. When the prediction accuracy reaches a predetermined standard, stop training to obtain a pre-trained charging pile fault probability prediction model.

5. The charging pile fault protection method according to claim 2, wherein In step S4, when the multimodal data meets at least one fault condition, it is determined that a fault has occurred. The fault conditions include: The ratio of the equivalent resistance to the reference value exceeds a preset threshold, and the equivalent resistance is calculated based on the current and voltage in the electrical data; The temperature change rate of any key heat-generating part exceeds a preset threshold, and the temperature change rate is calculated based on the temperature at the current moment and the temperature for a fixed duration before the current moment; The amplitude ratio of the sound frequency domain signal to the background noise in a specific frequency band exceeds a preset threshold; The determination criteria for the fault level are: Calculate the ratio of the equivalent resistance to the reference value, the temperature change rate, and the ratio between the amplitude ratio of the sound frequency domain signal and the background noise in a specific frequency band and the preset threshold respectively. When the highest value among the three is ∈[1, 2), it is determined as a minor fault. When the highest value among the three is ∈[2, 5), it is determined as a moderate fault. When the highest value among the three is ≥5, it is determined as a severe fault.

6. The charging pile fault protection method according to claim 5, characterized in that, The hierarchical protection in step S4 includes: Set a PTC-liquid metal composite switch in the main charging circuit of the charging pile. The PTC-liquid metal composite switch is provided with a PTC resistance part and a liquid metal part; In case of a minor fault, the PTC resistance part dynamically increases the resistance value based on the real-time temperature increase amount to limit the current growth; In case of a moderate fault, the PTC resistance part and the liquid metal part work together to further increase the resistance; In case of a severe fault, trigger the fuse to operate and cut off the main circuit.

7. The charging pile fault protection method according to claim 5, characterized in that, In step S5, the fault categories include software program faults and hardware faults, and the recovery measures include: For software program faults, call the backup program and data to restore the system operation; For hardware faults, if the faulty part is an electronic component that can be automatically reset, perform automatic reset; If the faulty part is an electronic component that cannot be automatically reset or an electronic component that fails after automatic reset, disconnect the connection between the electronic component and the system in the charging pile, and generate a fault report. The content of the fault report includes the name and model of the faulty component, the analysis of the fault cause, and the maintenance suggestions.

8. A charging pile fault protection system, characterized in that, For implementing the charging pile fault protection method as described in any one of claims 1-7, it includes: A data acquisition and processing module, which is used to continuously collect multi-modal data in real time and preprocess the multi-modal data at the current moment; A fault probability prediction module, which is used to input the preprocessed multi-modal data into a pre-trained charging pile fault probability prediction model and output a fault probability value; A fault judgment and early warning module, which is used to compare the fault probability value with a set threshold to judge whether there is a fault risk. If there is a fault risk, issue a fault early warning; A fault classification and protection execution module, which is used to classify and locate the fault according to the multi-modal data when a fault occurs, determine the fault level, and perform hierarchical protection based on the fault level; A model update module, which is used to update the parameters of the charging pile fault probability prediction model using the multi-modal data when a fault occurs; A recovery execution module, which is used to execute corresponding recovery measures according to the fault classification and location after the hierarchical protection is completed.

9. An electronic device, characterized in that, It includes a memory, a processor, and a computer program stored on the memory and executable on the processor. The processor is used to implement the steps of the charging pile fault protection method as described in any one of claims 1-7 when executing the computer program.

10. A storage medium, characterized in that, A computer program is stored on a storage medium. When the computer program is executed by a processor, it implements the steps of the charging pile fault protection method as described in any one of claims 1-7.

Images