Electric vehicle classification method, system and equipment based on multi-dimensional safety factors and storage medium

Abstract

The invention discloses an electric vehicle classification method, system and device based on a multi-dimensional safety factor and a storage medium, and is applied to the technical field of electric vehicle charging, and the method comprises the steps: determining and obtaining the multi-dimensional safety factor affecting the charging safety of an electric vehicle; obtaining a preliminary weight of the multi-dimensional safety factor by adopting an analytic hierarchy process, and correcting the preliminary weight by adopting an entropy weight method to obtain a final weight of the multi-dimensional safety factor; and based on the final weight of the multi-dimensional safety factor, carrying out electric vehicle classification by adopting a fuzzy C-means clustering algorithm. Various factors influencing the charging safety of the electric vehicle are deeply analyzed, vehicle classification is carried out based on the factors, the safety charging requirement of each vehicle is accurately matched, the charging safety of the electric vehicle is fundamentally improved, the accident rate is reduced, meanwhile, the service life of a battery is prolonged, and the battery replacement cost is reduced.

Description

Technical Field

[0001] This invention relates to the field of electric vehicle charging technology, and more specifically to a method, system, device, and storage medium for classifying electric vehicles based on multi-dimensional safety factors. Background Technology

[0002] Against the backdrop of global advocacy for green travel and sustainable development, electric vehicles are experiencing unprecedented growth in the automotive market due to their advantages such as zero emissions and low energy consumption. However, the explosive growth in the number of electric vehicles has also led to frequent charging safety incidents, becoming a bottleneck restricting the healthy development of the industry. These incidents not only cause direct losses such as vehicle damage and personal injury, but also trigger a crisis of public trust in the safety of electric vehicles, hindering potential consumers' purchasing decisions.

[0003] Traditional electric vehicle charging strategies are largely based on empirical vehicle classifications, such as setting uniform charging parameters according to vehicle type (sedans, SUVs, MPVs, etc.) and battery type (ternary lithium batteries, lithium iron phosphate batteries, etc.). This "one-size-fits-all" approach severely ignores the dynamic differences in key safety factors such as insulation performance, battery degradation, and environmental adaptability among individual vehicles. During actual use, each electric vehicle experiences continuous changes in battery health, charging equipment reliability, and sensitivity to environmental factors due to variations in mileage, charge / discharge cycles, and operating environment (high temperature, high humidity, low temperature, etc.). Simple classifications cannot accurately match the safe charging needs of each vehicle.

[0004] Therefore, how to provide a multi-dimensional safety factor-based electric vehicle classification method that can deeply analyze various factors affecting the safety of electric vehicle charging, classify vehicles based on this, accurately match the safety charging needs of each vehicle, fundamentally improve the safety of electric vehicle charging, reduce the accident rate, extend battery life, and reduce battery replacement costs is a problem that urgently needs to be solved by those skilled in the art. Summary of the Invention

[0005] In view of this, the present invention provides a method, system, device and storage medium for classifying electric vehicles based on multidimensional safety factors.

[0006] To achieve the above objectives, the present invention adopts the following technical solution: A method for classifying electric vehicles based on multidimensional safety factors, comprising: Step 1: Identify and obtain the multi-dimensional safety factors that affect the charging safety of electric vehicles; Step 2: Use the analytic hierarchy process (AHP) to obtain the initial weights of the multidimensional security factors, and then use the entropy weight method to correct the initial weights to obtain the final weights of the multidimensional security factors. Step 3: Based on the final weights of the multidimensional safety factors, the fuzzy C-means clustering algorithm is used to classify electric vehicles.

[0007] Optionally, in step 1, the multidimensional safety factor is specifically divided into three dimensions: battery performance, device safety, and environmental adaptability. Safety factors in battery performance include: battery health status and charging reception capability; battery health status is quantified by capacity decay rate and internal resistance growth rate; charging reception capability is quantified by maximum safe charging current. Safety factors in the equipment safety dimension include: the insulation level of the charging interface and the response speed of the protection device; among which, the insulation level of the charging interface is quantified by the insulation resistance; the response speed of the protection device is quantified by the overcurrent protection action time. Safety factors in the environmental adaptability dimension include: temperature sensitivity coefficient and altitude correction coefficient; among which, the temperature sensitivity coefficient is quantified by the charging efficiency temperature decay rate; the altitude correction coefficient is quantified by the influence rate of air pressure on discharge capacity.

[0008] Optionally, in step 1, after obtaining the multi-dimensional safety factors affecting the charging safety of electric vehicles, the method further includes: preprocessing the multi-dimensional safety factor data. Preprocessing, specifically: The data is normalized using the 0-1 normalization method, mapping all data to... The interval, and the use of the 3σ principle to remove outliers from the data.

[0009] Optionally, in step 2, the preliminary weights of the multidimensional security factors are obtained using the analytic hierarchy process (AHP), specifically as follows: Electric vehicle charging safety is taken as the target layer, the various dimensions affecting electric vehicle charging safety are taken as the criterion layer, and the safety factors contained in each dimension are taken as the solution layer. The relative importance of the indicators at the criterion layer and the alternative layer is scored using the 1-9 scale method to obtain the judgment matrices for the criterion layer and the alternative layer. The eigenvalue method is used to solve each judgment matrix to obtain the largest eigenvalue and the corresponding eigenvector, and the eigenvector is normalized to obtain the weight vector of each security factor. A consistency check is performed. When the random consistency ratio is less than the preset value, the initial weight is determined. When the random consistency ratio is greater than or equal to the preset value, the scoring is repeated until the random consistency ratio is less than the preset value.

[0010] Optionally, in step 2, the initial weights are corrected using the entropy weight method, specifically as follows: Calculate the information entropy and entropy weight of each multidimensional security factor, and then weight and fuse the entropy weight with the preliminary weights to obtain the final weights of the multidimensional security factors, as follows:

[0011] in, For the first The final weights of the multidimensional security factors; For the first Preliminary weights of multiple security factors; For the first Entropy weight of a multidimensional security factor; The number of multidimensional security factors.

[0012] Optionally, in step 2, based on the final weights of the multidimensional safety factors, a fuzzy C-means clustering algorithm is used to classify electric vehicles, specifically as follows: Introducing multidimensional factor membership functions to construct a classification space ;in, This is a battery performance vector that includes two metrics: battery health status and charging reception capability. The device safety vector includes two indicators: the insulation level of the charging interface and the response speed of the protection device. It is an environmental adaptability vector, which includes two indicators: temperature sensitivity coefficient and altitude correction coefficient. The iterative optimization objective function of the fuzzy C-means clustering algorithm is as follows:

[0013] in, The number of data points; The number of clusters; For the first The data point belongs to the th data point Membership degree of each cluster; For fuzzy index; For the first The data point and the first The Euclidean distances between the centers of the clusters are as follows:

[0014] in, The number of safety factors; For the first The final weight of each security factor; For the first The first data point One factor value; For the first The first cluster center One factor value; In each iteration, the membership degree and cluster center are updated based on the current membership degree matrix and cluster center, as follows:

[0015] in, This is an index variable for the class cluster, used to traverse all class clusters;

[0016] By iterating continuously until the change in the objective function is less than the set threshold, the resulting cluster centers are the final electric vehicle classification results.

[0017] This invention also provides an electric vehicle classification system based on multidimensional safety factors, utilizing a multidimensional safety factor-based electric vehicle classification method, comprising: Multidimensional safety factor determination and acquisition module; used to determine and acquire multidimensional safety factors affecting electric vehicle charging safety; Multidimensional safety factor weight calculation module: used to obtain the preliminary weights of the multidimensional safety factors using the analytic hierarchy process (AHP), and to correct the preliminary weights using the entropy weight method to obtain the final weights of the multidimensional safety factors; Electric vehicle classification module: used to classify electric vehicles based on the final weights of the multidimensional safety factors using a fuzzy C-means clustering algorithm.

[0018] The present invention also provides an electronic device, comprising: Memory, used to store computer programs; A processor is used to implement a method for classifying electric vehicles based on multidimensional safety factors when executing the computer program.

[0019] The present invention also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of a method for classifying electric vehicles based on multidimensional safety factors.

[0020] As can be seen from the above technical solution, compared with the prior art, this invention discloses a method, system, device, and storage medium for electric vehicle classification based on multi-dimensional safety factors. First, by constructing a multi-dimensional safety factor system, six key safety factors are selected from three dimensions: battery performance, equipment safety, and environmental adaptability. These factors include battery health status, charging reception capability, charging interface insulation level, protection device response speed, temperature sensitivity coefficient, and altitude correction coefficient. Corresponding quantitative indicators and evaluation methods are proposed. This system comprehensively covers the main factors affecting the charging safety of electric vehicles, providing a solid foundation for subsequent classification research. Second, by using the Analytic Hierarchy Process (AHP) combined with the entropy weight method to determine the weights of each safety factor, the weight allocation effectively integrates expert experience and the variability of the data itself, making the weight allocation more scientific and reasonable. Based on this, the Fuzzy C-means Clustering (FCM) algorithm is introduced to achieve dynamic classification of electric vehicles. By establishing a multi-dimensional factor membership function, the fuzziness and uncertainty between the various safety factors are fully considered, improving the accuracy and adaptability of the classification. Ultimately, based on the clustering results of the FCM algorithm, electric vehicles were divided into four safety levels, and differentiated charging strategies were developed for each level to accurately match the safe charging needs of each vehicle. This fundamentally improved the safety of electric vehicle charging, reduced the accident rate, extended battery life, and reduced battery replacement costs. Attached Figure Description

[0021] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0022] Figure 1 This is a schematic diagram of the method flow provided by the present invention. Detailed Implementation

[0023] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. 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 are within the scope of protection of the present invention.

[0024] Example 1: Embodiment 1 of this invention discloses a method for classifying electric vehicles based on multi-dimensional safety factors, such as... Figure 1 As shown, it includes: Step 1: Identify the multidimensional safety factors affecting electric vehicle charging safety; The multidimensional safety factor is specifically divided into three dimensions: battery performance, equipment safety, and environmental adaptability.

[0025] Safety factors in battery performance dimensions include: battery health status and charging reception capability; wherein, the battery health status is quantified by indicators such as capacity decay rate and internal resistance growth rate; and the charging reception capability is quantified by indicators such as maximum safe charging current.

[0026] Battery health status ( Capacity decay rate (CFR) is a key indicator for measuring battery performance, and its accurate evaluation is crucial for ensuring the safe operation of electric vehicles and extending battery life. This invention uses CFR as the key indicator. ) and internal resistance growth rate ( Two indicators are used to comprehensively assess the battery's health status.

[0027] Capacity decay rate ( This reflects the degree of degradation of battery storage capacity over time, among which... Current capacity and initial capacity The difference. As the number of charge-discharge cycles increases, a series of irreversible chemical reactions occur inside the battery, such as the loss of active materials and the thickening of the solid electrolyte interphase (SEI) film, all of which lead to a gradual decrease in battery capacity. Internal resistance growth rate ( This is used to characterize the increase in the degree of internal polarization of the battery, where The current internal resistance and the initial internal resistance The difference in internal resistance. An increase in internal resistance not only leads to increased energy loss during charging and discharging, but also exacerbates battery heating, further affecting battery performance and safety.

[0028] To obtain accurate data on capacity decay rate and internal resistance growth rate, this invention employs a method combining electrochemical impedance spectroscopy (EIS) and charge-discharge cycle testing. EIS measures the battery's AC impedance at different frequencies, enabling in-depth analysis of the internal electrochemical processes and obtaining key parameters such as internal resistance and charge transfer resistance. The charge-discharge cycle test subjects the battery to multiple charge-discharge cycles under different operating conditions, monitoring changes in capacity and internal resistance in real time. Finally, a dynamic model of battery health state based on dual indicators is established.

[0029] This model comprehensively considers the impact of capacity decay rate and internal resistance growth rate on battery health, enabling it to more accurately reflect the actual health status of the battery. For example, when the capacity decay rate reaches 10% and the internal resistance growth rate reaches 20%, the model calculates... A value of 75% indicates that the battery's health has declined to some extent, and its performance needs to be closely monitored.

[0030] Charging acceptance capability refers to the maximum charging current a battery can accept under safe conditions, and it directly affects the charging speed and efficiency of electric vehicles. This invention uses the maximum safe charging current (… The core indicator is battery temperature, state of charge (SOC), etc. C) and factors such as cycle life constrain charging reception capability.

[0031] Battery temperature significantly impacts charging acceptance. At low temperatures, the rate of internal chemical reactions slows, and resistance to ion diffusion increases, leading to a decrease in charging acceptance. Below 0°C, the maximum safe charging current may drop to less than 50% of that at room temperature. Conversely, at high temperatures, the battery's thermal stability deteriorates, and excessively high charging currents can cause overheating, increasing the risk of thermal runaway. Therefore, it is necessary to adjust the charging current in real-time based on battery temperature to ensure a safe and efficient charging process.

[0032] State of charge ( C) is also closely related to charging reception capability. When C is low, the battery has less polarization and can accept a larger charging current. With... As the charge (C) increases, the polarization inside the battery gradually intensifies, and the charge acceptance capability gradually decreases. When When the charge level reaches 80% or higher, the charging current should be appropriately reduced to avoid overcharging the battery.

[0033] To determine the maximum safe charging current, this invention employs a pulse charging test method. By applying a series of pulse currents of different amplitudes to the battery while simultaneously monitoring parameters such as battery voltage and temperature, the critical thermal runaway current threshold of the battery under different operating conditions is determined. Based on this, combined with the battery's temperature, Based on factors such as C and cycle life, a quantitative model of charging reception capability is constructed:

[0034] in, For battery temperature, In a charged state, The model measures cycle life. It accurately predicts the maximum safe charging current based on the battery's real-time state, providing crucial information for optimizing charging strategies. For example, when the battery temperature is 25°C, the SOC is 50%, and the cycle life is 500 cycles, the model calculates a maximum safe charging current of 100A. Using this current ensures both charging speed and battery safety.

[0035] Safety factors in the equipment safety dimension include: the insulation level of the charging interface and the response speed of the protection device; wherein, the insulation level of the charging interface is quantified by the insulation resistance; and the response speed of the protection device is quantified by the overcurrent protection action time.

[0036] The insulation class of the charging interface is a crucial indicator for ensuring the safe operation of charging equipment, and its performance directly affects the safety of personnel and equipment. This invention uses insulation resistance (…) As a core parameter, the insulation performance of the charging interface is rigorously evaluated in accordance with relevant standards.

[0037] A 500V DC withstand voltage tester was used to measure the insulation resistance of the charging interface to ground. This method can effectively detect the insulation performance of the interface under high voltage. Meanwhile, considering the influence of environmental factors such as humidity and dust pollution on insulation performance, an insulation performance degradation model was established.

[0038] in, For ambient humidity, Dust pollution level, This refers to the usage time. As humidity increases and dust pollution intensifies, the insulation resistance gradually decreases, and the longer the usage time, the more significant the degradation of insulation performance.

[0039] Based on the magnitude of insulation resistance, the insulation class of the charging interface is divided into Class I (…). Levels IV and IV There are four levels. Level I indicates excellent insulation performance, effectively preventing leakage accidents; Level II ( ) and Level III ( The insulation performance of each class decreases sequentially, requiring regular inspection and maintenance; Class IV indicates that the insulation resistance is too low, posing a serious safety hazard, and should be stopped immediately for repair or replacement.

[0040] The response speed of a protection device is one of the key indicators for measuring the safety of charging equipment. It directly determines whether the protection device can promptly cut off the circuit and prevent the accident from escalating when a fault occurs during charging. This invention addresses this issue by controlling the overcurrent protection action time (…). This is used to quantify the response speed of the protection device.

[0041] An oscilloscope is used to acquire the current-time curve during a short-circuit fault, and the time from the sudden change in current to the protection device cutting off the circuit is defined as the response time. The oscilloscope can accurately measure changes in current and time intervals, providing a reliable means to accurately obtain the response time of the protection device.

[0042] Establish a coupling model between response speed and temperature and aging degree:

[0043] in, To protect the operating temperature of the device, The degree of aging is a factor. Increased temperature and aging will both slow down the response speed of the protection device. When the temperature rises from 25°C to 50°C, the overcurrent protection action time may increase by about 20%; when the aging level reaches 50%, the response time may increase by more than 50%.

[0044] Set a security threshold ( When the response time of the protection device exceeds this threshold, it indicates that its performance has deteriorated and it needs to be repaired or replaced in a timely manner to ensure that the circuit can be quickly cut off in the event of a fault and to ensure charging safety.

[0045] Safety factors in the environmental adaptability dimension include: temperature sensitivity coefficient and altitude correction coefficient; wherein, the temperature sensitivity coefficient is quantified by the charging efficiency temperature decay rate; and the altitude correction coefficient is quantified by the influence rate of air pressure on discharge capacity.

[0046] The temperature sensitivity coefficient characterizes how sensitive battery charging efficiency is to temperature changes, and it is of great significance for optimizing charging strategies under different ambient temperatures. This invention uses the charging efficiency temperature decay rate (… The temperature sensitivity coefficient is defined by the following formula:

[0047] in, The charging efficiency is at room temperature (25℃). The charging efficiency is measured at high temperature (45℃).

[0048] Through variable-temperature environment charging experiments, battery charging tests were conducted under different temperature conditions. Parameters such as current, voltage, and time were recorded during the charging process, and the charging efficiency at different temperatures was calculated. Based on the experimental data, temperature decay curves for different battery types were fitted, and an environmental temperature compensation model was established.

[0049] in, This refers to the actual charging efficiency. The charging efficiency at standard temperature. This is the temperature compensation coefficient. This represents the difference between the ambient temperature and the standard temperature. The model can adjust the charging strategy in real time based on the ambient temperature to improve charging efficiency. For example, when the ambient temperature is 35℃, the charging efficiency calculated by the model is 85% of that at room temperature. In this case, the charging current can be appropriately reduced to avoid overheating of the battery while simultaneously improving charging efficiency.

[0050] Changes in altitude lead to changes in atmospheric pressure, which in turn affects the battery's discharge capacity and charging performance. This invention is based on the effect of air pressure on discharge capacity (…). To quantify the altitude correction factor, where, Discharge capacity at different altitudes and discharge capacity at standard altitude (0m) The difference.

[0051] The capacity decay characteristics of batteries at different altitudes (0-5000m) were tested using a high-altitude simulation chamber. The chamber simulated the air pressure environment at different altitudes, and charge-discharge tests were conducted on the batteries, recording the changes in discharge capacity. Based on the test data, a mathematical model of altitude-capacity decay was established.

[0052] in, Altitude and These are the model parameters. This model allows for the calculation of capacity decay rates at different altitudes, leading to the altitude correction coefficient. :

[0053] When charging in high-altitude areas, charging parameters are adjusted according to the altitude correction factor, such as appropriately increasing the charging voltage and extending the charging time, to ensure that the battery can be fully charged and improve the performance of electric vehicles in high-altitude areas. For example, in an area with an altitude of 3000m, the altitude correction factor calculated by the model is 0.8. In this case, the charging voltage should be increased by 20% to compensate for the battery capacity reduction caused by the increase in altitude.

[0054] The collected data often has problems such as inconsistent dimensions, missing data or anomalies. Therefore, after obtaining the multi-dimensional safety factors that affect the charging safety of electric vehicles, the process also includes: preprocessing the multi-dimensional safety factor data. Preprocessing, specifically: The data is normalized using the 0-1 normalization method, mapping all data to... Interval, eliminating the influence of dimensions:

[0055] in, This is the original data; and These are the minimum and maximum values ​​of the indicator, respectively.

[0056] In addition, the 3σ principle is used to remove outliers from the data. That is, if the deviation of a data point from the mean exceeds 3 times the standard deviation, it is judged as an outlier. Suppose the mean of a certain battery health status index is 0.8 and the standard deviation is 0.1. If a certain data point is 1.2, then the data point exceeds the 3σ range (0.8 ± 3 × 0.1) and should be removed.

[0057] After normalization and outlier removal, a standardized dataset is formed. This corresponds to six safety factor indicators, providing high-quality data support for subsequent classification analysis.

[0058] Step 2: Use the analytic hierarchy process (AHP) to obtain the preliminary weights of the multidimensional security factors, and then use the entropy weight method to correct the preliminary weights to obtain the final weights of the multidimensional security factors.

[0059] In step 2, the preliminary weights of the multidimensional security factors are obtained using the analytic hierarchy process (AHP), specifically as follows: Electric vehicle charging safety is taken as the target layer, the various dimensions affecting electric vehicle charging safety are taken as the criterion layer, and the safety factors contained in each dimension are taken as the solution layer. The relative importance of the indicators in the criterion layer and the scheme layer is scored using the 1-9 scale to obtain the judgment matrix for the criterion layer and the scheme layer. Regarding the 1-9 scale, 1 indicates that the two factors are equally important; 3 indicates that one factor is slightly more important than the other; 5 indicates that one factor is significantly more important than the other; 7 indicates that one factor is strongly more important than the other; 9 indicates that one factor is extremely more important than the other; 2, 4, 6, and 8 are the median values ​​of the above adjacent judgments. For example, when the importance of one factor is between slightly more important and significantly more important than the other, it can be represented by 4.

[0060] The eigenvalue method is used to solve each judgment matrix to obtain the largest eigenvalue and the corresponding eigenvector, and the eigenvector is normalized to obtain the weight vector of each security factor. The core principle of the eigenvalue method is: weight vector W Essentially, it's a judgment matrix. The corresponding largest eigenvalue The eigenvectors. Specifically, by solving the homogeneous linear equation system. (here) (representing the identity matrix), which can obtain the largest eigenvalue. The corresponding eigenvector. To ensure that this eigenvector satisfies the practical meaning and properties of a weight vector—that is, all elements are non-negative and their sum is 1—the obtained eigenvector needs to be normalized. The final vector is the weight vector. .

[0061] A consistency check is performed. If the random consistency ratio is less than a preset value, the preliminary weight determination is completed. If the random consistency ratio is greater than or equal to the preset value, the scoring is repeated until the random consistency ratio is less than the preset value. The random consistency ratio is as follows:

[0062] in, The random consistency ratio; As a consistency indicator; It is a random consistency indicator; To determine the largest eigenvalue in the matrix; To determine the order of a matrix.

[0063] The initial weights are corrected using the entropy weight method, specifically as follows: Entropy weighting is an objective weighting method based on the degree of variation in data. The greater the degree of variation in data, the more information it carries, and the greater its corresponding weight. The information entropy and entropy weight of each multidimensional safety factor are calculated. If a certain indicator shows a large difference in value among different vehicles, it indicates that the indicator contributes significantly to vehicle classification, and its entropy weight will increase accordingly.

[0064] The entropy weight is then weighted and fused with the initial weight to obtain the final weight of the multidimensional security factor, as follows:

[0065] in, For the first The final weights of the multidimensional security factors; For the first Preliminary weights of multiple security factors; For the first Entropy weight of a multidimensional security factor; The number of multidimensional security factors.

[0066] After correction using the entropy weight method, the weights of each factor are more objective and accurate, and can better reflect the actual situation.

[0067] Based on the final weights of multidimensional safety factors, fuzzy C-means clustering (FCM) algorithm is used to classify electric vehicles. This algorithm can effectively handle the fuzziness and uncertainty of data and is suitable for classification problems of multidimensional and dynamically changing safety factors in electric vehicles. Specifically: Introducing multidimensional factor membership functions to construct a classification space ;in, This is a battery performance vector that includes two metrics: battery health status and charging reception capability. The device safety vector includes two indicators: the insulation level of the charging interface and the response speed of the protection device. It is an environmental adaptability vector, which includes two indicators: temperature sensitivity coefficient and altitude correction coefficient. The core idea of ​​the FCM algorithm is to achieve optimal membership of each data point to each cluster center by iteratively optimizing the objective function, thereby realizing data clustering. Its iterative optimization objective function is as follows:

[0068] in, The number of data points; The number of clusters; For the first The data point belongs to the th data point Membership degree of each cluster; The fuzzy index (usually taken as 2); For the first The data point and the first The Euclidean distances between the centers of the clusters are as follows:

[0069] in, The number of safety factors; For the first The final weight of each security factor; For the first The first data point One factor value; For the first The first cluster center One factor value; In each iteration, the membership degree and cluster center are updated based on the current membership degree matrix and cluster center, as follows:

[0070] in, This is an index variable for the class cluster, used to traverse all class clusters;

[0071] By iterating continuously until the change in the objective function is less than a set threshold, the resulting cluster centers are the final electric vehicle classification results. This iterative optimization process fully considers the interrelationships between various safety factors and the distribution characteristics of the data, achieving dynamic clustering of vehicle safety levels.

[0072] The clustering results obtained by the fuzzy C-means clustering algorithm are divided into four safety levels. Differentiated charging strategies are formulated for vehicles of different safety levels, as follows: Level I (Safety Priority): All factors are better than the safety threshold, suitable for fast charging mode. These vehicles have batteries in good health, with capacity decay rate and internal resistance growth rate within normal ranges. They have strong charging acceptance capabilities and can withstand large charging currents; the charging interface has a high insulation level with an insulation resistance greater than 100MΩ; the protection device responds quickly, with overcurrent protection action time less than 50μs; they have strong environmental adaptability, with minimal impact from temperature sensitivity coefficients and altitude correction coefficients on charging performance. In fast charging mode, they can complete charging quickly and efficiently while ensuring charging safety.

[0073] Level II (Balanced): Some factors degrade to the edge, requiring limitation of charging current. At this level, some safety factors of the vehicle show a certain degree of decline, but are still within acceptable ranges. For example, battery health may show slight capacity decay or internal resistance increase, resulting in a decrease in charging acceptance; the insulation level of the charging interface may drop to Level II (100MΩ > 100MΩ). (≥50MΩ), the response speed of the protection device will be slightly slower. To ensure charging safety, it is necessary to appropriately limit the charging current to avoid safety accidents caused by overheating or other abnormalities during the charging process.

[0074] Level III (Risk Warning): Single factor exceeds threshold, initiating charging protection strategy. When a safety factor exceeds a safety threshold, it indicates a certain safety risk to the vehicle. For example, if the capacity decay rate in the battery health status indicators exceeds 15%, or the insulation resistance of the charging interface is below 10MΩ, the charging protection strategy should be activated immediately, such as reducing charging power or stopping charging, and a warning signal should be issued in a timely manner to remind users and relevant management personnel to inspect and maintain the vehicle.

[0075] Level IV (High Risk): Multiple factors exceed limits, charging is prohibited and a maintenance warning is triggered. If multiple safety factors simultaneously exceed safety thresholds, it indicates a serious safety hazard in the vehicle. Charging must be prohibited to prevent serious accidents. A maintenance warning is also triggered, notifying professional repair personnel to conduct a comprehensive inspection of the vehicle, identify and fix any safety hazards, and only resume normal charging after the problems are resolved.

[0076] In the field of smart charging piles, the output of the classification model is deeply integrated with the control strategy of the charging pile, achieving precise adaptation of charging parameters. When a vehicle connects to the charging pile, the charging pile obtains multi-dimensional safety factor data through the communication system with the vehicle and inputs it into the classification model for real-time analysis.

[0077] Based on the classification results, charging stations can provide personalized charging strategies for vehicles with different safety levels. For Level I (safety-first) vehicles, charging stations can provide fast charging at up to 1.5C, quickly replenishing the vehicle's charge to meet users' urgent travel needs. This is because these vehicles exhibit excellent safety performance and can withstand higher charging currents; fast charging not only avoids damage to the vehicle but also significantly reduces charging time.

[0078] For Level II (Balanced) vehicles, considering the marginal degradation of some safety factors, the charging station will automatically limit the charging current to 1.0C to ensure both charging efficiency and safety during the charging process. This moderate current limitation can prevent battery overheating or other safety issues caused by excessive charging current, protecting the vehicle's battery and charging equipment.

[0079] For Level III (risk warning type) and Level IV (high-risk type) vehicles, the charging station will immediately activate trickle charging mode and work in real time with the vehicle's Battery Management System (BMS) to monitor the charging process comprehensively. Trickle charging mode charges the battery with a smaller current, effectively reducing battery polarization and minimizing battery wear. Simultaneously, the real-time linkage with the BMS can promptly detect and handle abnormal situations during charging, such as battery overheating or overvoltage, ensuring charging safety. Once an abnormality is detected, the charging station will immediately stop charging, issue an alarm, and notify the user and relevant maintenance personnel for handling.

[0080] Through this "one vehicle, one policy" charging strategy, smart charging stations can provide the most suitable charging parameters based on the actual safety status of each vehicle, greatly improving charging safety and efficiency, and bringing a more convenient and reliable charging experience to electric vehicle users.

[0081] In the entire life cycle management of batteries, the classification results play an important role, providing a scientific basis for decisions on the cascade utilization and recycling of batteries.

[0082] When electric vehicle batteries reach a certain age or their performance deteriorates to a certain level, they need to be evaluated to determine their suitability for reuse. Analyzing the multi-dimensional safety factors of batteries using a classification model can accurately assess their remaining value and performance status. While Class I and Class II batteries may no longer meet the performance requirements for electric vehicles, they still have significant value in scenarios with relatively lower battery performance requirements, such as energy storage. These batteries can be retested, sorted, and reassembled for use in energy storage power stations, backup power supplies, and other fields, maximizing resource utilization.

[0083] For Class III and Class IV batteries, given their numerous safety concerns and performance issues, direct dismantling and recycling is a more suitable option. During dismantling and recycling, advanced physical and chemical methods are employed to efficiently recover valuable metals (such as lithium, cobalt, and nickel) from the batteries, reducing resource waste and environmental pollution. Simultaneously, the recovered metals are refined and purified, enabling their reuse in battery production and other fields, thus achieving resource recycling.

[0084] By adopting a battery lifecycle management strategy based on classification results, we can effectively improve the utilization efficiency of battery resources, reduce the cost of using electric vehicles, reduce the impact on the environment, and promote the development of the electric vehicle industry towards a green and sustainable direction.

[0085] Example 2: Embodiment 2 of the present invention discloses an electric vehicle classification system based on multidimensional safety factors, comprising: Multidimensional safety factor determination and acquisition module; used to determine and acquire multidimensional safety factors affecting electric vehicle charging safety; Multidimensional safety factor weight calculation module: used to obtain the preliminary weights of the multidimensional safety factors using the analytic hierarchy process (AHP), and to correct the preliminary weights using the entropy weight method to obtain the final weights of the multidimensional safety factors; Electric vehicle classification module: used to classify electric vehicles based on the final weights of the multidimensional safety factors using a fuzzy C-means clustering algorithm.

[0086] Example 3: Embodiment 3 of the present invention also discloses an electronic device, comprising: Memory, used to store computer programs; A processor is used to implement a method for classifying electric vehicles based on multidimensional safety factors when executing the computer program.

[0087] Example 4: Embodiment 4 of the present invention discloses a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of a method for classifying electric vehicles based on multidimensional safety factors.

[0088] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the apparatus disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the description is relatively simple; relevant parts can be referred to the method section.

[0089] The above description of the disclosed embodiments enables those skilled in the art to make or use the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for classifying electric vehicles based on multidimensional safety factors, characterized in that, include: Step 1: Identify and obtain the multi-dimensional safety factors that affect the charging safety of electric vehicles; Step 2: Use the analytic hierarchy process (AHP) to obtain the preliminary weights of the multidimensional security factors, and then use the entropy weight method to correct the preliminary weights to obtain the final weights of the multidimensional security factors. Step 3: Based on the final weights of the multidimensional safety factors, fuzzy C-means clustering algorithm is used to classify electric vehicles.

2. The electric vehicle classification method based on multi-dimensional safety factors according to claim 1, characterized in that, In step 1, the multidimensional safety factor is specifically divided into three dimensions: battery performance, device safety, and environmental adaptability. Safety factors in battery performance include: battery health status and charging reception capability; wherein, the battery health status is quantified by capacity decay rate and internal resistance growth rate; and the charging reception capability is quantified by the maximum safe charging current. Safety factors in the equipment safety dimension include: the insulation level of the charging interface and the response speed of the protection device; wherein, the insulation level of the charging interface is quantified by the insulation resistance; and the response speed of the protection device is quantified by the overcurrent protection action time. Safety factors in the environmental adaptability dimension include: temperature sensitivity coefficient and altitude correction coefficient; wherein, the temperature sensitivity coefficient is quantified by the charging efficiency temperature decay rate; and the altitude correction coefficient is quantified by the influence rate of air pressure on discharge capacity.

3. The electric vehicle classification method based on multi-dimensional safety factors according to claim 1, characterized in that, In step 1, after obtaining the multi-dimensional safety factors affecting the charging safety of electric vehicles, the following steps are also included: preprocessing the multi-dimensional safety factor data. The preprocessing specifically includes: The data is normalized using the 0-1 normalization method, mapping all data to... The interval, and the use of the 3σ principle to remove outliers from the data.

4. The electric vehicle classification method based on multi-dimensional safety factors according to claim 1, characterized in that, In step 2, the preliminary weights of the multidimensional security factors are obtained using the analytic hierarchy process (AHP), specifically as follows: Electric vehicle charging safety is taken as the target layer, the various dimensions affecting electric vehicle charging safety are taken as the criterion layer, and the safety factors contained in each dimension are taken as the solution layer. The relative importance of the indicators of the criterion layer and the scheme layer is scored based on the 1-9 scaling method to obtain the judgment matrix of the criterion layer and the scheme layer. The maximum eigenvalue and the corresponding eigenvector are obtained by solving each judgment matrix using the eigenvalue method, and the eigenvector is normalized to obtain the weight vector of each security factor. A consistency check is performed. When the random consistency ratio is less than a preset value, the preliminary weight is determined. When the random consistency ratio is greater than or equal to the preset value, the scoring is repeated until the random consistency ratio is less than the preset value.

5. The electric vehicle classification method based on multi-dimensional safety factors according to claim 1, characterized in that, In step 2, the initial weights are corrected using the entropy weight method, specifically as follows: Calculate the information entropy and entropy weight of each multidimensional security factor, and then weight and fuse the entropy weight with the preliminary weight to obtain the final weight of the multidimensional security factor, as follows: in, For the first The final weights of the multidimensional security factors; For the first Preliminary weights of multiple security factors; For the first Entropy weight of a multidimensional security factor; The number of multidimensional security factors.

6. The electric vehicle classification method based on multi-dimensional safety factors according to claim 1, characterized in that, In step 2, based on the final weights of the multidimensional safety factors, fuzzy C-means clustering algorithm is used to classify electric vehicles, specifically as follows: Introducing multidimensional factor membership functions to construct a classification space ;in, This is a battery performance vector that includes two metrics: battery health status and charging reception capability. The device safety vector includes two indicators: the insulation level of the charging interface and the response speed of the protection device. It is an environmental adaptability vector, which includes two indicators: temperature sensitivity coefficient and altitude correction coefficient. The iterative optimization objective function of the fuzzy C-means clustering algorithm is as follows: in, The number of data points; The number of clusters; For the first The data point belongs to the th data point The membership degree of each cluster; For fuzzy index; For the first The data point and the first The Euclidean distances between the centers of the clusters are as follows: in, The number of safety factors; For the first The final weight of each security factor; For the first The first data point One factor value; For the first The first cluster center One factor value; In each iteration, the membership degree and cluster center are updated based on the current membership degree matrix and cluster center, as follows: in, This is an index variable for the class cluster, used to traverse all class clusters; By iterating continuously until the change in the objective function is less than the set threshold, the resulting cluster centers are the final electric vehicle classification results.

7. An electric vehicle classification system based on multi-dimensional safety factors, utilizing the electric vehicle classification method based on multi-dimensional safety factors as described in any one of claims 1-6, characterized in that, include: Multidimensional safety factor determination and acquisition module; Used to identify and acquire multidimensional safety factors that affect the charging safety of electric vehicles; Multidimensional safety factor weight calculation module: used to obtain the preliminary weights of the multidimensional safety factors using the analytic hierarchy process (AHP), and to correct the preliminary weights using the entropy weight method to obtain the final weights of the multidimensional safety factors; Electric vehicle classification module: used to classify electric vehicles based on the final weights of the multidimensional safety factors using a fuzzy C-means clustering algorithm.

8. An electronic device, characterized in that, include: Memory, used to store computer programs; A processor, configured to implement the steps of the electric vehicle classification method based on multidimensional safety factors as described in any one of claims 1-6 when executing the computer program.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the steps of the electric vehicle classification method based on multidimensional safety factors as described in any one of claims 1-6.

Images