Safety diagnosis method, device and equipment for new energy automobile and storage medium

Abstract

The invention discloses a safety diagnosis method, device and equipment for a new energy automobile and a storage medium. The method comprises the steps that battery state time sequence data and external environment data of the new energy automobile are collected through an edge end, and the collected data are preprocessed; calculating a thermal correlation characteristic value and an electrical correlation characteristic value based on the preprocessed data, and performing preliminary safety diagnosis and early warning judgment based on the thermal correlation characteristic value and the electrical correlation characteristic value; under the condition that the preliminary safety diagnosis result is that early warning is not triggered, calculating an overall risk score corresponding to the new energy vehicle based on a pre-constructed overall risk scoring model and a coupling risk knowledge base issued by the cloud; and determining an early warning level corresponding to the new energy vehicle based on the overall risk score and the preset risk threshold. The technical effects of improving the risk diagnosis accuracy and guaranteeing the safety response timeliness of the battery are achieved.

Description

Technical Field

[0001] This invention relates to the field of safety diagnostic technology for new energy vehicles, and in particular to a safety diagnostic method, device, equipment, and storage medium for new energy vehicles. Background Technology

[0002] With the rapid development of the new energy vehicle industry, the electrification and intelligence of vehicles are constantly improving, and their safety issues are becoming increasingly prominent. Core systems of new energy vehicles, such as high-energy-density battery systems, may experience sudden safety events during operation due to cell aging, thermal management failure, or physical impact, such as thermal runaway, short circuits, and insulation failures. If these issues are not diagnosed and addressed in a timely manner, they can easily lead to serious accidents.

[0003] Existing technologies have solved the problem that establishing safety early warning strategies is relatively simple due to the limited computing power of the BMS itself, making it difficult to achieve real-time control of battery system safety. However, they do not comprehensively consider the coupling effect of battery state time-series data and external environmental data, cannot accurately capture the role of the environment in battery risk, lack dynamic overall risk scoring and flexible hierarchical early warning mechanisms, and the early warning rules are mostly fixed settings. Furthermore, they rely excessively on cloud analysis in cloud-edge collaboration and lack independent real-time diagnostic and early warning rendering capabilities. Summary of the Invention

[0004] This invention provides a safety diagnostic method, device, equipment, and storage medium for new energy vehicles, so as to achieve accurate identification and graded early warning of battery risks in new energy vehicles.

[0005] According to one aspect of the present invention, a safety diagnostic method for new energy vehicles is provided, the method comprising:

[0006] The system collects battery status time-series data and external environment data of new energy vehicles at the edge and preprocesses the collected data.

[0007] Based on the preprocessed data, thermally related characteristic values ​​and electrically related characteristic values ​​are calculated, and preliminary safety diagnosis and early warning judgment are made based on the thermally related characteristic values ​​and electrically related characteristic values.

[0008] If the initial safety diagnosis result is that no warning has been triggered, the overall risk score corresponding to the new energy vehicle is calculated based on the pre-built overall risk scoring model and the coupled risk knowledge base distributed from the cloud.

[0009] The warning level corresponding to the new energy vehicle is determined based on the overall risk score and the preset risk threshold.

[0010] According to another aspect of the present invention, a safety diagnostic device for new energy vehicles is provided, the device comprising:

[0011] The data acquisition module is used to collect battery status time-series data and external environment data of new energy vehicles through the edge terminal, and to preprocess the collected data.

[0012] The preliminary safety diagnosis module is used to calculate thermally related feature values ​​and electrically related feature values ​​based on the preprocessed data, and to perform preliminary safety diagnosis and early warning judgment based on the thermally related feature values ​​and electrically related feature values.

[0013] The overall risk scoring module is used to calculate the overall risk score corresponding to the new energy vehicle based on a pre-built overall risk scoring model and a coupled risk knowledge base distributed from the cloud, when the preliminary safety diagnosis result is that no warning has been triggered.

[0014] The early warning level determination module is used to determine the early warning level corresponding to the new energy vehicle based on the overall risk score and the preset risk threshold.

[0015] According to another aspect of the present invention, an electronic device is provided, the electronic device comprising:

[0016] At least one processor;

[0017] and memory that is communicatively connected to at least one processor;

[0018] The memory stores a computer program that can be executed by at least one processor, and the computer program is executed by at least one processor so that at least one processor can execute the safety diagnosis method for new energy vehicles according to any embodiment of the present invention.

[0019] According to another aspect of the present invention, a computer-readable storage medium is provided, which stores computer instructions for causing a processor to execute and implement the safety diagnostic method for new energy vehicles according to any embodiment of the present invention.

[0020] The technical solution of this invention collects battery status time-series data and external environmental data of new energy vehicles at the edge, and preprocesses the collected data; calculates thermally related feature values ​​and electrically related feature values ​​based on the preprocessed data, and performs preliminary safety diagnosis and early warning judgment based on the thermally related feature values ​​and the electrically related feature values; if the preliminary safety diagnosis result is that no early warning is triggered, calculates the overall risk score corresponding to the new energy vehicle based on a pre-constructed overall risk scoring model and a coupled risk knowledge base distributed from the cloud; and determines the early warning level corresponding to the new energy vehicle based on the overall risk score and the preset risk threshold. This solves the technical problems of existing technologies that do not comprehensively consider the coupling effect of battery and environmental data, lack dynamic risk scoring and flexible early warning mechanisms, and achieves the technical effects of improving the accuracy of risk diagnosis, adapting to the differences in risks across multiple scenarios, and ensuring the timeliness of battery safety response.

[0021] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of the present invention, nor is it intended to limit the scope of the invention. Other features of the invention will become readily apparent from the following description. Attached Figure Description

[0022] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0023] Figure 1 A flowchart of a safety diagnostic method for new energy vehicles provided in an embodiment of the present invention;

[0024] Figure 2 A flowchart of another safety diagnosis method for new energy vehicles provided in an embodiment of the present invention;

[0025] Figure 3 This is a schematic diagram of the structure of a safety diagnostic device for new energy vehicles provided in an embodiment of the present invention;

[0026] Figure 4 A schematic diagram of the structure of an electronic device for implementing a safety diagnostic method for new energy vehicles according to an embodiment of the present invention. Detailed Implementation

[0027] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0028] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0029] Figure 1 This is a flowchart illustrating a safety diagnostic method for new energy vehicles provided in an embodiment of the present invention. This embodiment is applicable to the safety diagnostic situation of new energy vehicles. The method can be executed by a safety diagnostic device for new energy vehicles, which can be implemented in hardware and / or software and can be configured in an electronic device. Figure 1 As shown, the method specifically includes the following steps:

[0030] S110. Collect battery status time-series data and external environment data of new energy vehicles through the edge terminal, and preprocess the collected data.

[0031] In this context, edge computing can be understood as hardware units deployed locally on the new energy vehicle, possessing data acquisition and preliminary computing capabilities, such as onboard controllers and edge computing modules. Battery status time-series data can be understood as dynamic data continuously collected over time, reflecting the battery's operating status, such as the monitored values ​​of parameters like voltage, current, and temperature at different times. External environment data can be understood as environmental parameters of the external environment in which the new energy vehicle operates, such as ambient temperature, humidity, vibration intensity, and other external factors that may affect battery risks. Preprocessing can be understood as the operation of organizing and optimizing the collected raw data, with the aim of improving data quality. Common operations include removing invalid values ​​and completing missing data.

[0032] Specifically, the vehicle-mounted edge device collects real-time data on the battery's status over time during use, as well as data on the vehicle's environment such as temperature and humidity. This raw data is then preprocessed, for example, by deleting invalid data caused by sensor malfunctions and using linear interpolation to fill in data missing due to signal interruptions.

[0033] Optionally, the battery state time-series data includes at least two of the following: continuous monitoring data of battery voltage, current, temperature, and remaining charge changing over time; the external environment data includes at least two of the following: data of the environment in which the new energy vehicle is located; the preprocessing includes removing invalid values, completing missing data through linear interpolation, and verifying data accuracy using cross-validation.

[0034] Here, remaining battery power can be understood as the battery's current available power. Cross-validation can be understood as a method to verify the accuracy of the data.

[0035] Specifically, battery status time-series data should include at least two of the following: voltage, current, temperature, and remaining charge. External environment data should include at least two of the following: ambient temperature, humidity, vibration intensity, and altitude. Preprocessing specifically includes removing invalid values, such as abnormal high temperature values ​​caused by sensor failure; using linear interpolation to complete missing data, such as completing current data when the signal is interrupted; and using cross-validation to verify data accuracy, such as using the data from the first 10 minutes to verify the rationality of the data from the last 10 minutes.

[0036] Preferably, differentiated preprocessing strategies can be adopted for different types of data to adapt to data characteristics.

[0037] S120. Calculate thermally related characteristic values ​​and electrically related characteristic values ​​based on the preprocessed data, and perform preliminary safety diagnosis and early warning judgment based on the thermally related characteristic values ​​and electrically related characteristic values.

[0038] Thermally relevant characteristic values ​​can be understood as quantitative indicators extracted from data that are related to the battery's thermal state, such as the rate of temperature change and temperature gradient, used to reflect the battery's thermal risk. Electrically relevant characteristic values ​​can be understood as quantitative indicators extracted from data that are related to the battery's electrical performance, such as voltage fluctuation amplitude and current change rate, used to reflect the battery's electrical risk.

[0039] Specifically, based on the pre-processed valid data, thermally related characteristic values ​​(such as the battery temperature change rate in the past 5 minutes) and electrically related characteristic values ​​(such as the voltage fluctuation amplitude in the same time period) are calculated through feature engineering. These two types of characteristic values ​​are then compared with preset safety thresholds. If either characteristic value exceeds the limit, the highest level warning is triggered directly to remind the user or the vehicle system to intervene in a timely manner.

[0040] Preferably, the preset threshold can be set according to the differences in battery type to adapt to the safety characteristics of different batteries.

[0041] Optionally, the thermally related characteristic values ​​include the battery temperature change rate, temperature gradient, and heat accumulation per unit time; the electrically related characteristic values ​​include the battery voltage fluctuation amplitude, current change rate, and power decay rate per unit usage cycle.

[0042] The temperature gradient can be understood as the temperature difference between different parts of the battery. The heat accumulation per unit time can be understood as the amount of heat accumulated in the battery per unit time. The capacity decay rate per unit usage cycle can be understood as the percentage decrease in battery capacity within one usage cycle (such as a complete charge-discharge cycle).

[0043] Specifically, thermally related characteristics include battery temperature change rate (e.g., temperature rise per minute), temperature gradient (e.g., temperature difference between the positive and negative electrodes of the battery), and heat accumulation per unit time (e.g., heat accumulated in the battery per minute); electrically related characteristics include battery voltage fluctuation amplitude (e.g., maximum voltage fluctuation during charging), current change rate (e.g., how fast the current changes during discharge), and capacity decay rate per unit usage cycle (e.g., the percentage decrease in capacity from the initial value after one charge and discharge cycle).

[0044] Preferably, the frequency of feature value calculation can be adjusted according to the battery usage scenario (such as fast charging or slow charging).

[0045] Optionally, the preliminary safety diagnosis and early warning judgment based on the thermally related feature value and the electrically related feature value includes: comparing the thermally related feature value with a preset thermal feature safety threshold, and comparing the electrically related feature value with a preset electrical feature safety threshold; if the thermally related feature value exceeds the preset thermal feature safety threshold, or the electrically related feature value exceeds the preset electrical feature safety threshold, then the highest level early warning is triggered; if the thermally related feature value does not exceed the preset thermal feature safety threshold and the electrically related feature value does not exceed the preset electrical feature safety threshold, then it is determined that no early warning has been triggered.

[0046] Specifically, the calculated thermal-related characteristic values ​​are first compared with preset thermal characteristic safety thresholds, and the electrical-related characteristic values ​​are compared with preset electrical characteristic safety thresholds. If either the thermal-related or electrical characteristic value exceeds the limit, the highest level warning is immediately triggered (e.g., the vehicle system issues an audible and visual alarm). If neither exceeds the limit, it is determined that no warning has been triggered, and the process proceeds to the subsequent overall risk score calculation stage. The preset thermal and electrical characteristic safety thresholds can be pre-set and adjusted based on experience; this embodiment does not impose specific restrictions on them.

[0047] Preferably, when the highest level of warning is triggered, on-board emergency measures (such as cutting off the charging circuit) can be activated simultaneously to reduce safety risks.

[0048] S130. If the preliminary safety diagnosis result indicates that no warning has been triggered, calculate the overall risk score corresponding to the new energy vehicle based on the pre-built overall risk scoring model and the coupled risk knowledge base distributed from the cloud.

[0049] The overall risk scoring model can be understood as a pre-built mathematical model used to comprehensively calculate the overall risk of the battery, which can integrate thermal and electrical related indicators to output a quantitative risk value. The coupled risk knowledge base can be understood as a database distributed from the cloud and storing association rules for different risk scenarios, used to determine whether thermal risk and electrical risk have mutual influence. The overall risk score can be understood as a numerical value that quantifies the overall safety risk of the battery, calculated through the overall risk scoring model.

[0050] Specifically, if the initial warning is not triggered, the pre-built overall risk scoring model is invoked, and combined with the coupled risk knowledge base distributed in the cloud, the thermal risk index and the electrical risk index are calculated separately. Then, based on the knowledge base, it is determined whether there is a coupling relationship between the two (such as the electrical risk amplifying the thermal risk in a high humidity environment). Finally, the two types of indicators are merged to obtain a quantitative overall risk score.

[0051] Preferably, the model parameters can be dynamically adjusted according to the battery's service life. For example, the weight of the thermal risk index can be appropriately increased for older batteries to improve the accuracy of the scoring.

[0052] Optionally, the calculation of the overall risk score based on the pre-built overall risk scoring model and the coupled risk knowledge base distributed in the cloud includes:

[0053] The thermal risk index is obtained by summing the normalized thermal-related feature values ​​according to preset weights using the overall risk scoring model, and the electrical risk index is obtained by summing the normalized electrical-related feature values ​​according to preset weights.

[0054] Query the coupling risk knowledge base to determine whether there is a scenario-coupled relationship between the current thermal risk indicators and electrical risk indicators;

[0055] If there is no coupling relationship, the thermal risk index and the electrical risk index are integrated according to a preset ratio to obtain the overall risk score;

[0056] If a coupling relationship exists, the coupling coefficient of the corresponding scenario in the knowledge base is called, and the overall risk score is calculated based on the coupling coefficient and the preset fusion formula.

[0057] Normalization can be understood as a process of converting feature values ​​from different ranges into a unified value range. The thermal risk index can be understood as the value obtained by weighting and summing the normalized thermal-related feature values ​​according to preset weights. The electrical risk index can be understood as the value obtained by weighting and summing the normalized electrical-related feature values ​​according to preset weights. The coupling coefficient can be understood as a coefficient stored in the coupling risk knowledge base, used to quantify the degree of mutual influence between thermal and electrical risks; different coupling scenarios correspond to different coefficients.

[0058] Specifically, the thermal and electrical characteristic values ​​are first normalized. Then, using the overall risk scoring model, the thermal risk index is obtained by weighting and summing the values ​​according to preset weights (e.g., temperature change rate weight 0.4, temperature gradient weight 0.3, and heat accumulation per unit time weight 0.3). Similarly, the electrical risk index is obtained. Next, the coupled risk knowledge base is queried to determine whether there is a coupling relationship between the current thermal and electrical risks (e.g., in a high-temperature and high-humidity environment, thermal and electrical risks are coupled). If there is no coupling, the overall risk score is obtained by fusing the values ​​according to preset proportions (e.g., thermal risk index accounts for 40%, electrical risk index accounts for 60%). If there is coupling, the coupling coefficient for the corresponding scenario (e.g., the coupling coefficient for a high-temperature and high-humidity scenario is 1.2) is called and substituted into the preset fusion formula, for example, overall risk score = (thermal risk index × 0.4 + electrical risk index × 0.6) × coupling coefficient). The preset weights and proportions can be preset and adjusted based on experience; this embodiment does not impose specific restrictions on them.

[0059] Preferably, the preset weights can be dynamically adjusted based on historical fault data. For example, if historical data shows that the temperature gradient has a greater impact on the fault, then its weight will be increased.

[0060] S140. Determine the warning level corresponding to the new energy vehicle based on the overall risk score and the preset risk threshold.

[0061] The preset risk threshold can be understood as a pre-set numerical standard used to judge the risk level. Different thresholds correspond to different warning levels and can be preset and adjusted based on experience. This embodiment does not impose specific restrictions on it.

[0062] Specifically, the calculated overall risk score is compared with a preset risk threshold. For example, an overall risk score of 0-30 corresponds to low risk, an overall risk score of 31-60 corresponds to medium risk, and an overall risk score of 61-100 corresponds to high risk. Based on the overall risk score and the preset risk threshold, the warning level corresponding to the battery of the current new energy vehicle is determined, providing a basis for subsequent safety measures.

[0063] Preferably, the warning level can be displayed intuitively on the vehicle's in-vehicle display screen, such as green for low risk and red for high risk, so that users can quickly know the risk status.

[0064] Optionally, the method further includes:

[0065] The battery status time-series data collected at the edge, along with external environment data, preprocessed data, calculated thermal-related characteristic values, electrical-related characteristic values, and overall risk score, are packaged and transmitted to the cloud through encryption combined with dynamic key management.

[0066] Encryption can be understood as the operation of encoding data, including AES-256 symmetric encryption and asymmetric encryption. Dynamic key management can be understood as a mechanism for dynamically updating and managing the keys used for encryption.

[0067] Specifically, the edge device packages the collected raw data (battery status time-series data, external environment data), preprocessed data, calculated thermal-related characteristic values, electrical-related characteristic values, and overall risk score, and transmits the data to the cloud using encryption (such as AES-256 symmetric encryption) combined with dynamic key management to ensure data security during transmission and prevent the leakage of sensitive information.

[0068] Preferably, the principle of minimum necessity can be used to filter transmitted data, such as transmitting only abnormal data and key feature values, thereby reducing bandwidth consumption.

[0069] Optionally, the method further includes:

[0070] The cloud-based system iteratively optimizes the parameters of the overall risk scoring model based on data uploaded from at least one edge device of a new energy vehicle, and updates the coupling rules and coefficients of different scenarios in the coupled risk knowledge base.

[0071] Specifically, after receiving data uploaded from the edge of at least one new energy vehicle, the cloud iteratively optimizes the parameters of the overall risk scoring model based on this massive amount of data (such as adjusting the weights of thermal and electrical feature values), and updates the coupling rules and coefficients of different scenarios in the coupled risk knowledge base (such as adding coupling rules for low temperature + high current scenarios), so that the model and knowledge base are more adapted to actual use scenarios.

[0072] Preferably, the cloud can perform a comprehensive optimization of the model and knowledge base at preset time intervals, such as once a month, to ensure continuous improvement in diagnostic capabilities.

[0073] Furthermore, the cloud will encrypt and transmit the optimized overall risk scoring model parameters, the updated coupled risk knowledge base, and the adjusted preset risk thresholds back to the edge. After receiving them, the edge will replace the old local parameters and knowledge base, complete the upgrade of security diagnostic capabilities, and form a closed loop of security diagnosis of "data collection - diagnosis and analysis - cloud optimization - capability update".

[0074] Preferably, the cloud-based system trains both a thermal risk logistic regression model and an electrical risk logistic regression model using data uploaded from at least one edge device of a new energy vehicle. The thermal risk logistic regression model is trained using thermally relevant feature values ​​as input and actual thermal fault results as output, while the electrical risk logistic regression model is trained using electrically relevant feature values ​​as input and actual electrical fault results as output. The model parameters are periodically iteratively optimized based on newly added data to improve the model's prediction accuracy.

[0075] The technical solution of this invention collects battery status time-series data and external environmental data of new energy vehicles at the edge, and preprocesses the collected data; calculates thermally related feature values ​​and electrically related feature values ​​based on the preprocessed data, and performs preliminary safety diagnosis and early warning judgment based on the thermally related feature values ​​and the electrically related feature values; if the preliminary safety diagnosis result is that no early warning is triggered, calculates the overall risk score corresponding to the new energy vehicle based on a pre-constructed overall risk scoring model and a coupled risk knowledge base distributed from the cloud; and determines the early warning level corresponding to the new energy vehicle based on the overall risk score and the preset risk threshold. This solves the technical problems of existing technologies that do not comprehensively consider the coupling effect of battery and environmental data, lack dynamic risk scoring and flexible early warning mechanisms, and achieves the technical effects of improving the accuracy of risk diagnosis, adapting to the differences in risks across multiple scenarios, and ensuring the timeliness of battery safety response.

[0076] Figure 2 This is a flowchart of another safety diagnostic method for new energy vehicles provided by an embodiment of the present invention. Based on the above embodiments, this embodiment is an optimization of the above embodiments, and its specific implementation can be found in the technical solution of this embodiment. Technical terms that are the same as or corresponding to those in the above embodiments will not be repeated here. Figure 2 As shown, the method specifically includes the following steps:

[0077] S210: Collect battery status data and external temperature data, bind timestamp information and perform preprocessing to construct a basic status feature set.

[0078] Specifically, by setting up multiple sensors to collect battery status data and external temperature data at preset frequencies, binding high-precision timestamps and performing preprocessing, a basic state feature set is constructed. The battery status data includes the voltage of each cell in the battery pack, the center temperature of each cell module, current, SOC (State of Charge / Remaining Capacity), SOH (State of Health), charging power, discharging power, and insulation resistance. Preprocessing includes: outlier handling, calculation of thermally related features, calculation of electrically related features, and standardization.

[0079] Furthermore, the calculation of thermally related features includes:

[0080] Based on battery status data, acquire temperature data at each temperature measurement point;

[0081] Based on the temperature data, the temperature gradient at each measurement point is calculated, and the fastest heating gradient and the highest temperature are obtained using the maximum value function; the third central moment of the temperature data is calculated to obtain the temperature distribution skewness.

[0082] ;

[0083] Where n is the number of battery cell module centers. It is the center temperature of the i-th cell module. It is the average temperature. It represents the standard deviation of all temperature sample data. Temperature distribution skewness is used to quantify the asymmetry of temperature distribution. If right skewness occurs, it indicates that the temperature at a few temperature measurement points is significantly higher than that at most temperature measurement points, suggesting a local overheating trend.

[0084] The maximum rate of change of temperature difference between adjacent cell module pairs in the computational space is used as the fastest heating rate of the adjacent point, thereby generating thermally correlated features.

[0085] The calculation of electrical correlation characteristics includes:

[0086] Based on battery status data, obtain voltage and insulation resistance data for each cell module;

[0087] Based on voltage data, the voltage difference between all cell modules is calculated, and the maximum voltage difference is obtained using the maximum value function; the reciprocal of the insulation resistance data is calculated as the reverse insulation resistance feature; and electrical-related features are generated by combining the battery's basic parameters, including current, charging power, and discharging power.

[0088] S220. Calculate the feature weight coefficients, and perform weighted operations based on the basic state feature set and feature weight coefficients to obtain the basic thermal risk index and the basic electrical risk index.

[0089] Specifically, the feature weight coefficients stored locally are called to perform weighted summation on thermally related features and electrically related features respectively to obtain the basic thermal risk index and the basic electrical risk index, and the feature weight coefficients are iteratively optimized.

[0090] Furthermore, the iterative optimization of feature weight coefficients includes:

[0091] Based on historical data, we constructed a logistic regression model for thermal risk and a logistic regression model for electrical risk.

[0092] Using thermally related feature values ​​as input and thermal fault binary labels indicating whether a thermal fault has occurred as output, a thermal risk logistic regression model is trained, and the regression coefficients of all thermally related features are obtained as the corresponding feature weight coefficients. Using electrically related feature values ​​as input and electrical fault binary labels indicating whether an electrical fault has occurred as output, an electrical risk logistic regression model is trained, and the regression coefficients of all electrically related features are obtained as the corresponding feature weight coefficients.

[0093] Each regression coefficient represents the relative importance of each feature in risk assessment and is directly used as the parameter weights for calculating the basic thermal risk and electrical risk indicators. These coefficients are then distributed to the edge via OTA (Over-the-Air) or periodic communication. The cloud can periodically retrain the model and update the feature weight coefficients based on changes in global data, enabling continuous model evolution.

[0094] S230. Based on the battery status, calculate the thermal risk status index and the electrical risk status index.

[0095] A battery’s risk performance is closely related to its current state of charge and health status. Based on the thermal risk index and electrical risk index calculated from general characteristics, combined with the battery’s state of charge (SOC) and state of health (SOH) as influencing factors for safety status diagnosis, a more accurate risk score can be obtained.

[0096] Furthermore, the calculation of thermal risk state indices and electrical risk state indices includes:

[0097] The mean of all fault samples is calculated as the fault sample mean, which represents the baseline level of each feature when a fault occurs.

[0098] Divide the intervals and calculate the mean of the partition features, including: dividing all fault samples into low SOC intervals, medium SOC intervals and high SOC intervals according to the SOC value, and calculating the mean of the features for the low SOC interval, medium SOC interval and high SOC interval respectively.

[0099] Based on the SOH value, all fault samples are divided into low SOH intervals and high SOH intervals, and the characteristic mean values ​​of the low SOH interval and high SOH interval are calculated respectively.

[0100] The SOC (State of Charge) and SOH (State of Alarm) baseline coefficients are calculated to quantify the impact of other states on risk by setting benchmark states. The medium SOC and high SOH ranges are considered safe states, as they are not believed to significantly amplify risk, and are thus set as benchmark states. Specifically, the SOC baseline coefficient is the ratio of the characteristic mean of the fault samples to the characteristic mean of the medium SOC range, and the SOH baseline coefficient is the ratio of the characteristic mean of the fault samples to the characteristic mean of the high SOH range.

[0101] Calculate the SOC adjustment coefficient and SOH adjustment coefficient for each interval to quantify the risk amplification or reduction factor of the current state relative to the baseline state. The SOC adjustment coefficient includes low, medium, and high SOC adjustment coefficients, which are the products of the ratio of the characteristic mean of the interval to the characteristic mean of the fault samples and the SOC baseline coefficient, respectively. The SOH adjustment coefficient includes low and high SOH adjustment coefficients, which are the products of the ratio of the characteristic mean of the interval to the characteristic mean of the fault samples and the SOH baseline coefficient, respectively.

[0102] Based on the real-time SOC and SOH values ​​of the current battery, determine its corresponding SOC and SOH ranges, and then determine the corresponding SOC adjustment coefficient and SOH adjustment coefficient.

[0103] The thermal risk status index is obtained by multiplying the basic thermal risk index with the corresponding SOC adjustment coefficient and SOH adjustment coefficient. The electrical risk status index is obtained by multiplying the basic electrical risk index with the corresponding SOC adjustment coefficient and SOH adjustment coefficient.

[0104] S240. Obtain the thermal coupling amplification factor and the electrical coupling amplification factor based on the coupling risk knowledge base, and calculate the comprehensive thermal risk index and the comprehensive electrical risk index.

[0105] Specifically, a coupling risk knowledge base is constructed to map the thermal coupling amplification factor and electrical coupling amplification factor corresponding to each coupling scenario, and to calculate the comprehensive thermal risk index and the comprehensive electrical risk index. The comprehensive thermal risk index is the product of the thermal coupling amplification factor and the thermal risk state index, and the comprehensive electrical risk index is the product of the electrical coupling amplification factor and the electrical risk state index.

[0106] Battery risks are often not caused by a single factor, but are the result of multiple state and behavioral factors being coupled and amplified together. Therefore, it is necessary to build a coupled risk knowledge base and store it at the edge, and to judge in real time whether coupled scenarios exist and adjust the calculation of risk scores.

[0107] Furthermore, constructing a coupled risk knowledge base includes:

[0108] Construct a set of state features and a set of behavior features, wherein the state features include at least external temperature, SOC, and SOH, and the behavior features include at least charging power and discharging power;

[0109] Select one or more features from the state feature set and the behavior feature set respectively to form a coupled feature group;

[0110] Construct a risk prediction model and calculate the thermal coupling amplification factor and the electrical coupling amplification factor;

[0111] Each coupling feature group is associated with its corresponding coupling amplification coefficient and stored in the coupling risk knowledge base.

[0112] Furthermore, constructing coupled feature sets includes:

[0113] Discretized category features include: assigning degree labels to each feature in the state feature set and behavior feature set by setting targeted thresholds, generating state-degree feature sets and behavior-degree feature sets. Specifically, external temperature, SOC, charging power, and discharging power are all divided into three categories: low, medium, and high; SOH is divided into two categories: low and high.

[0114] N features are selected from the state-degree feature set, and M features are selected from the behavior-degree feature set. These are then fused into a single coupled feature group. Different coupled feature groups represent different risk coupling scenarios. Here, N is any positive integer not greater than the number of features in the state-degree feature set, and M is any positive integer not greater than the number of features in the behavior-degree feature set.

[0115] Furthermore, the calculation of thermal coupling amplification factor and electrical coupling amplification factor includes:

[0116] A coupled feature matrix is ​​constructed based on coupled feature groups. The coupled feature matrix includes: basic features and interaction features. The basic features include the original feature values ​​of state features and behavior features. The interaction features correspond one-to-one with the coupled feature groups and are identified by binary labels.

[0117] Constructing and training risk prediction models includes: constructing a thermal risk prediction model, inputting a coupling feature matrix, outputting the corresponding thermal fault binary labels, and training it fully until the model converges; constructing an electrical risk prediction model, inputting a coupling feature matrix, outputting the corresponding electrical fault binary labels, and training it fully until the model converges.

[0118] After the model is trained, the contribution of coupled feature groups to risk is quantified using model interpretation techniques, including:

[0119] Input the coupling feature matrix and the thermal risk prediction model into the SHAP analysis tool to calculate the average SHAP value of each coupling feature group across all samples. The average SHAP value is the coupling risk importance score for the corresponding coupling feature group; a higher score indicates a greater contribution of that coupling feature group to the model's prediction of high risk. If the average SHAP value is not greater than 0, the thermal coupling amplification factor is set to 1 by default; otherwise, the thermal coupling amplification factor is set to the normalized result of the average SHAP value.

[0120] Input the coupling feature matrix and electrical risk prediction model into the SHAP analysis tool to calculate the average SHAP value of each coupling feature group on all samples. If the average SHAP value is not greater than 0, the electrical coupling amplification factor is set to 1 by default; otherwise, the electrical coupling amplification factor is set to the normalized result of the average SHAP value.

[0121] S250. Calculate the overall risk score, conduct root cause analysis, and implement targeted protection.

[0122] The thermal and electrical risks of new energy vehicle battery systems are not independent. Assessing the impact of thermal or electrical risk comprehensive indicators on safety status separately will seriously underestimate the actual hazards. Therefore, the combined impact of thermal and electrical risks should be taken into account when calculating the overall risk score.

[0123] Furthermore, the calculation of the overall risk score includes:

[0124] A comprehensive risk characteristic matrix is ​​constructed, including: a comprehensive thermal risk index, a comprehensive electrical risk index, and a collaborative risk index. The collaborative risk index, which is the product of the comprehensive thermal risk index and the comprehensive electrical risk index, is used to quantify the interaction.

[0125] A comprehensive risk model is constructed based on a machine learning model. The input is a pre-constructed comprehensive risk feature matrix, and the output is a comprehensive risk level label. The model is then trained thoroughly until it converges. The number of comprehensive risk level labels corresponds to the number of warning levels.

[0126] Based on the trained overall risk model, the weight coefficients corresponding to the comprehensive thermal risk index, comprehensive electrical risk index, and collaborative risk index are determined through model interpretation techniques and normalization.

[0127] Based on the calculated comprehensive thermal risk index, comprehensive electrical risk index, and collaborative risk index, and their corresponding weighting coefficients, the overall risk score is calculated by weighted summation of the comprehensive thermal risk index, comprehensive electrical risk index, and collaborative risk index.

[0128] The calculated overall risk score is compared with a preset threshold. If the minimum warning threshold is not reached, no warning is issued. If the threshold is reached, the corresponding level of warning is triggered.

[0129] Once an alert is triggered, the vehicle needs to quickly identify the primary risk in order to take immediate initial protective measures. Simultaneously, detailed data needs to be uploaded to the cloud for in-depth root cause analysis to develop effective protection strategies, forming a complete closed loop of diagnosis, analysis, and optimization.

[0130] Further, root cause analysis includes:

[0131] The dominant risk type is determined based on the comprehensive thermal risk index, comprehensive electrical risk index, and collaborative risk index, generating a preliminary diagnostic result. If the ratio of the comprehensive thermal risk index to the comprehensive electrical risk index is greater than a preset threshold, it is preliminarily judged as thermal risk dominant; if the ratio is less than the preset threshold, it is preliminarily judged as electrical risk dominant; otherwise, it is preliminarily judged as collaborative risk dominant.

[0132] By integrating the basic state feature set, overall risk score, and preliminary diagnostic results, a root cause analysis feature set is generated. This feature set is then input into a pre-trained XGBoost multi-classification model, which outputs the specific root cause categories and corresponding confidence levels of new energy vehicle battery safety risks.

[0133] A multi-classification model is constructed based on machine learning. It takes a root cause analysis feature set as input and outputs a probability distribution of preset root cause labels, selecting the label with the highest probability as the candidate root cause. The contribution of each feature to the candidate root cause is quantified using the SHAP value; a higher SHAP value indicates a stronger contribution, thus further identifying the core triggers and providing direction for the selection of protection strategies. To ensure the accuracy of the root cause analysis, cosine similarity of features is calculated to filter cases in historical data with similarity exceeding a preset similarity threshold. The final triggers of these highly similar cases are compared to cross-validate the current analysis results, clarifying the accuracy of the root cause analysis.

[0134] The root cause analysis results are organized into a structured format to ensure that the edge can directly parse and associate them with protection strategies. Based on the dominant risk type, core trigger, and warning level in the root cause analysis results, targeted protection is immediately implemented according to the protection strategy. At the same time, the implementation process of the protection strategy is monitored in real time and dynamically adjusted. After each monitoring cycle, the protection effect is compared with that of similar historical cases. If the rate of decline of a certain indicator is lower than the preset proportion of its corresponding historical average, a strategy optimization instruction is issued to further adjust the protection actions in a targeted manner.

[0135] The technical solution of this invention combines real-time edge detection with deep cloud analysis to quantify the coupling effect of battery state characteristics and behavioral characteristics, accurately assess vehicle safety status, locate dominant risks and core causes, and effectively improve the real-time nature of safety diagnosis, the accuracy of risk identification, and the pertinence of protective measures.

[0136] Figure 3 This is a schematic diagram of the structure of a safety diagnostic device for new energy vehicles provided in an embodiment of the present invention. Figure 3 As shown, the device includes: a data acquisition module 310, a preliminary safety diagnosis module 320, an overall risk scoring module 330, and a warning level determination module 340.

[0137] The system includes a data acquisition module 310, which collects battery status time-series data and external environment data of the new energy vehicle through an edge device and preprocesses the collected data; a preliminary safety diagnosis module 320, which calculates thermally related feature values ​​and electrically related feature values ​​based on the preprocessed data, and performs preliminary safety diagnosis and early warning judgment based on the thermally related feature values ​​and electrically related feature values; an overall risk scoring module 330, which calculates the overall risk score corresponding to the new energy vehicle based on a pre-built overall risk scoring model and a coupled risk knowledge base distributed from the cloud when the preliminary safety diagnosis result is that no early warning has been triggered; and an early warning level determination module 340, which determines the early warning level corresponding to the new energy vehicle based on the overall risk score and the preset risk threshold.

[0138] The technical solution of this invention collects battery status time-series data and external environmental data of new energy vehicles at the edge, and preprocesses the collected data; calculates thermally related feature values ​​and electrically related feature values ​​based on the preprocessed data, and performs preliminary safety diagnosis and early warning judgment based on the thermally related feature values ​​and the electrically related feature values; if no early warning is triggered, calculates the overall risk score corresponding to the new energy vehicle based on a pre-constructed overall risk scoring model and a coupled risk knowledge base distributed from the cloud; and determines the early warning level corresponding to the new energy vehicle based on the overall risk score and the preset risk threshold. This solves the technical problems of existing technologies that do not comprehensively consider the coupled influence of battery and environmental data, lack dynamic risk scoring and flexible early warning mechanisms, and achieves technical effects such as improving the accuracy of risk diagnosis, adapting to risk differences in multiple scenarios, and ensuring timely battery safety response.

[0139] In some optional embodiments, the battery state time-series data includes at least two of the following: continuous monitoring data of battery voltage, current, temperature, and remaining charge changing over time; the external environment data includes at least two of the following: temperature, humidity, vibration intensity, and altitude data of the environment in which the new energy vehicle is located; the preprocessing includes removing invalid values, completing missing data through linear interpolation, and verifying data accuracy using cross-validation.

[0140] In some alternative embodiments, the preliminary security diagnostic module includes:

[0141] The feature value comparison unit is used to compare the thermally related feature value with a preset thermal feature safety threshold and the electrically related feature value with a preset electrical feature safety threshold.

[0142] The first early warning judgment unit is used to trigger the highest level early warning if the thermally related characteristic value exceeds a preset thermal characteristic safety threshold or the electrically related characteristic value exceeds a preset electrical characteristic safety threshold.

[0143] The second early warning judgment unit is used to determine that no early warning has been triggered if the thermally related characteristic value does not exceed the preset thermal characteristic safety threshold and the electrically related characteristic value does not exceed the preset electrical characteristic safety threshold.

[0144] In some optional embodiments, the thermally related characteristic values ​​include the battery temperature change rate, temperature gradient, and heat accumulation per unit time; the electrically related characteristic values ​​include the battery voltage fluctuation amplitude, current change rate, and power decay rate per unit usage cycle.

[0145] In some optional embodiments, the overall risk scoring module includes:

[0146] The risk index calculation unit is used to obtain the thermal risk index by weighting and summing the normalized thermal-related characteristic values ​​according to preset weights through the overall risk scoring model, and to obtain the electrical risk index by weighting and summing the normalized electrical-related characteristic values ​​according to preset weights.

[0147] The knowledge base query unit is used to query the coupled risk knowledge base to determine whether there is a scenario-coupled relationship between the current thermal risk index and the electrical risk index.

[0148] The first risk scoring unit is used to merge thermal risk indicators and electrical risk indicators according to a preset ratio to obtain an overall risk score if there is no coupling relationship.

[0149] The second risk scoring determination unit is used to call the coupling coefficient of the corresponding scenario in the knowledge base if there is a coupling relationship, and calculate the overall risk score based on the coupling coefficient and the preset fusion formula.

[0150] In some alternative embodiments, the apparatus further includes:

[0151] The data transmission module is used to package the battery status time-series data collected at the edge, external environment data, preprocessed data, calculated thermal-related characteristic values, electrical-related characteristic values, and overall risk score, and transmit them to the cloud through encryption combined with dynamic key management.

[0152] In some alternative embodiments, the apparatus further includes:

[0153] The coefficient update module is used to iteratively optimize the parameters of the overall risk scoring model in the cloud based on data uploaded from at least one new energy vehicle edge terminal, and update the coupling rules and coefficients of different scenarios in the coupled risk knowledge base.

[0154] The safety diagnostic device for new energy vehicles provided in this embodiment of the invention can execute the safety diagnostic method for new energy vehicles provided in any embodiment of the invention, and has the corresponding functional modules and beneficial effects of the method.

[0155] Figure 4 This is a schematic diagram of the structure of an electronic device for implementing the safety diagnostic method for new energy vehicles according to embodiments of the present invention. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital assistants, cellular phones, smartphones, wearable devices (e.g., helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the invention described and / or claimed herein.

[0156] like Figure 4 As shown, the electronic device 10 includes at least one processor 11 and a memory, such as a read-only memory (ROM) 12 or a random access memory (RAM) 13, communicatively connected to the at least one processor 11. The memory stores computer programs executable by the at least one processor. The processor 11 can perform various appropriate actions and processes based on the computer program stored in the ROM 12 or loaded into the RAM 13 from storage unit 18. The RAM 13 can also store various programs and data required for the operation of the electronic device 10. The processor 11, ROM 12, and RAM 13 are interconnected via a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0157] Multiple components in electronic device 10 are connected to I / O interface 15, including: input unit 16, such as keyboard, mouse, etc.; output unit 17, such as various types of displays, speakers, etc.; storage unit 18, such as disk, optical disk, etc.; and communication unit 19, such as network card, modem, wireless transceiver, etc. Communication unit 19 allows electronic device 10 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0158] Processor 11 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. Processor 11 performs the various methods and processes described above, such as the method of safety diagnosis of new energy vehicles.

[0159] In some embodiments, the safety diagnostics of the new energy vehicle method can be implemented as a computer program tangibly contained in a computer-readable storage medium, such as storage unit 18. In some embodiments, part or all of the computer program can be loaded and / or installed on electronic device 10 via ROM 12 and / or communication unit 19. When the computer program is loaded into RAM 13 and executed by processor 11, one or more steps of the safety diagnostics of the new energy vehicle method described above can be performed. Alternatively, in other embodiments, processor 11 can be configured to perform the safety diagnostics of the new energy vehicle method by any other suitable means (e.g., by means of firmware).

[0160] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations may include: implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0161] Computer programs used to implement the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when executed by the processor, the computer programs cause the functions / operations specified in the flowcharts and / or block diagrams to be performed. The computer programs may be executed entirely on a machine, partially on a machine, or as a standalone software package, partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0162] In the context of this invention, a computer-readable storage medium can be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, apparatus, or device. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination thereof. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0163] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the electronic device. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0164] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or middleware components (e.g., application servers), or frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include local area networks (LANs), wide area networks (WANs), blockchain networks, and the Internet.

[0165] A computing system can include clients and servers. Clients and servers are generally located far apart and typically interact through communication networks. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or cloud host, which is a hosting product within the cloud computing service system to address the shortcomings of traditional physical hosts and VPS services, such as high management difficulty and weak business scalability.

[0166] It should be understood that the various forms of processes shown above can be used, with steps reordered, added, or deleted. For example, the steps described in this invention can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of this invention can be achieved, and this is not limited herein.

[0167] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. A safety diagnostic method for new energy vehicles, characterized in that, include: The system collects battery status time-series data and external environment data of new energy vehicles at the edge and preprocesses the collected data. Based on the preprocessed data, thermally related characteristic values ​​and electrically related characteristic values ​​are calculated, and preliminary safety diagnosis and early warning judgment are made based on the thermally related characteristic values ​​and electrically related characteristic values. If the initial safety diagnosis result is that no warning has been triggered, the overall risk score corresponding to the new energy vehicle is calculated based on the pre-built overall risk scoring model and the coupled risk knowledge base distributed from the cloud. The warning level corresponding to the new energy vehicle is determined based on the overall risk score and the preset risk threshold.

2. The method according to claim 1, characterized in that, The battery status time-series data includes at least two of the following: continuous monitoring data of battery voltage, current, temperature, and remaining charge changing over time; the external environment data includes at least two of the following: temperature, humidity, vibration intensity, and altitude data of the environment in which the new energy vehicle is located; the preprocessing includes removing invalid values, completing missing data through linear interpolation, and verifying data accuracy using cross-validation.

3. The method according to claim 1, characterized in that, The preliminary safety diagnosis and early warning judgment based on the thermally related feature values ​​and the electrically related feature values ​​includes: The thermally related feature values ​​are compared with preset thermal feature safety thresholds, and the electrically related feature values ​​are compared with preset electrical feature safety thresholds. If the thermally related characteristic value exceeds a preset thermal characteristic safety threshold, or the electrically related characteristic value exceeds a preset electrical characteristic safety threshold, then the highest level warning is triggered. If the thermally related characteristic value does not exceed the preset thermal characteristic safety threshold and the electrically related characteristic value does not exceed the preset electrical characteristic safety threshold, then it is determined that no warning has been triggered.

4. The method according to claim 1, characterized in that, The thermally related characteristic values ​​include the battery temperature change rate, temperature gradient, and heat accumulation per unit time; the electrically related characteristic values ​​include the battery voltage fluctuation amplitude, current change rate, and power decay rate per unit usage cycle.

5. The method according to claim 1, characterized in that, The calculation of the overall risk score based on the pre-built overall risk scoring model and the coupled risk knowledge base distributed in the cloud includes: The thermal risk index is obtained by summing the normalized thermal-related feature values ​​according to preset weights using the overall risk scoring model, and the electrical risk index is obtained by summing the normalized electrical-related feature values ​​according to preset weights. Query the coupling risk knowledge base to determine whether there is a scenario-coupled relationship between the current thermal risk indicators and electrical risk indicators; If there is no coupling relationship, the thermal risk index and the electrical risk index are integrated according to a preset ratio to obtain the overall risk score; If a coupling relationship exists, the coupling coefficient of the corresponding scenario in the knowledge base is called, and the overall risk score is calculated based on the coupling coefficient and the preset fusion formula.

6. The method according to claim 1, characterized in that, Also includes: The battery status time-series data collected at the edge, along with external environment data, preprocessed data, calculated thermal-related characteristic values, electrical-related characteristic values, and overall risk score, are packaged and transmitted to the cloud through encryption combined with dynamic key management.

7. The method according to claim 6, characterized in that, Also includes: The cloud-based system iteratively optimizes the parameters of the overall risk scoring model based on data uploaded from at least one edge device of a new energy vehicle, and updates the coupling rules and coefficients of different scenarios in the coupled risk knowledge base.

8. A safety diagnostic device for new energy vehicles, characterized in that, include: The data acquisition module is used to collect battery status time-series data and external environment data of new energy vehicles through the edge terminal, and to preprocess the collected data. The preliminary safety diagnosis module is used to calculate thermally related feature values ​​and electrically related feature values ​​based on the preprocessed data, and to perform preliminary safety diagnosis and early warning judgment based on the thermally related feature values ​​and electrically related feature values. The overall risk scoring module is used to calculate the overall risk score corresponding to the new energy vehicle based on a pre-built overall risk scoring model and a coupled risk knowledge base distributed from the cloud, when the preliminary safety diagnosis result is that no warning has been triggered. The early warning level determination module is used to determine the early warning level corresponding to the new energy vehicle based on the overall risk score and the preset risk threshold.

9. An electronic device, characterized in that, The electronic device includes: At least one processor; and a memory communicatively connected to the at least one processor; The memory stores a computer program that can be executed by the at least one processor, and the computer program is executed by the at least one processor to enable the at least one processor to perform the safety diagnostic method for new energy vehicles according to any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions that are used to cause a processor to execute the safety diagnostic method for new energy vehicles as described in any one of claims 1-7.

Images