A method and system for constructing an electric vehicle charging safety feature database

Abstract

The present invention discloses a method and system for constructing an electric vehicle charging safety feature database, the method comprising: collecting vehicle-side data, charging-side data and environmental data during the charging process of the electric vehicle through a data acquisition device and uploading them to a server; the server classifies a variety of characteristic factors based on the vehicle-side data, charging-side data and environmental data; constructing a big data knowledge graph according to the classification results of the various characteristic factors; extracting and storing data according to the big data knowledge graph to obtain an electric vehicle charging safety feature database; the method can provide strong support for the formation of a subsequent charging safety early warning model.

Description

Technical Field

[0001] The present invention relates to the field of electric vehicle safety technology, and in particular to a method and system for constructing an electric vehicle charging safety feature database. Background Art

[0002] For the charging process of electric vehicles, the factors affecting charging safety can be divided into four aspects: safety factors on the grid side, safety factors on the charging equipment side, safety factors on the vehicle side, and safety factors on the monitoring system. At the same time, charging safety is also affected by spatiotemporal behavioral factors such as time and space uncertainty, equipment uncertainty, and vehicle uncertainty of electric vehicle charging; the rationality of charging safety is also particularly important for guiding electric vehicles to charge in an orderly manner, alleviating load tension in local areas, and reducing grid load during peak hours; in addition, deficiencies in equipment, technology, monitoring, and management will also cause unsafe vehicle-grid interaction processes. How to analyze the various complex factors that affect vehicle-grid interaction safety, provide basic services for studying the vehicle-grid interaction safety indicator system, designing a comprehensive evaluation method for the overall operating performance of each charging device, and establishing a charging safety early warning model, has become an urgent problem to be solved.

[0003] In the face of the above-mentioned complex influencing factors, the traditional data processing and safety influencing factor analysis methods can no longer meet the needs of electric vehicle charging safety. Therefore, it is necessary to introduce various intelligent algorithms and big data model processing methods. For example, patent document CN116039433B proposes a vehicle charging safety detection system and method based on big data, including: obtaining the battery capacity specifications, remaining power and travel data of the target vehicle; obtaining the historical charging records of the target vehicle; obtaining the location information and time point data when the target vehicle is connected to the charging pile; obtaining the collected data for encrypted storage; obtaining the historical charging times and all charging durations of the target vehicle, analyzing the current battery capacity of the target vehicle, and analyzing the current range required for charging; calculating the different charging modes and corresponding charging durations of the target vehicle; analyzing the historical charging current records of the target vehicle, and conducting a comprehensive analysis of the charging mode; and conducting safety detection on the charging current, voltage and temperature of the target vehicle after it is connected to the charging pile. This method has limited integrity in charging safety data collection, involving only partial battery information, charging location time information and historical charging records, and cannot monitor and collect charging safety data in all stages and aspects; it only considers vehicle-side influencing factors, and the analysis of safety influencing factors is not comprehensive and intuitive, which is not conducive to the subsequent safety warning work. For example, patent document CN116680553A discloses a source information fusion electric vehicle charging safety monitoring and graded warning method, including the steps of: collecting multi-source information such as charging equipment data, electric vehicle data, environmental data, infrared thermal imaging monitoring video information flow, etc. during electric vehicle charging; classifying the collected multi-source information for screening, denoising, dimensionality reduction and normalization, and dividing it into training set and test set; using deep learning method to construct a new type of electric vehicle charging dynamic warning network, and using group adaptive optimization algorithm to determine its hyperparameters; training the dynamic warning network through the divided data set, and verifying the effectiveness of the network through the test set; deploying the trained network on the system monitoring cloud platform, and using multi-source information fusion technology to perform real-time safety monitoring and graded warning of the charging process of electric vehicles. This method does not collect charging pile operation data, and the environmental data only considers temperature, ignoring the impact of other environmental factors such as humidity on charging safety; the analysis of charging safety influencing factors only stays at the charging site, without considering other factors such as network testing and platform side, which affects the reliability of early warning. The charging safety feature database has not been formed, which is not conducive to data classification and storage, affecting the data demand for future algorithm training, and thus affecting the accuracy of early warning. Summary of the invention

[0004] The present invention provides a method and system for constructing an electric vehicle charging safety feature database, which can provide support for the formation of a subsequent charging safety early warning model.

[0005] A method for constructing an electric vehicle charging safety feature database, comprising:

[0006] The data acquisition device is used to collect vehicle-side data, charging-side data, and environmental data during the charging process of the electric vehicle and upload the data to the server;

[0007] The server classifies multiple characteristic factors based on the vehicle-side data, charging-side data, and environmental data;

[0008] Construct a big data knowledge graph based on the classification results of multiple characteristic factors;

[0009] Data is extracted and stored according to the big data knowledge graph to obtain an electric vehicle charging safety feature database.

[0010] Furthermore, the vehicle-side data includes the state of charge of the vehicle power battery, the vehicle identification code, the charging voltage, the charging current, the battery temperature, the maximum allowable charging voltage / current of the single cell, the maximum allowable charging voltage / charging current of the entire battery, the voltage / current state of the single cell, the total voltage of the vehicle power battery, the nominal total energy of the vehicle power battery, the rated capacity of the vehicle power battery, the battery charging capacity, the battery type, the battery manufacturer, the battery pack serial number, the battery production date, the number of times the battery pack is charged, and the battery pack property identification from the electric vehicle BMS system;

[0011] The charging end data includes the charger serial number, charger number, charging station name, charger protocol version number, charger / charging station area code, charger maximum output voltage and minimum output voltage, charger maximum output current and minimum output current, charger voltage output value and current output value, charger charging capacity, charger input voltage / current, charger output voltage / current, charging equipment over-temperature rate, charging output over-current rate, charging output over-voltage rate, charging output under-voltage rate, maximum chargeable power, charger shutdown reason, charger cumulative fault times, charger fault reason, charging program abnormality rate and communication equipment fault rate;

[0012] The environmental data includes weather status data, charging gun temperature, pile body temperature, charger temperature, communication equipment temperature, smoke concentration inside the charger, and vehicle charging environment temperature and humidity.

[0013] Furthermore, the server classifies safety influencing factors based on the vehicle-side data, charging-side data, and environmental data, including:

[0014] Classify the vehicle-end data, charging-end data, and environmental data according to security features to obtain feature classification result data;

[0015] According to the vehicle-side data, charging-side data and environmental data, the spatiotemporal behavior factors of the electric vehicle are analyzed based on the kernel density estimation method to obtain the charging rules of the electric vehicle under different working conditions and obtain the charging rule data;

[0016] Perform early warning analysis based on the vehicle-side data, charging-side data, and environmental data to obtain early warning data;

[0017] The vehicle-end data, charging-end data, and environmental data are classified according to the safety process to obtain safety process classification result data.

[0018] Furthermore, the feature classification result data includes charging dynamic monitoring data, charging electrical parameter standard data, charging vehicle identification data, charger identification data, charging electrical safety data, charging pile fault data, temperature and humidity dynamic monitoring data and weather status data.

[0019] Furthermore, according to the vehicle-side data, charging-side data and environmental data, the spatiotemporal behavior factors of the electric vehicle are analyzed based on the kernel density estimation method to obtain the charging rules of the electric vehicle under different working conditions, including:

[0020] Preprocessing the vehicle-side data, charging-side data, and environmental data;

[0021] The least square method is used to fit the data of the highest output voltage of the charger, the lowest output voltage of the charger, the maximum output current of the charger, the minimum output current of the charger, the voltage output value of the charger, the current output value of the charger, the charging capacity of the charger, the voltage state of the single battery, and the current state of the single battery of the same electric vehicle after preprocessing under different weather conditions, and the fitting curves of the highest output voltage of the charger, the lowest output voltage of the charger, the maximum output current of the charger, the minimum output current of the charger, the voltage output value of the charger, the current output value of the charger, the charging capacity of the charger, the voltage state of the single battery, and the current state of the single battery of the same electric vehicle in different seasons are obtained;

[0022] Compare the obtained fitting curve with the target value curve provided by the car company to obtain the characteristic data offset in different time periods;

[0023] Setting a kernel function bandwidth, taking the kernel function bandwidth as a search interval, and counting the number of feature data offsets falling within the search interval through a sliding interval;

[0024] Determine the size of each grid of the output histogram, and calculate the density contribution value of each feature data offset to each grid in the sliding interval through the kernel function;

[0025] Assign a density value to each grid, which is the accumulation of the contribution value of each feature data offset to the grid density within the sliding interval of the grid;

[0026] Output the density value of each grid to form a probability density statistical histogram;

[0027] A probability density curve is obtained according to the probability density statistical histogram, and a kernel density estimation function is obtained when the probability density curve tends to be smooth. The characteristic data offset fitting curve in different seasons is calculated according to the kernel density estimation function to obtain the charging regularity data.

[0028] Furthermore, the kernel density estimation function is as follows:

[0029]

[0030] Among them, f n (x) is the kernel density estimation function, n is the sample size of the feature data offset, h is the bandwidth, x is the abscissa of the kernel density estimation function, and is a point in a continuous feature data offset interval. i is the offset of the i-th feature data in a seasonal time series sample, and k(.) represents the kernel function.

[0031] Furthermore, early warning analysis is performed based on the vehicle-side data, charging-side data, and environmental data to obtain early warning data, including:

[0032] According to the vehicle identification code, the number of times the same vehicle is charged at different times, as well as the charging voltage, charging current, charger voltage output value, charger current output value, single cell voltage state, battery temperature and charger charging capacity of each charge, are tracked, and early warning indicators are calculated, wherein the early warning indicators include charging voltage exceeding the maximum allowable charging voltage of the entire battery, charging current exceeding the maximum allowable charging current of the entire battery, charging overcurrent, vehicle battery overvoltage, charger output voltage out of tolerance, charger output current out of tolerance, single cell overvoltage, battery overtemperature, charging capacity out of range, battery imbalance, abnormal battery temperature rise and unchanged charging capacity;

[0033] Different thresholds are set respectively, and the calculated warning item indicators are compared with different thresholds to determine whether they exceed the thresholds, and obtain warning data when they exceed different thresholds.

[0034] Furthermore, the safety process classification results include manufacturer safety influencing factors, charging operator safety influencing factors and user safety influencing factors;

[0035] The big data knowledge graph is constructed using visualization software, and the vehicle-side data, charging-side data, environmental data, charging dynamic monitoring data, charging electrical parameter standard data, charging vehicle identification data, charger identification data, charging electrical safety data, charging pile failure data, temperature and humidity dynamic monitoring data, weather status data, charging regularity data, warning data under different thresholds, manufacturer safety influencing factors, charging operator safety influencing factors and user safety influencing factors are described according to entities, attributes and relationship elements.

[0036] Furthermore, the data acquisition device includes a control module, a BMS communication interface module, a DTU communication interface module, a 485 communication module, a debugging interface module, an analog input interface module and a digital input interface module; the BMS communication interface module, the DTU communication interface module, the 485 communication module, the power interface module, the debugging interface module, the analog input interface module and the digital input interface module are all connected to the control module, the BMS communication interface module is used to connect to the BMS system to collect vehicle-end data and charging-end data, the DTU communication interface module is used to connect to the DTU device to collect vehicle-end data and charging-end data, the 485 communication module is used to upload the vehicle-end data, charging-end data and environmental data to the server, the debugging interface module is used to debug the data acquisition device, and the analog input interface module and the digital input interface module are used to connect to external sensors to collect the environmental data.

[0037] A system for constructing an electric vehicle charging safety feature database applied to the above method comprises:

[0038] A data acquisition device, used to collect vehicle-side data, charging-side data, and environmental data during the charging process of the electric vehicle and upload them to a server;

[0039] The server classifies multiple characteristic factors based on the vehicle-side data, charging-side data and environmental data; constructs a big data knowledge graph according to the classification results of the multiple characteristic factors; extracts and stores data according to the big data knowledge graph to obtain an electric vehicle charging safety feature database.

[0040] Furthermore, the vehicle-side data includes the state of charge of the vehicle power battery, the vehicle identification code, the charging voltage, the charging current, the battery temperature, the maximum allowable charging voltage / current of the single cell, the maximum allowable charging voltage / charging current of the entire battery, the voltage / current state of the single cell, the total voltage of the vehicle power battery, the nominal total energy of the vehicle power battery, the rated capacity of the vehicle power battery, the battery charging capacity, the battery type, the battery manufacturer, the battery pack serial number, the battery production date, the number of times the battery pack is charged, and the battery pack property identification from the electric vehicle BMS system;

[0041] The charging end data includes the charger serial number, charger number, charging station name, charger protocol version number, charger / charging station area code, charger maximum output voltage and minimum output voltage, charger maximum output current and minimum output current, charger voltage output value and current output value, charger charging capacity, charger input voltage / current, charger output voltage / current, charging equipment over-temperature rate, charging output over-current rate, charging output over-voltage rate, charging output under-voltage rate, maximum chargeable power, charger shutdown reason, charger cumulative fault times, charger fault reason, charging program abnormality rate and communication equipment fault rate;

[0042] The environmental data includes weather status data, charging gun temperature, pile body temperature, charger temperature, communication equipment temperature, smoke concentration inside the charger, and vehicle charging environment temperature and humidity.

[0043] Furthermore, the server classifies safety influencing factors based on the vehicle-side data, charging-side data, and environmental data, including:

[0044] Classify the vehicle-end data, charging-end data, and environmental data according to security features to obtain feature classification result data;

[0045] According to the vehicle-side data, charging-side data and environmental data, the spatiotemporal behavior factors of the electric vehicle are analyzed based on the kernel density estimation method to obtain the charging rules of the electric vehicle under different working conditions and obtain the charging rule data;

[0046] Perform early warning analysis based on the vehicle-side data, charging-side data, and environmental data to obtain early warning data;

[0047] The vehicle-end data, charging-end data, and environmental data are classified according to the safety process to obtain safety process classification result data.

[0048] Furthermore, the feature classification result data includes charging dynamic monitoring data, charging electrical parameter standard data, charging vehicle identification data, charger identification data, charging electrical safety data, charging pile fault data, temperature and humidity dynamic monitoring data and weather status data.

[0049] Furthermore, the server analyzes the spatiotemporal behavior factors of the electric vehicle based on the vehicle-side data, the charging-side data and the environmental data based on the kernel density estimation method to obtain the charging rules of the electric vehicle under different working conditions, including:

[0050] Preprocessing the vehicle-side data, charging-side data, and environmental data;

[0051] The least square method is used to fit the data of the highest output voltage of the charger, the lowest output voltage of the charger, the maximum output current of the charger, the minimum output current of the charger, the voltage output value of the charger, the current output value of the charger, the charging capacity of the charger, the voltage state of the single battery, and the current state of the single battery of the same electric vehicle after preprocessing under different weather conditions, and the fitting curves of the highest output voltage of the charger, the lowest output voltage of the charger, the maximum output current of the charger, the minimum output current of the charger, the voltage output value of the charger, the current output value of the charger, the charging capacity of the charger, the voltage state of the single battery, and the current state of the single battery of the same electric vehicle in different seasons are obtained;

[0052] Compare the obtained fitting curve with the target value curve provided by the car company to obtain the characteristic data offset in different time periods;

[0053] Setting a kernel function bandwidth, taking the kernel function bandwidth as a search interval, and counting the number of feature data offsets falling within the search interval through a sliding interval;

[0054] Determine the size of each grid of the output histogram, and calculate the density contribution value of each feature data offset to each grid in the sliding interval through the kernel function;

[0055] Assign a density value to each grid, which is the accumulation of the contribution value of each feature data offset to the grid density within the sliding interval of the grid;

[0056] Output the density value of each grid to form a probability density statistical histogram;

[0057] A probability density curve is obtained according to the probability density statistical histogram, and a kernel density estimation function is obtained when the probability density curve tends to be smooth. The characteristic data offset fitting curve in different seasons is calculated according to the kernel density estimation function to obtain the charging regularity data.

[0058] Furthermore, the kernel density estimation function is as follows:

[0059]

[0060] Among them, f n (x) is the kernel density estimation function, n is the sample size of the feature data offset, h is the bandwidth, x is the abscissa of the kernel density estimation function, and is a point in a continuous feature data offset interval. i is the offset of the i-th feature data in a seasonal time series sample, and k(.) represents the kernel function.

[0061] Furthermore, the server performs early warning analysis based on the vehicle-side data, charging-side data, and environmental data to obtain early warning data, including:

[0062] According to the vehicle identification code, the number of times the same vehicle is charged at different times, as well as the charging voltage, charging current, charger voltage output value, charger current output value, single cell voltage state, battery temperature and charger charging capacity of each charge, are tracked, and early warning indicators are calculated, wherein the early warning indicators include charging voltage exceeding the maximum allowable charging voltage of the entire battery, charging current exceeding the maximum allowable charging current of the entire battery, charging overcurrent, vehicle battery overvoltage, charger output voltage out of tolerance, charger output current out of tolerance, single cell overvoltage, battery overtemperature, charging capacity out of range, battery imbalance, abnormal battery temperature rise and unchanged charging capacity;

[0063] Different thresholds are set respectively, and the calculated warning item indicators are compared with different thresholds to determine whether they exceed the thresholds, and obtain warning data when they exceed different thresholds.

[0064] Furthermore, the safety process classification results include manufacturer safety influencing factors, charging operator safety influencing factors and user safety influencing factors;

[0065] The server uses visualization software to construct the big data knowledge graph, and describes the vehicle-side data, charging-side data, environmental data, charging dynamic monitoring data, charging electrical parameter standard data, charging vehicle identification data, charger identification data, charging electrical safety data, charging pile failure data, temperature and humidity dynamic monitoring data, weather status data, charging regularity data, warning data under different thresholds, manufacturer safety influencing factors, charging operator safety influencing factors and user safety influencing factors according to entities, attributes and relationship elements.

[0066] Furthermore, the data acquisition device includes a control module, a BMS communication interface module, a DTU communication interface module, a 485 communication module, a debugging interface module, an analog input interface module and a digital input interface module; the BMS communication interface module, the DTU communication interface module, the 485 communication module, the power interface module, the debugging interface module, the analog input interface module and the digital input interface module are all connected to the control module, the BMS communication interface module is used to connect to the BMS system to collect vehicle-end data and charging-end data, the DTU communication interface module is used to connect to the DTU device to collect vehicle-end data and charging-end data, the 485 communication module is used to upload the vehicle-end data, charging-end data and environmental data to the server, the debugging interface module is used to debug the data acquisition device, and the analog input interface module and the digital input interface module are used to connect to external sensors to collect the environmental data.

[0067] The method and system for constructing an electric vehicle charging safety feature database provided by the present invention have at least the following beneficial effects:

[0068] The charging data acquisition device can collect multi-source data such as the charging pile itself, charging process, environment, etc., providing reliable data support for database construction; it can analyze the factors affecting charging safety from multiple angles such as vehicle-grid interaction, spatiotemporal behavior, charging safety, and the entire charging process, which is conducive to forming a more comprehensive safety early warning system; it uses big data knowledge graph analysis technology to analyze charging safety, so that safety influencing factors can be visualized, and can more intuitively understand the correspondence between safety feature quantities and safety influencing factors, which is conducive to the classification and extraction of safety feature quantities and the formation of a charging safety feature database; it can realize the analysis of multiple complex influencing factors of charging safety, and realize the classification and extraction of data from different sources and types. The construction of a charging safety feature database can provide strong support for the formation of subsequent charging safety early warning models. BRIEF DESCRIPTION OF THE DRAWINGS

[0069] Figure 1 The present invention provides a flowchart of an embodiment of a method for constructing an electric vehicle charging safety feature database.

[0070] Figure 2 The present invention provides a flow chart of an embodiment of a data acquisition device in a method for constructing an electric vehicle charging safety feature database.

[0071] Figure 3 The present invention provides a flowchart of an embodiment of analyzing the charging rules of electric vehicles in the method for constructing the electric vehicle charging safety feature database provided by the present invention.

[0072] Figure 4 A flowchart of an embodiment of a device for constructing an electric vehicle charging safety feature database provided by the present invention. DETAILED DESCRIPTION

[0073] In order to better understand the above technical solution, the above technical solution will be described in detail below in conjunction with the accompanying drawings and specific implementation methods.

[0074] refer to Figure 1 In some embodiments, a method for constructing an electric vehicle charging safety feature database is provided, comprising:

[0075] S1, collecting vehicle-side data, charging-side data and environmental data during the charging process of the electric vehicle through a data acquisition device and uploading them to a server;

[0076] S2. The server classifies safety influencing factors based on the vehicle-side data, charging-side data and environmental data;

[0077] S3. Construct a big data knowledge graph based on the classification results of security impact factors;

[0078] S4. Extract, classify and store security feature data based on the big data knowledge graph to obtain an electric vehicle charging safety feature database.

[0079] Specifically, refer to Figure 2 In some embodiments, the data acquisition device includes a control module 1, a BMS communication interface module 2, a DTU communication interface module 3, a 485 communication module 4, a debugging interface module 5, an analog input interface module 6, and a digital input interface module 7; the BMS communication interface module 2, the DTU communication interface module 3, the 485 communication module 4, the debugging interface module 5, the analog input interface module 6, and the digital input interface module 7 are all connected to the control module 1, the BMS communication interface module 2 is used to connect to the BMS system to collect vehicle-end data and charging-end data, the DTU communication interface module 3 is used to connect to the DTU device to collect vehicle-end data and charging-end data, the 485 communication module 4 is used to upload the vehicle-end data, charging-end data, and environmental data to a server, the debugging interface module 5 is used to debug the data acquisition device, and the analog input interface module 6 and the digital input interface module 7 are used to connect to external sensors to collect the environmental data.

[0080] Among them, the BMS communication interface module 2 is based on the CAN (Controller Area Network) communication protocol, which can perform two-way data transmission and is used to collect data from the charging end and the vehicle end during the charging process.

[0081] The DTU communication interface module 3 is a reserved interface. Some charging pile companies do not provide the vehicle-pile CAN communication bus interface, so they need to connect to the charging pile TCU module for data collection. The collected data is consistent with the BMS communication interface module 2.

[0082] Furthermore, the BMS communication interface module 2 and the DTU communication interface module 3 are based on the CAN (controller area network) communication protocol, and can perform two-way data transmission, which is used to collect vehicle-side data and charging-side data during the charging process; the 485 communication module 4 is used to output the parsed vehicle-side data and charging-side data. In some embodiments, the 485 serial communication protocol is used to communicate with the 4G module or the 5G module to achieve data upload and server query instructions; the debugging interface module 5 is used for device data acquisition code burning and address configuration and output data monitoring. In some embodiments, the device also includes a power module and an indicator light module. The power module is a 12V DC constant voltage power supply, and the indicator light module can display whether each interface function is working normally.

[0083] Specifically, the vehicle-side data includes the state of charge of the vehicle power battery, the vehicle identification code, the charging voltage, the charging current, the battery temperature, the maximum allowable charging voltage / current of the single cell, the maximum allowable charging voltage / charging current of the entire battery, the voltage / current state of the single cell, the total voltage of the vehicle power battery, the nominal total energy of the vehicle power battery, the rated capacity of the vehicle power battery, the battery charging capacity, the battery type, the battery manufacturer, the battery pack serial number, the battery production date, the number of times the battery pack is charged, and the battery pack property identification from the electric vehicle BMS system;

[0084] The charging end data includes the charger serial number, charger number, charging station name, charger protocol version number, charger / charging station area code, charger maximum output voltage and minimum output voltage, charger maximum output current and minimum output current, charger voltage output value and current output value, charger charging capacity, charger input voltage / current, charger output voltage / current, charging equipment over-temperature rate, charging output over-current rate, charging output over-voltage rate, charging output under-voltage rate, maximum chargeable power, charger shutdown reason, charger cumulative fault times, charger fault reason, charging program abnormality rate and communication equipment fault rate;

[0085] The environmental data includes weather status data, charging gun temperature, pile body temperature, charger temperature, communication equipment temperature, smoke concentration inside the charger, and vehicle charging environment temperature and humidity.

[0086] Environmental data is specifically completed by smoke sensors, temperature sensors, humidity sensors, infrared cameras, chassis scanning and other modules. The output of each module is input into the data acquisition device through the analog input and digital input module (see the hardware structure diagram of the data acquisition device) interface.

[0087] Furthermore, the server platform issues a query command to the data acquisition device, which parses the original charging message of the BMS or DTU input interface according to the communication protocol between the non-on-board conductive charger and the battery management system of GB / T 27930 electric vehicles, and outputs it to the 4G module through the 485 serial port. The parsed data message is uploaded to the server for storage and processing through the 4G module in the network transparent transmission mode, and the next query command issued by the server is received at the same time, and the command is transmitted to the data acquisition device through 485.

[0088] Further, refer to Figure 3 In step S2, the server classifies multiple characteristic factors based on the vehicle-side data, charging-side data and environmental data, including:

[0089] S21, classifying the vehicle-end data, charging-end data, and environmental data according to security features to obtain feature classification result data;

[0090] S22, analyzing the spatiotemporal behavior factors of the electric vehicle based on the vehicle-side data, the charging-side data and the environmental data based on the kernel density estimation method, obtaining the charging rules of the electric vehicle under different working conditions, and obtaining the charging rule data;

[0091] S23, performing early warning analysis based on the vehicle-side data, charging-side data, and environmental data to obtain early warning data;

[0092] S24, classify the vehicle-end data, charging-end data and environmental data according to the safety process to obtain safety process classification result data.

[0093] Specifically, in step S21, the feature classification results include charging dynamic monitoring data, charging electrical parameter standard data, charging vehicle identification data, charger identification data, charging electrical safety data, charging pile fault data, temperature and humidity dynamic monitoring data, and weather status data.

[0094] The dynamic charging monitoring data includes the charging capacity of the charger, the input voltage / current of the charger, the output voltage / current of the charger, the state of charge of the vehicle power battery, and the voltage / current state of the single battery;

[0095] The charging electrical parameter standard data includes the maximum allowable charging voltage / current of the single battery, the nominal total energy of the vehicle power battery and the rated capacity of the vehicle power battery system;

[0096] The charging vehicle identification data includes the vehicle identification code, battery type, battery manufacturer, battery pack serial number, battery production date, battery pack charging times and battery pack property identification;

[0097] The charger identification data includes the charger serial number, charger number, charging station name, charger protocol version number and charger / charging station location code;

[0098] The charging electrical safety data includes the charging equipment over-temperature rate, charging output over-current rate, charging output over-voltage rate, charging output under-voltage rate and maximum chargeable power;

[0099] The charging pile fault data includes the charging shutdown reason, the cumulative number of charger failures, the charger failure reason, the charging program abnormality rate and the communication equipment failure rate;

[0100] The temperature and humidity dynamic monitoring data include the charging gun temperature, pile body temperature, charger temperature, battery temperature, communication equipment temperature, smoke concentration inside the charger, and vehicle charging environment temperature and humidity.

[0101] Furthermore, in step S22, according to the vehicle-side data, charging-side data and environmental data, the spatiotemporal behavior factors of the electric vehicle are analyzed based on the kernel density estimation method to obtain the charging law of the electric vehicle under different working conditions, and obtain the charging law data, including:

[0102] S221, pre-processing the vehicle-end data, charging-end data, and environmental data;

[0103] S222, using the least squares method to fit the data of the highest output voltage of the charger, the lowest output voltage of the charger, the maximum output current of the charger, the minimum output current of the charger, the voltage output value of the charger, the current output value of the charger, the charging capacity of the charger, the voltage state of the single battery, and the current state of the single battery of the same electric vehicle after preprocessing under different weather conditions, to obtain fitting curves of the highest output voltage of the charger, the lowest output voltage of the charger, the maximum output current of the charger, the minimum output current of the charger, the voltage output value of the charger, the current output value of the charger, the charging capacity of the charger, the voltage state of the single battery, and the current state of the single battery of the same electric vehicle in different seasons;

[0104] S223, comparing the obtained fitting curve with the target value curve provided by the automobile manufacturer to obtain characteristic data offsets in different time periods;

[0105] S224, setting a kernel function bandwidth, taking the kernel function bandwidth as a search interval, and counting the number of feature data offsets falling within the search interval through a sliding interval;

[0106] S225, determining the size of each grid of the output bar graph, and calculating the density contribution value of each feature data offset to each grid in the sliding interval through a kernel function;

[0107] S226, assigning a density value to each grid, where the value assigned is the accumulation of contribution values ​​of each feature data offset within the sliding interval within the grid to the density of the grid;

[0108] S227, output the density value of each grid to form a probability density statistical histogram;

[0109] S228. Obtain a probability density curve according to the probability density statistical histogram, obtain a kernel density estimation function when the probability density curve tends to be smooth, calculate and obtain a characteristic data offset fitting curve in different seasons according to the kernel density estimation function, and obtain charging regularity data.

[0110] Specifically, in step S221, for parameters such as the charger's maximum output voltage, the charger's minimum output voltage, the charger's maximum output current, the charger's minimum output current, the charger's voltage output value, the charger's current output value, the charger's charging capacity, the battery's voltage status, and the battery's current status under different weather conditions, the missing, omitted, and abnormal values ​​are repaired using the mean method, arithmetic difference method, and median method to complete preprocessing.

[0111] Furthermore, in step S222, the characteristic data curve model is assumed to be y=f(x,θ), where y is the output, x is the input, and θ is the curve parameter. The criterion for estimating the parameters is the sum of square errors Q of the model, and the least squares method is to find the parameter estimate that minimizes Q. The basic formula is as follows:

[0112]

[0113] Among them, y i For x i The real data value collected under the condition is n, and n is the number of sampling points. The iterative algorithm is used to continuously calculate the Until the convergence condition is met, the parameter θ is obtained, thus obtaining the fitting curve.

[0114] Further, in step S228, the kernel density estimation function is as follows:

[0115]

[0116] Among them, f n (x) is the kernel density estimation function, n is the sample size of the feature data offset, h is the bandwidth, x is the abscissa of the kernel density estimation function, and is a point in a continuous feature data offset interval. i is the offset of the i-th characteristic data in a seasonal time series sample, and k(.) represents the kernel function. h is also called the smoothing parameter, which controls the relative weight of each sample by affecting the value of the independent variable in the kernel function, thereby affecting the accuracy of the fitted probability density curve. The kernel function usually selects a symmetric unimodal probability density function centered at 0.

[0117] Furthermore, in step S23, early warning analysis is performed based on the vehicle-side data, charging-side data, and environmental data to obtain early warning data, including:

[0118] According to the vehicle identification code, the number of times the same vehicle is charged at different times, as well as the charging voltage, charging current, charger voltage output value, charger current output value, single cell voltage state, battery temperature and charger charging capacity of each charge, are tracked, and early warning indicators are calculated, wherein the early warning indicators include charging voltage exceeding the maximum allowable charging voltage of the entire battery, charging current exceeding the maximum allowable charging current of the entire battery, charging overcurrent, vehicle battery overvoltage, charger output voltage out of tolerance, charger output current out of tolerance, single cell overvoltage, battery overtemperature, battery charging capacity out of range, battery imbalance, abnormal battery temperature rise and unchanged charging capacity;

[0119] Different thresholds are set respectively, and the calculated warning item indicators are compared with different thresholds to determine whether they exceed the thresholds, and obtain warning data when they exceed different thresholds.

[0120] As an optional implementation, three thresholds can be set: a first threshold, a second threshold, and a third threshold, where the first threshold is greater than the second threshold, and the second threshold is greater than the third threshold. Data greater than the first threshold, data greater than the second threshold and less than the first threshold, and data greater than the third threshold and less than the second threshold are stored separately.

[0121] It needs to be further explained that: as the number of charging times and charged vehicles increases, the database will gradually increase and will be classified according to the alarm level, vehicle type, trial period, location and time of occurrence, and used to count the characteristics of vehicles involved in accidents. As the number of charging times increases, the threshold may change.

[0122] Furthermore, in step S24, by analyzing the factors affecting the safety of the entire charging process, a safe charging strategy can be proposed, which mainly includes three aspects: equipment manufacturers, charging operators, and users. Manufacturers must do a good job in quality monitoring and equipment maintenance of charging equipment, charging operators must provide comprehensive operating services and safety management, and users must use charging equipment correctly according to specifications. Only when the three parties reach a consensus and reduce the probability of accidents during the charging process from all aspects can the charging safety of the entire life cycle of the charging process be guaranteed. Specifically:

[0123] Manufacturer safety influencing factors include: vehicle identification code, vehicle power battery state of charge from the electric vehicle BMS system, charging voltage, charging current, battery temperature, maximum allowable charging voltage / current of single cells, maximum allowable charging voltage / current of the vehicle battery, total voltage of the vehicle power battery, nominal total energy of the vehicle power battery, rated capacity of the vehicle power battery and battery charging capacity.

[0124] Factors affecting the safety of charging operators include: charging gun temperature, pile body temperature, charger temperature, communication equipment temperature, smoke concentration inside the charger, vehicle charging environment temperature and humidity, charger maximum and minimum output voltage, charger maximum and minimum output current, charger output voltage / current, charger charging capacity, and single cell battery voltage / current status.

[0125] Factors affecting user safety include: whether the user operates the charging equipment correctly and how many times the user maintains the vehicle properly.

[0126] Furthermore, in step S3, visualization software is used to construct the big data knowledge graph, and the vehicle-side data, charging-side data, environmental data, charging dynamic monitoring data, charging electrical parameter standard data, charging vehicle identification data, charger identification data, charging electrical safety data, charging pile failure data, temperature and humidity dynamic monitoring data, weather status data, charging regularity data, warning data under different thresholds, manufacturer safety influencing factors, charging operator safety influencing factors and user safety influencing factors are described according to entities, attributes and relationship elements.

[0127] Knowledge graphs can effectively display the relationships between different entities in the real world in a graphical way, and can realize semantic analysis and knowledge reasoning and updating, helping people to better understand information, discover knowledge, solve problems and make decisions. The construction of knowledge graphs has promoted the standardization and efficiency of data governance in big data processing.

[0128] According to the elements of knowledge graph such as entities, attributes, and relationships, as well as triple descriptions in the form of entities, attributes, and attribute values, we can correspond to the database name, table name, attribute name, attribute value and other information in the database and build it in the database.

[0129] The construction of knowledge graphs promotes the standardization and efficiency of data governance in big data processing, which is necessary. The database is constructed based on the triple descriptions of entities, attributes, attribute values, etc. in the knowledge graph, and is the specific data support for the contents of knowledge graph entities, attributes, and relationships.

[0130] Furthermore, in step S4, security feature data is extracted and classified according to the big data knowledge graph, and a MySQL database is used to store local or cloud data, thereby establishing a database of the above data and influencing factors.

[0131] refer to Figure 4 In some embodiments, a system for constructing an electric vehicle charging safety feature database applied to the above method is provided, comprising:

[0132] The data collection device 201 is used to collect vehicle-side data, charging-side data and environmental data during the charging process of the electric vehicle and upload them to the server;

[0133] Server 202 classifies safety influencing factors based on the vehicle-side data, charging-side data and environmental data; constructs a big data knowledge graph according to the classification results of safety influencing factors; extracts, classifies and stores safety feature data according to the big data knowledge graph to obtain an electric vehicle charging safety feature database.

[0134] Furthermore, the vehicle-side data includes the state of charge of the vehicle power battery, the vehicle identification code, the charging voltage, the charging current, the battery temperature, the maximum allowable charging voltage / current of the single cell, the maximum allowable charging voltage / charging current of the entire battery, the voltage / current state of the single cell, the total voltage of the vehicle power battery, the nominal total energy of the vehicle power battery, the rated capacity of the vehicle power battery, the battery charging capacity, the battery type, the battery manufacturer, the battery pack serial number, the battery production date, the number of times the battery pack is charged, and the battery pack property identification from the electric vehicle BMS system;

[0135] The charging end data includes the charger serial number, charger number, charging station name, charger protocol version number, charger / charging station area code, charger maximum output voltage and minimum output voltage, charger maximum output current and minimum output current, charger voltage output value and current output value, charger charging capacity, charger input voltage / current, charger output voltage / current, charging equipment over-temperature rate, charging output over-current rate, charging output over-voltage rate, charging output under-voltage rate, maximum chargeable power, charger shutdown reason, charger cumulative fault times, charger fault reason, charging program abnormality rate and communication equipment fault rate;

[0136] The environmental data includes weather status data, charging gun temperature, pile body temperature, charger temperature, communication equipment temperature, smoke concentration inside the charger, and vehicle charging environment temperature and humidity.

[0137] Furthermore, the server classifies safety influencing factors based on the vehicle-side data, charging-side data, and environmental data, including:

[0138] Classify the vehicle-end data, charging-end data, and environmental data according to security features to obtain feature classification result data;

[0139] According to the vehicle-side data, charging-side data and environmental data, the spatiotemporal behavior factors of the electric vehicle are analyzed based on the kernel density estimation method to obtain the charging rules of the electric vehicle under different working conditions and obtain the charging rule data;

[0140] Perform early warning analysis based on the vehicle-side data, charging-side data, and environmental data to obtain early warning data;

[0141] The vehicle-end data, charging-end data, and environmental data are classified according to the safety process to obtain safety process classification result data.

[0142] Furthermore, the feature classification result data includes charging dynamic monitoring data, charging electrical parameter standard data, charging vehicle identification data, charger identification data, charging electrical safety data, charging pile fault data, temperature and humidity dynamic monitoring data and weather status data.

[0143] Furthermore, the server analyzes the spatiotemporal behavior factors of the electric vehicle based on the vehicle-side data, the charging-side data and the environmental data based on the kernel density estimation method to obtain the charging rules of the electric vehicle under different working conditions, including:

[0144] Preprocessing the vehicle-side data, charging-side data, and environmental data;

[0145] The least square method is used to fit the data of the highest output voltage of the charger, the lowest output voltage of the charger, the maximum output current of the charger, the minimum output current of the charger, the voltage output value of the charger, the current output value of the charger, the charging capacity of the charger, the voltage state of the single battery, and the current state of the single battery of the same electric vehicle after preprocessing under different weather conditions, and the fitting curves of the highest output voltage of the charger, the lowest output voltage of the charger, the maximum output current of the charger, the minimum output current of the charger, the voltage output value of the charger, the current output value of the charger, the charging capacity of the charger, the voltage state of the single battery, and the current state of the single battery of the same electric vehicle in different seasons are obtained;

[0146] Compare the obtained fitting curve with the target value curve provided by the car company to obtain the characteristic data offset in different time periods;

[0147] Setting a kernel function bandwidth, taking the kernel function bandwidth as a search interval, and counting the number of feature data offsets falling within the search interval through a sliding interval;

[0148] Determine the size of each grid of the output histogram, and calculate the density contribution value of each feature data offset to each grid in the sliding interval through the kernel function;

[0149] Assign a density value to each grid, which is the accumulation of the contribution value of each feature data offset to the grid density within the sliding interval of the grid;

[0150] Output the density value of each grid to form a probability density statistical histogram;

[0151] A probability density curve is obtained according to the probability density statistical histogram, and a kernel density estimation function is obtained when the probability density curve tends to be smooth. The characteristic data offset fitting curve in different seasons is calculated according to the kernel density estimation function to obtain the charging regularity data.

[0152] Furthermore, the kernel density estimation function is as follows:

[0153]

[0154] Among them, f n (x) is the kernel density estimation function, n is the sample size of the feature data offset, h is the bandwidth, x is the abscissa of the kernel density estimation function, and is a point in a continuous feature data offset interval. i is the offset of the i-th feature data in a seasonal time series sample, and k(.) represents the kernel function.

[0155] Furthermore, the server performs early warning analysis based on the vehicle-side data, charging-side data, and environmental data to obtain early warning data, including:

[0156] According to the vehicle identification code, the number of times the same vehicle is charged at different times, as well as the charging voltage, charging current, charger voltage output value, charger current output value, single cell voltage state, battery temperature and charger charging capacity of each charge, are tracked, and early warning indicators are calculated, wherein the early warning indicators include charging voltage exceeding the maximum allowable charging voltage of the entire battery, charging current exceeding the maximum allowable charging current of the entire battery, charging overcurrent, vehicle battery overvoltage, charger output voltage out of tolerance, charger output current out of tolerance, single cell overvoltage, battery overtemperature, charging capacity out of range, battery imbalance, abnormal battery temperature rise and unchanged charging capacity;

[0157] Different thresholds are set respectively, and the calculated warning item indicators are compared with different thresholds to determine whether they exceed the thresholds, and obtain warning data when they exceed different thresholds.

[0158] Furthermore, the safety process classification results include manufacturer safety influencing factors, charging operator safety influencing factors and user safety influencing factors;

[0159] The server uses visualization software to construct the big data knowledge graph, and describes the vehicle-side data, charging-side data, environmental data, charging dynamic monitoring data, charging electrical parameter standard data, charging vehicle identification data, charger identification data, charging electrical safety data, charging pile failure data, temperature and humidity dynamic monitoring data, weather status data, charging regularity data, warning data under different thresholds, manufacturer safety influencing factors, charging operator safety influencing factors and user safety influencing factors according to entities, attributes and relationship elements.

[0160] Furthermore, the data acquisition device includes a control module, a BMS communication interface module, a DTU communication interface module, a 485 communication module, a debugging interface module, an analog input interface module and a digital input interface module; the BMS communication interface module, the DTU communication interface module, the 485 communication module, the power interface module, the debugging interface module, the analog input interface module and the digital input interface module are all connected to the control module, the BMS communication interface module is used to connect to the BMS system to collect vehicle-end data and charging-end data, the DTU communication interface module is used to connect to the DTU device to collect vehicle-end data and charging-end data, the 485 communication module is used to upload the vehicle-end data, charging-end data and environmental data to the server, the debugging interface module is used to debug the data acquisition device, and the analog input interface module and the digital input interface module are used to connect to external sensors to collect the environmental data.

[0161] The method and system for constructing an electric vehicle charging safety feature database provided in the above embodiment have at least the following beneficial effects:

[0162] The charging data acquisition device can collect multi-source data such as the charging pile body, charging process, and environment, providing reliable data support for database construction; it analyzes the factors affecting charging safety from multiple angles such as vehicle-grid interaction, spatiotemporal behavior, charging safety, and the entire charging process, which is conducive to forming a more comprehensive safety early warning system; it uses big data knowledge graph analysis technology to analyze charging safety, so that safety influencing factors can be visualized, and the corresponding relationship between safety feature quantities and safety influencing factors can be more intuitively understood, which is conducive to the classification and extraction of safety feature quantities and the formation of a charging safety feature database. This invention can realize the analysis of multiple complex influencing factors of charging safety, and realize the classification and extraction of data from different sources and types. The construction of a charging safety feature database can provide strong support for the formation of a subsequent charging safety early warning model.

[0163] Although preferred embodiments of the present invention have been described, additional changes and modifications may be made to these embodiments by those skilled in the art once the basic inventive concepts are known. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention. Obviously, those skilled in the art may make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A method for constructing an electric vehicle charging safety feature database, characterized in that: include: The data acquisition device is used to collect vehicle-side data, charging-side data, and environmental data during the charging process of the electric vehicle and upload the data to the server; The server classifies multiple characteristic factors based on the vehicle-side data, charging-side data, and environmental data; Construct a big data knowledge graph based on the classification results of multiple characteristic factors; Extract and store data according to the big data knowledge graph to obtain an electric vehicle charging safety feature database; The server classifies safety influencing factors based on the vehicle-side data, the charging-side data, and the environmental data, including: Classify the vehicle-end data, charging-end data, and environmental data according to security features to obtain feature classification result data; According to the vehicle-side data, charging-side data and environmental data, the spatiotemporal behavior factors of the electric vehicle are analyzed based on the kernel density estimation method to obtain the charging rules of the electric vehicle under different working conditions and obtain the charging rule data; Perform early warning analysis based on the vehicle-side data, charging-side data, and environmental data to obtain early warning data; Classify the vehicle-end data, charging-end data, and environmental data according to the safety process to obtain safety process classification result data; The feature classification result data includes charging dynamic monitoring data, charging electrical parameter standard data, charging vehicle identification data, charger identification data, charging electrical safety data, charging pile failure data, temperature and humidity dynamic monitoring data, and weather status data; The safety process classification results include manufacturer safety influencing factors, charging operator safety influencing factors and user safety influencing factors; The big data knowledge graph is constructed using visualization software, and the vehicle-side data, charging-side data, environmental data, charging dynamic monitoring data, charging electrical parameter standard data, charging vehicle identification data, charger identification data, charging electrical safety data, charging pile failure data, temperature and humidity dynamic monitoring data, weather status data, charging regularity data, warning data under different thresholds, manufacturer safety influencing factors, charging operator safety influencing factors and user safety influencing factors are described according to entities, attributes and relationship elements.

2. The method according to claim 1, characterized in that The vehicle-side data includes the state of charge of the vehicle power battery, the vehicle identification code, the charging voltage, the charging current, the battery temperature, the maximum allowable charging voltage / current of the single cell, the maximum allowable charging voltage / charging current of the entire battery, the voltage / current status of the single cell, the total voltage of the vehicle power battery, the nominal total energy of the vehicle power battery, the rated capacity of the vehicle power battery, the battery charging capacity, the battery type, the battery manufacturer, the battery pack serial number, the battery production date, the number of times the battery pack is charged, and the battery pack property identification from the electric vehicle BMS system; The charging end data includes the charger serial number, charger number, charging station name, charger protocol version number, charger / charging station area code, charger maximum output voltage and minimum output voltage, charger maximum output current and minimum output current, charger voltage output value and current output value, charger charging capacity, charger input voltage / current, charger output voltage / current, charging equipment over-temperature rate, charging output over-current rate, charging output over-voltage rate, charging output under-voltage rate, maximum chargeable power, charger shutdown reason, charger cumulative fault times, charger fault reason, charging program abnormality rate and communication equipment fault rate; The environmental data includes weather status data, charging gun temperature, pile body temperature, charger temperature, communication equipment temperature, smoke concentration inside the charger, and vehicle charging environment temperature and humidity.

3. The method according to claim 1, characterized in that According to the vehicle-side data, charging-side data and environmental data, the spatiotemporal behavior factors of electric vehicles are analyzed based on the kernel density estimation method to obtain the charging rules of electric vehicles under different working conditions, including: Preprocessing the vehicle-side data, charging-side data, and environmental data; The least square method is used to fit the data of the highest output voltage of the charger, the lowest output voltage of the charger, the maximum output current of the charger, the minimum output current of the charger, the voltage output value of the charger, the current output value of the charger, the charging capacity of the charger, the voltage state of the single battery, and the current state of the single battery of the same electric vehicle after preprocessing under different weather conditions, and the fitting curves of the highest output voltage of the charger, the lowest output voltage of the charger, the maximum output current of the charger, the minimum output current of the charger, the voltage output value of the charger, the current output value of the charger, the charging capacity of the charger, the voltage state of the single battery, and the current state of the single battery of the same electric vehicle in different seasons are obtained; Compare the obtained fitting curve with the target value curve provided by the car company to obtain the characteristic data offset in different time periods; Setting a kernel function bandwidth, taking the kernel function bandwidth as a search interval, and counting the number of feature data offsets falling within the search interval through a sliding interval; Determine the size of each grid of the output histogram, and calculate the density contribution value of each feature data offset to each grid in the sliding interval through the kernel function; Assign a density value to each grid, which is the accumulation of the contribution value of each feature data offset to the grid density within the sliding interval of the grid; Output the density value of each grid to form a probability density statistical histogram; A probability density curve is obtained according to the probability density statistical histogram, and a kernel density estimation function is obtained when the probability density curve tends to be smooth. The characteristic data offset fitting curve in different seasons is calculated according to the kernel density estimation function to obtain the charging regularity data.

4. The method according to claim 3, characterized in that The kernel density estimation function is as follows: Among them, f n (x) is the kernel density estimation function, n is the sample size of the feature data offset, h is the bandwidth, x is the abscissa of the kernel density estimation function, and is a point in a continuous feature data offset interval. i is the offset of the i-th feature data in a seasonal time series sample, and k(.) represents the kernel function.

5. The method according to claim 1, characterized in that Perform early warning analysis based on the vehicle-side data, charging-side data, and environmental data to obtain early warning data, including: According to the vehicle identification code, the number of times the same vehicle is charged at different times, as well as the charging voltage, charging current, charger voltage output value, charger current output value, single cell voltage state, battery temperature and charger charging capacity of each charge, are tracked, and early warning indicators are calculated, wherein the early warning indicators include charging voltage exceeding the maximum allowable charging voltage of the entire battery, charging current exceeding the maximum allowable charging current of the entire battery, charging overcurrent, vehicle battery overvoltage, charger output voltage out of tolerance, charger output current out of tolerance, single cell overvoltage, battery overtemperature, charging capacity out of range, battery imbalance, abnormal battery temperature rise and unchanged charging capacity; Different thresholds are set respectively, and the calculated warning item indicators are compared with different thresholds to determine whether they exceed the thresholds, and obtain warning data when they exceed different thresholds.

6. The method according to claim 1, characterized in that The data acquisition device includes a control module, a BMS communication interface module, a DTU communication interface module, a 485 communication module, a debugging interface module, an analog input interface module and a digital input interface module; the BMS communication interface module, the DTU communication interface module, the 485 communication module, the power interface module, the debugging interface module, the analog input interface module and the digital input interface module are all connected to the control module, the BMS communication interface module is used to connect to the BMS system to collect vehicle-end data and charging-end data, the DTU communication interface module is used to connect to the DTU device to collect vehicle-end data and charging-end data, the 485 communication module is used to upload the vehicle-end data, charging-end data and environmental data to a server, the debugging interface module is used to debug the data acquisition device, and the analog input interface module and the digital input interface module are used to connect to external sensors to collect the environmental data.

7. A system for constructing an electric vehicle charging safety feature database according to any of the methods described in claims 1-6, characterized in that: include: A data acquisition device, used to collect vehicle-side data, charging-side data, and environmental data during the charging process of the electric vehicle and upload them to a server; The server classifies multiple characteristic factors based on the vehicle-side data, the charging-side data, and the environmental data; Construct a big data knowledge graph based on the classification results of multiple characteristic factors; extract and store data based on the big data knowledge graph to obtain an electric vehicle charging safety feature database; The server classifies safety influencing factors based on the vehicle-side data, the charging-side data, and the environmental data, including: Classify the vehicle-end data, charging-end data, and environmental data according to security features to obtain feature classification result data; According to the vehicle-side data, charging-side data and environmental data, the spatiotemporal behavior factors of the electric vehicle are analyzed based on the kernel density estimation method to obtain the charging rules of the electric vehicle under different working conditions and obtain the charging rule data; Perform early warning analysis based on the vehicle-side data, charging-side data, and environmental data to obtain early warning data; Classify the vehicle-end data, charging-end data, and environmental data according to the safety process to obtain safety process classification result data; The feature classification result data includes charging dynamic monitoring data, charging electrical parameter standard data, charging vehicle identification data, charger identification data, charging electrical safety data, charging pile failure data, temperature and humidity dynamic monitoring data, and weather status data; The safety process classification results include manufacturer safety influencing factors, charging operator safety influencing factors and user safety influencing factors; The big data knowledge graph is constructed using visualization software, and the vehicle-side data, charging-side data, environmental data, charging dynamic monitoring data, charging electrical parameter standard data, charging vehicle identification data, charger identification data, charging electrical safety data, charging pile failure data, temperature and humidity dynamic monitoring data, weather status data, charging regularity data, warning data under different thresholds, manufacturer safety influencing factors, charging operator safety influencing factors and user safety influencing factors are described according to entities, attributes and relationship elements.

Images