Risk assessment method and device, terminal and storage medium

Abstract

The invention relates to the technical field of battery management, and discloses a risk assessment method and device, a terminal and a storage medium, and the method comprises the steps: obtaining the health state variation of each target battery cell through a first charging and discharging operation, screening out a first battery cell of a first risk level from all the target battery cells based on the health state variation; obtaining the self-discharge rate growth rate of each second battery monomer except the first battery monomer in all the target battery monomers, and screening out a third battery monomer of the first risk level again based on the self-discharge rate growth rate; and obtaining multi-dimensional response parameters of each fourth battery monomer except the third battery monomer in all the second battery monomers during execution of the second charging and discharging operation, and obtaining a risk level of each fourth battery monomer in combination with the risk assessment model. According to the invention, nondestructive and accurate safety risk grading evaluation of the battery after slight thermal abuse can be realized.

Description

Technical Field

[0001] This application relates to the field of battery management technology, and in particular to a risk assessment method, device, terminal and storage medium. Background Technology

[0002] With the rapid development of new energy vehicles and energy storage systems, lithium-ion batteries have become the mainstream energy storage device due to their advantages such as high energy density, long cycle life, and lightweight design. However, during battery use, minor thermal abuse events such as valve opening and micro-short circuits may occur due to overcharging, short circuits, mechanical damage, or localized heat accumulation. Although these events do not lead to severe thermal runaway, they may cause irreversible electrochemical degradation within individual battery cells, thereby affecting the safety and reliability of the entire module.

[0003] In existing technologies, the assessment of battery safety status largely relies on online monitoring parameters such as voltage, temperature, and internal resistance during operation, or on establishing health status models based on historical charge and discharge data for trend prediction. However, when a battery experiences minor thermal abuse, its macroscopic performance parameters, such as capacity and voltage curves, may not show significant changes, making it difficult for traditional methods to identify potential risks in a timely manner. Summary of the Invention

[0004] In view of this, embodiments of this application provide a risk assessment method, apparatus, terminal, and storage medium to achieve non-destructive and accurate safety risk classification assessment of batteries after experiencing minor thermal abuse.

[0005] In a first aspect, embodiments of this application provide a risk assessment method for assessing a single battery cell, the method comprising: The health status change of each target battery cell is obtained through the first charge and discharge operation, and the first battery cell with the first risk level is selected from all the target battery cells based on the health status change; the target battery cell is the battery cell other than the valve-opening battery cell. Obtain the self-discharge rate growth rate of each second battery cell other than the first battery cell in all the target battery cells, and further filter out the third battery cells that are at the first risk level based on the self-discharge rate growth rate. The multidimensional response parameters of each fourth battery cell (excluding the third battery cell) in the second battery cell are obtained when performing the second charge-discharge operation, and combined with the risk assessment model, the risk level of each fourth battery cell is obtained; among them, the first risk level has the highest risk.

[0006] In an optional implementation, the step of acquiring the change in health status of each target battery cell through a first charge-discharge operation, and selecting the first battery cell belonging to the first risk level from all the target battery cells based on the change in health status, includes: A first charge-discharge operation is performed on each of the target battery cells to determine the change in the health status of the corresponding target battery cell based on the capacity change of each target battery cell after the first charge-discharge operation. The target battery cell whose health status change is greater than or equal to the first judgment threshold is designated as the first battery cell of the first risk level.

[0007] In an optional implementation, before performing the first charge-discharge operation on each of the target battery cells, the method further includes: obtaining the initial nominal capacity of each of the target battery cells; The step of performing the first charge-discharge operation on each of the target battery cells to determine the corresponding change in health status based on the capacity change of each target battery cell after the first charge-discharge operation includes: Perform at least two first charge-discharge operations on each of the target battery cells, and obtain the average battery cell capacity based on the battery cell capacity after each first charge-discharge operation. The health status of a battery cell is obtained based on the initial nominal capacity and the average battery cell capacity of the target battery cell, and the change in health status is determined based on the health status of the battery cell.

[0008] In an optional implementation, obtaining the self-discharge rate growth rate of each of the second battery cells (excluding the first battery cell) among all the target battery cells, and further screening out the third battery cells that meet the first risk level based on the self-discharge rate growth rate, includes: Perform a second charge-discharge operation and a resting operation on each of the second battery cells to obtain the self-discharge rate growth rate of each of the second battery cells relative to the initial self-discharge rate. The battery cell whose self-discharge rate growth rate is greater than or equal to the second judgment threshold is designated as the third battery cell of the first risk level.

[0009] In an optional implementation, obtaining the self-discharge rate growth rate of each of the second battery cells relative to the initial self-discharge rate at the time of manufacture includes: The self-discharge rate of each second battery cell is calculated based on the obtained static voltage at the start and end times within a preset time before the static operation ends. The self-discharge rate growth rate of each second battery cell is determined based on the current self-discharge rate of each second battery cell, the initial self-discharge rate at the time of manufacture, and the adjustment coefficient.

[0010] In an optional implementation, the charging operation in the second charge-discharge operation is a constant current constant voltage charging operation; the constant current constant voltage charging operation includes a constant current charging stage and a constant voltage charging stage. The process of obtaining multidimensional response parameters of each of the fourth battery cells (excluding the third battery cell) during the second charge-discharge operation, and combining these parameters with a risk assessment model to obtain the risk level of each of the fourth battery cells, includes: The voltage standard deviation of each of the fourth battery cells during the constant current charging phase, the duration of the constant voltage charging phase, the temperature change characteristics during the charging operation of the second charge-discharge operation, and the voltage change obtained after the resting operation are extracted as the multidimensional response parameters. The multidimensional response parameters are input into the risk assessment model to obtain the risk score of each of the fourth battery cells; The risk of each of the fourth battery cells is ranked based on the risk score, and the risk level of each of the fourth battery cells is determined based on the ranking result; wherein, the risk level of the fourth battery cell includes a second risk level and a third risk level; the risk of the second risk level is higher than the risk of the third risk level.

[0011] In an optional implementation, the construction of the risk assessment model includes: Multiple training sample battery cells with known risk states are acquired, the second charge-discharge operation is performed, and the corresponding multidimensional response parameters are collected as input features; wherein, the known risk states are used as classification labels. The random forest algorithm is used to model the training dataset consisting of the input features and the classification labels, and the input features with importance scores higher than a preset importance threshold are selected as key features by the importance ranking method. After normalizing the importance scores of the key features, they are used as the weight coefficients of the corresponding key features in the risk score, so as to obtain a risk assessment model composed of the key features and their corresponding weight coefficients.

[0012] In an optional implementation, the target battery cell is a battery cell that extends outward from the valve-opening battery cell within a preset threshold range.

[0013] In an optional implementation, the adjustment coefficient is dynamically adjusted based on the cumulative number of cycles, capacity decay rate, and ambient temperature of each of the second battery cells.

[0014] Secondly, embodiments of this application provide a risk assessment device for assessing individual battery cells, the device comprising: The initial screening module is used to obtain the change in health status of each target battery cell through the first charge and discharge operation, and to screen out the first battery cell with the first risk level from all the target battery cells based on the change in health status; the target battery cell is a battery cell other than the valve-opening battery cell. The re-screening module is used to obtain the self-discharge rate growth rate of each second battery cell other than the first battery cell in all the target battery cells, and to screen out the third battery cells that are at the first risk level based on the self-discharge rate growth rate. The risk determination module is used to obtain the multi-dimensional response parameters of each fourth battery cell (excluding the third battery cell) in all the second battery cells when performing the second charge and discharge operation, and combine them with the risk assessment model to obtain the risk level of each fourth battery cell; wherein, the first risk level has the highest risk.

[0015] Thirdly, embodiments of this application provide a terminal device, the terminal device including a processor and a memory, the memory storing a computer program, and the processor executing the computer program to implement the risk assessment method described above.

[0016] Fourthly, embodiments of this application provide a computer-readable storage medium storing a computer program that, when executed on a processor, implements the aforementioned risk assessment method.

[0017] The embodiments of this application have the following beneficial effects: Firstly, this application performs preliminary screening based on changes in health status, quickly identifying high-risk battery cells due to significant capacity decay and preventing them from entering subsequent use stages. Then, by introducing the self-discharge rate growth rate as a deeper indicator, it effectively distinguishes between abnormal self-discharge caused by thermal abuse and natural decay during normal service, thereby improving the accuracy of the judgment. For battery cells not eliminated in the first two rounds of screening, its multi-dimensional response parameters during the charging and discharging process are further extracted and combined with a machine learning model for comprehensive scoring, achieving refined stratification of battery cells other than those with high risk. It is understood that the entire evaluation process does not require disassembling the battery or damaging its structure; it is entirely based on standard charging and discharging operations and non-invasive data acquisition, which not only ensures the reusability of the battery but also significantly reduces testing costs and time. Through this method, this application can identify battery cells that still have usable value to the greatest extent possible while ensuring safety. Attached Figure Description

[0018] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 A first flowchart of the risk assessment method according to an embodiment of this application is shown; Figure 2 A second flowchart of the risk assessment method according to an embodiment of this application is shown; Figure 3 A schematic diagram of the third process of the risk assessment method according to an embodiment of this application is shown; Figure 4 A schematic diagram of the fourth process of the risk assessment method according to an embodiment of this application is shown; Figure 5 The fifth flowchart of the risk assessment method according to an embodiment of this application is shown; Figure 6 A sixth flowchart of the risk assessment method according to an embodiment of this application is shown; Figure 7 The seventh flowchart of the risk assessment method according to an embodiment of this application is shown; Figure 8 A schematic diagram of a risk assessment device according to an embodiment of this application is shown. Detailed Implementation

[0020] The technical solutions in the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments.

[0021] The components of the embodiments of this application described and illustrated in the accompanying drawings can be arranged and designed in a variety of different configurations. Therefore, the following detailed description of the embodiments of this application provided in the drawings is not intended to limit the scope of the claimed application, but merely to illustrate selected embodiments of the application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0022] In the following text, the terms "comprising," "having," and their cognates, which may be used in various embodiments of this application, are intended only to indicate a particular feature, number, step, operation, element, component, or combination thereof, and should not be construed as primarily excluding the presence of one or more other features, numbers, steps, operations, elements, components, or combinations thereof, or adding the possibility of one or more combinations thereof. Furthermore, the terms "first," "second," "third," etc., are used only for distinguishing descriptions and should not be construed as indicating or implying relative importance.

[0023] Unless otherwise specified, all terms used herein (including technical and scientific terms) shall have the same meaning as commonly understood by one of ordinary skill in the art to which the various embodiments of this application pertain. Terms (such as those defined in commonly used dictionaries) shall be interpreted as having the same meaning as in their contextual meaning in the relevant technical field and shall not be construed as having an idealized or overly formal meaning, unless clearly defined in the various embodiments of this application.

[0024] The following detailed description of some embodiments of this application is provided in conjunction with the accompanying drawings. Unless otherwise specified, the following embodiments and features can be combined with each other.

[0025] The risk assessment method will be explained below with reference to some specific examples.

[0026] Figure 1 A flowchart illustrating a risk assessment method according to an embodiment of this application is shown. Exemplarily, this risk assessment method is used to assess the safety status of surrounding battery cells that remain outwardly intact after a minor thermal abuse event, such as valve opening, has occurred in a battery cell within a battery module, in order to identify potentially high-risk battery cells. The method includes steps S110-S130: Step S110: The health status change of each target battery cell is obtained through the first charge and discharge operation, and the first battery cell with the first risk level is selected from all target battery cells based on the health status change.

[0027] The target battery cell refers to any battery cell other than the valve-opening battery cell. Specifically, the target battery cell is the battery cell extending outward from the valve-opening battery cell within a preset threshold range. This preset threshold range can be understood as the spatial area covered by a preset number of loops extending along rows and columns in the plane of the battery module, with the valve-opening battery cell as the core. The preset number of loops can be 1 loop, 2 loops, etc., and can be set as needed. For example, in a module composed of square battery cells, this area includes multiple neighboring battery cells surrounding the central battery cell, thus effectively covering surrounding units that may be affected by heat conduction.

[0028] In some implementations, determining whether a battery belongs to the target battery cell requires meeting the following physical integrity conditions: no visible damage, intact insulation layer, closed top cover, and measurable positive and negative terminal voltages within the normal operating voltage range. For ternary lithium-ion batteries, the positive and negative terminal voltages should be between 2.8V and 4.25V; for lithium iron phosphate batteries, the positive and negative terminal voltages should be between 2.5V and 3.65V. These voltage ranges are set based on the actual full-charge and discharge cut-off voltages of the battery chemical system. If different specifications exist for the same type of battery, adjustments are made according to their factory parameters, ensuring that the voltage is measurable and within the normal operating voltage range.

[0029] In some implementations, such as Figure 2 As shown, step S110 includes steps S210-S220: Step S210: Perform a first charge-discharge operation on each target battery cell to determine the change in the health status of the corresponding target battery cell based on the capacity change of each target battery cell after the first charge-discharge operation.

[0030] In some implementations, prior to step S210, the initial nominal capacity of each target battery cell is obtained.

[0031] The initial nominal capacity refers to the rated capacity of each target battery cell recorded at the time of manufacture, in ampere-hours (Ah). The initial nominal capacity of each battery cell can be denoted as... ;in This represents the initial nominal capacity of the Nth target battery cell. The initial nominal capacity of each of the above target battery cells can be obtained from the production database, BMS historical records, or product labels. If original data is lacking for individual battery cells, the typical value of the same batch and model of battery cells can be used as a substitute.

[0032] like Figure 3 As shown, step S210 specifically includes steps S310-S320: Step S310: Perform at least two first charge-discharge operations on each target battery cell, and obtain the average battery cell capacity based on the battery cell capacity after each first charge-discharge operation.

[0033] Exemplarily, the first charge-discharge operation is performed at a standard ambient temperature (e.g., within the range of 20°C to 30°C). The charging process employs a constant current and constant voltage method, charging at a current of 0.33 times the current nominal capacity (i.e., 0.33C) until the full-charge voltage is reached. For ternary lithium-ion batteries, the full-charge voltage is typically 4.25V; for lithium iron phosphate batteries, the full-charge voltage is typically 3.65V. Subsequently, the charging mode is switched to constant voltage until the charging current decays to 0.05C, at which point charging stops. After charging is complete, a first preset time is allowed (the first preset time can be any value between 20 and 60 minutes, such as 30 minutes). In this embodiment, the first preset time is primarily used to eliminate polarization effects during charging, allowing the charge distribution within the battery cells to tend towards equilibrium, thereby obtaining a stable open-circuit voltage state.

[0034] The discharge process also uses a constant current discharge of 0.33C until the discharge reaches the specified discharge cutoff voltage of the battery. For ternary lithium-ion batteries, the discharge cutoff voltage is usually 2.8V, and for lithium iron phosphate batteries, the discharge cutoff voltage is usually 2.5V. After the discharge is completed, the battery is left to stand for the first preset time.

[0035] Repeat the above complete cycle at least three times, recording the discharge capacity calculated using the coulomb integral method during each discharge process, denoted as C1, C2, and C3 respectively. To improve the accuracy of the data, the first charge-discharge operation can be repeated four times, discarding the data obtained the first time and using the discharge capacities C2, C3, and C4 obtained the second to fourth times. The arithmetic mean of the three capacities is taken as the average cell capacity of the battery cell, i.e.: ;in, This represents the average cell capacity of the Nth target battery cell. This embodiment uses this average value to eliminate random errors in a single test, thereby improving the stability and repeatability of the measurement results.

[0036] Step S320: Obtain the health status of the target battery cell based on its initial nominal capacity and average battery cell capacity, and determine the change in health status based on the battery cell health status.

[0037] The health status of a single battery cell can be determined using the formula... Calculated, where, The health status of the Nth target battery cell after experiencing slight thermal abuse can be used to reflect the ratio of its current usable capacity to its factory condition; the change in the health status of each target battery cell can be calculated using the formula... Calculated; where, This refers to the change in the health status of the Nth target battery cell, which can be used to represent the degree of relative capacity loss caused by thermal abuse events.

[0038] Step S220: The target battery cell whose health status change is greater than or equal to the first judgment threshold is designated as the first battery cell of the first risk level.

[0039] The first judgment threshold is a pre-set empirical value, ranging from 0.3 to 0.4, such as 0.35. That is, when a certain battery cell... At that time, it was determined that significant irreversible damage had occurred, posing a high safety risk, and should be directly classified as the highest risk level, no longer participating in subsequent assessment processes. Therefore, it is only necessary to... The corresponding battery cells will then undergo a subsequent evaluation process.

[0040] This embodiment can achieve preliminary screening of target battery cells through the above steps, and quickly eliminate high-risk individuals with severe macroscopic performance degradation due to thermal abuse.

[0041] Step S120: Obtain the self-discharge rate growth rate of each second battery cell (excluding the first battery cell) among all target battery cells, and further filter out the third battery cells that are classified as the first risk level based on the self-discharge rate growth rate.

[0042] In this step, the second battery cell refers to the target battery cells whose health status change is less than the first judgment threshold after the initial screening in step S110. These battery cells do not show significant capacity degradation, but may still have hidden damage such as internal micro-short circuits, SEI film rupture, or electrolyte decomposition, leading to abnormally aggravated self-discharge behavior. Therefore, their dynamic electrochemical characteristics need to be further evaluated to identify potentially high-risk individuals.

[0043] In some implementations, such as Figure 4 As shown, step S120 includes steps S410-S420: Step S410: Perform a second charge-discharge operation and a resting operation on each second battery cell to obtain the self-discharge rate growth rate of each second battery cell relative to the initial self-discharge rate.

[0044] The initial self-discharge rate is the self-discharge rate of the battery cell at the time of manufacture.

[0045] In this step, the second charge-discharge operation is used to restore each of the second battery cells to a stable fully charged state, ensuring the consistency and comparability of measurement results during the subsequent resting process. This process includes at least one complete discharge, one charge cycle, and one long resting operation after full charge.

[0046] Specifically, each second battery cell is first charged using a constant current and constant voltage method at a charging rate of 0.33C until fully charged. The cutoff voltage for constant current charging is 0.05C. After full charge, the cells are left to stand for a second preset time, which can be any value between 20 and 60 minutes, such as 30 minutes. Then, a constant current discharge is performed at a discharge rate of 0.33C. For ternary lithium batteries, the discharge cutoff voltage can be 2.8V; for lithium iron phosphate batteries, the discharge cutoff voltage can be 2.5V. After discharge, the cells are left to stand for a second preset time to allow them to reach an equilibrium state. Then, a constant current and constant voltage method is used: the cells are charged at a constant current of 0.33C until fully charged. Charging stops when the current decreases to 0.05C. After charging, the cells are left to stand for another second preset time. Finally, a resting period is initiated, placing each charged second battery cell in a constant temperature environment for a third preset time, which can be 24 hours. During this period, the static voltage of each second battery cell is continuously collected at a sampling frequency of no less than once every 10 minutes to ensure data resolution.

[0047] Exemplary, such as Figure 5 As shown, obtaining the growth rate of the self-discharge rate of each second battery cell relative to the initial self-discharge rate includes steps S510-S520: Step S510: Obtain the settling voltage of each second battery cell at the start and end times within a preset time before the end of the settling operation, and calculate the self-discharge rate of each second battery cell based on the obtained settling voltage at the start and end times.

[0048] Since the first 12 hours of the resting operation may experience unsteady voltage decay due to charging polarization, and the voltage change in the last 12 hours better reflects the actual self-discharge behavior, the preset time before the end of the resting operation can be the last 12 hours of the resting operation process. For the Nth second battery cell, the resting voltage at the start time is obtained. and the resting voltage at the end When determining the self-discharge rate, the formula for calculating the self-discharge rate can be used. Calculated, where, Let be the self-discharge rate of the Nth battery.

[0049] Step S520: Based on the current self-discharge rate of each second battery cell, the initial self-discharge rate at the time of manufacture, and the adjustment coefficient, determine the self-discharge rate growth rate of each second battery cell.

[0050] The self-discharge rate growth rate of each second battery cell in this step It can be done through formula The calculated self-discharge rate growth rate represents the relative acceleration of self-discharge behavior after excluding normal aging factors. Here, the initial self-discharge rate at the time of manufacture is included. This refers to the baseline value measured by the same method for each battery cell during the production stage or initial use. If the original data is missing, a typical value of the same model battery can be used instead. An adjustment coefficient is introduced in this step. It is used to correct the natural increase in self-discharge rate caused by normal aging (such as the increase in the number of cumulative cycles, the influence of long-term operating temperature, etc.) and avoid misjudgment.

[0051] Adjustment coefficient The adjustment coefficient can be dynamically adjusted based on the cumulative number of cycles, capacity decay, and ambient temperature of each second battery cell. Specifically, the adjustment coefficient is positively correlated with the cumulative number of cycles, capacity decay, and ambient temperature. The specific value of the adjustment coefficient can be determined based on experience or obtained by looking up a table.

[0052] Step S420: The battery cells whose self-discharge rate growth rate is greater than or equal to the second judgment threshold are designated as the third battery cells of the first risk level.

[0053] The second judgment threshold can be a pre-set empirical value, with a range of 20% to 40%, such as 30%. When a certain second battery cell... At a certain percentage, it was considered that significant side reactions or micro-short circuit risks had occurred internally. Although no significant capacity loss had occurred, these cells posed a high safety hazard. Therefore, these cells were classified as the third-risk cells of the first level and were excluded from subsequent evaluations. Only those cells with a risk level of 90% were considered. The second battery cell will be evaluated subsequently.

[0054] The judgment process in step S120 can achieve in-depth screening of edge cells with normal surface conditions of individual battery cells, thereby identifying early degradation phenomena that are difficult to detect by capacity testing alone, thus improving the sensitivity and reliability of the overall evaluation system.

[0055] Step S130: Obtain the multidimensional response parameters of each fourth battery cell (excluding the third battery cell) when performing the second charge-discharge operation, and combine them with the risk assessment model to obtain the risk level of each fourth battery cell.

[0056] The fourth category of battery cells refers to those that survived the first two rounds of screening, neither classified as the first-risk level due to significant capacity decay nor as belonging to that level due to abnormally high self-discharge rates. These cells exhibit normal macroscopic performance and basic electrochemical behavior, but may still show subtle differences in dynamic response, requiring more refined classification through higher-dimensional behavioral characteristic analysis. This step aims to comprehensively evaluate this group and identify battery cells with relatively higher risks.

[0057] The first risk level is the highest, corresponding to battery cells with significant safety hazards that should no longer be used. The fourth risk level is further divided into second and third risk levels to guide whether it can be used in applications with relatively lower safety requirements, such as energy storage. The second risk level is higher than the third risk level. It can be understood that if the first risk level is high-risk, then the second and third risk levels correspond to medium and low risk, respectively.

[0058] In some implementations, such as Figure 6 As shown, step S130 includes steps S610-S630: Step S610: Extract the voltage standard deviation of each fourth battery cell during the constant current charging stage, the duration of the constant voltage charging stage, the temperature change characteristics during the charging operation of the second charge and discharge operation, and the voltage change obtained after the resting operation, as multidimensional response parameters.

[0059] Specifically, for each fourth battery cell, high-frequency voltage, current, and temperature data are simultaneously collected during its second charge-discharge operation. Based on this raw data, the temperature change characteristics during charging and the voltage change obtained after the resting operation are calculated as multi-dimensional response parameters.

[0060] First, extract the voltage standard deviation of the second constant current charging stage in the second charge-discharge operation. This parameter reflects the fluctuation of the terminal voltage of a single battery cell during constant current charging and is a key indicator for measuring the uniformity and stability of its internal reaction. The formula for calculating the voltage standard deviation is: ;in, Let be the voltage value measured at time t. The average voltage during the entire constant current phase is denoted by , and n is the total number of sampling points. If the voltage fluctuation of a certain battery cell is significantly higher than that of other cells in the same group, it may indicate problems such as uneven internal resistance, electrode coating defects, or microstructural damage.

[0061] Secondly, the duration of the constant voltage charging phase is extracted, and the proportion of constant voltage duration is calculated by combining it with the duration of the constant current phase. The calculation formula is as follows: ;in, This refers to the duration after the Nth battery cell enters the constant voltage stage. This refers to the time required for constant current charging. The larger this ratio, the more significant the decrease in the charge acceptance capacity of the battery cell. This is usually related to SEI film thickening, loss of active lithium, or pore blockage, and is an important sign of early aging.

[0062] Next, it is necessary to extract the temperature change characteristics during the charging process. This includes, but is not limited to, the standard deviation of the temperature during the charging process. The highest temperature rises and maximum temperature difference .in, This indicates temperature fluctuations during the constant current phase, used to assess the consistency of the thermal management system. Defined as the difference between the highest temperature recorded during charging and the initial temperature, it reflects the overall heat generation level; It can refer to the temperature difference between different locations on the surface of a single battery cell, or the temperature difference between it and other adjacent battery cells, and is used to identify local hot spots and prevent the spread of thermal runaway.

[0063] In addition, it is also necessary to extract the voltage change obtained after the resting operation. This parameter can be obtained based on the open-circuit voltage relaxation behavior after full charging and resting for 30 minutes. Record the voltage values ​​at the initial time (t=0) and at the 30th minute, and calculate the difference. ;in, This is the voltage value at the end of the rest period after the second charging operation. This represents the voltage value at the initial setting. The OCV of a new battery cell stabilizes relatively quickly after it is fully charged. The voltage drop is relatively small; however, battery cells that have experienced slight thermal abuse will show a more obvious slow voltage drop due to interface instability or continuous side reactions, so this parameter is highly sensitive to potential degradation.

[0064] Step S620: Input the multidimensional response parameters into the risk assessment model to obtain the risk score of each fourth battery cell.

[0065] The risk assessment model is a pre-trained machine learning model that can output a quantitative score based on the input multidimensional response parameters to reflect the risk level of a single battery cell.

[0066] In some implementations, such as Figure 7 As shown, the construction of the risk assessment model includes steps S710-S730: Step S710: Obtain multiple training sample battery cells with known risk states, perform the second charge / discharge operation, and collect the corresponding multidimensional response parameters as input features.

[0067] The known risk status serves as the classification label.

[0068] The training sample battery cells can be derived from historical experimental data or retired batteries from actual operation, covering typical individuals with different health states and failure modes. Each sample battery cell is operated according to the same process as the battery cell under test, and its voltage, temperature, and other signals are collected simultaneously. The aforementioned multidimensional response parameters are extracted to form the input feature vector. At the same time, assign category labels to them. The value can be either 0 or 1. Label 1 corresponds to a battery cell confirmed to be high-risk, such as the individual identified as the third battery cell in step S120; label 0 corresponds to a low-risk individual that has been operating stably for a long time and has no abnormal records. This constructs a system consisting of several (… The training dataset consists of .

[0069] Step S720: The random forest algorithm is used to model the training dataset consisting of input features and classification labels, and the input features with importance scores higher than the preset importance threshold are selected as key features by the importance ranking method.

[0070] In this step, we can first standardize the features in the training dataset, for example, by using the min-max scaling method. This method aims to eliminate the influence of different physical dimensions.

[0071] Next, set the parameters for the random forest model, such as setting the number of decision trees B=100 and the maximum depth of each tree. =10, minimum number of leaf node samples =5. The Bootstrap method is used to extract multiple (e.g., 100) subsets of the training set with replacement, each subset being approximately 63.2% the size of the original set. At each node split, three features are randomly selected to participate in the selection of the optimal splitting variables, introducing randomness to reduce the risk of overfitting.

[0072] For each node, Gini impurity can be used as the splitting criterion, i.e. ;in, This represents the proportion of samples of class k in the current node. Recursively split until the maximum depth is reached, the Gini value is below 0.01, or the number of samples in the node is less than [a certain value]. This process is repeated 100 times to generate 100 independent decision trees, which are then used to form a random forest model. The leaf nodes of each tree output the probability of the risk category.

[0073] After model training, the contribution of each input feature is evaluated using a ranking importance method. Specifically, baseline accuracy is calculated using full prediction on the test set. Then, the values ​​of each feature are randomly shuffled (while keeping other features unchanged), and predictions are re-performed and the new accuracy is calculated. The difference between the two is the importance score of that feature, i.e. If the model performance drops significantly after one of the features is shuffled, it indicates that the feature has strong discriminative ability and should be retained.

[0074] Furthermore, all importance scores can be normalized, and the normalization process is as follows: After sorting the normalized importance scores in descending order, a preset importance threshold is set. Only retain The features are used as key features. Among them, The value range is from 0.05 to 0.15, and can be 0.1.

[0075] Step S730: After normalizing the importance scores of the key features, use them as the weight coefficients of the corresponding key features in the risk score to obtain a risk assessment model composed of key features and corresponding weight coefficients.

[0076] Specifically, assuming m key features are selected, their normalized importance scores are respectively , ··· These importance scores are then further standardized into weighting coefficients. ,Right now It is understandable that this weight vector, together with the key features, can form the final risk assessment model for online prediction.

[0077] When a new sample is input, the model can use the formula Calculate its risk score The higher the score, the closer the battery cell is to a high-risk group. is the normalized input value for the i-th key feature.

[0078] Step S630: Based on the risk score, the risk of each fourth battery cell is ranked, and the risk level of each fourth battery cell is determined based on the ranking result.

[0079] In this step, the risk scores of all fourth-generation battery cells are sorted in ascending order to generate an ordered sequence. Then, cells at the bottom 95 percentile are designated as medium-risk; the remaining cells are classified as low-risk. For example, if there are 100 fourth-generation battery cells, the five with the highest risk scores are classified as medium-risk, and it is recommended to monitor them closely or downgrade their use; the remaining 95 are considered low-risk and can be used for secondary utilization under safe conditions. This embodiment uses a percentile classification method, which is not limited by absolute thresholds and can dynamically adjust the warning line according to the overall status of the current batch, avoiding the influence of individual extreme values ​​on the overall judgment.

[0080] This embodiment constructs a multi-level battery cell safety risk assessment system, enabling accurate and non-destructive identification of the safety status of battery cells with intact surrounding structures after a minor thermal abuse event. This method utilizes multi-dimensional characteristics of the battery during charging and discharging, such as capacity decay, self-discharge behavior evolution, and dynamic voltage and temperature response, combined with machine learning models to achieve refined risk stratification. This effectively avoids the omissions and misjudgments caused by existing assessment methods that rely on macroscopic parameters or destructive testing. The entire assessment process does not require battery disassembly and will not cause any permanent damage to the battery, demonstrating good engineering feasibility. Furthermore, this embodiment, through step-by-step screening of key indicators such as changes in health status and self-discharge rate growth rate, can not only quickly eliminate high-risk individuals but also further identify potential marginal abnormal cells within the low-risk group, thereby improving the safety and economy of battery system reuse.

[0081] Figure 8 A schematic diagram of a risk assessment device according to an embodiment of this application is shown. Exemplarily, the risk assessment device includes: The initial screening module 100 is used to obtain the change in health status of each target battery cell through the first charge and discharge operation, and to screen out the first battery cell with the first risk level from all target battery cells based on the change in health status; the target battery cell is the battery cell other than the valve-opening battery cell.

[0082] The rescreening module 200 is used to obtain the self-discharge rate growth rate of each second battery cell other than the first battery cell in all target battery cells, and to screen out the third battery cells with the first risk level based on the self-discharge rate growth rate.

[0083] The risk determination module 300 is used to obtain the multi-dimensional response parameters of each of the fourth battery cells (excluding the third battery cell) during the second charge-discharge operation, and combine them with the risk assessment model to obtain the risk level of each fourth battery cell. Among them, the first risk level has the highest risk.

[0084] It is understood that the apparatus in this embodiment corresponds to the risk assessment method in the above embodiments, and the options in the above embodiments are also applicable to this embodiment, so they will not be described again here.

[0085] This application also provides a terminal device, exemplary of which includes a processor and a memory, wherein the memory stores a computer program, and the processor executes the computer program to enable the terminal device to perform the functions of the various modules in the above-described risk assessment method or risk assessment device.

[0086] The processor can be an integrated circuit chip with signal processing capabilities. The processor can be a general-purpose processor, including at least one of a Central Processing Unit (CPU), Graphics Processing Unit (GPU), Network Processor (NP), Digital Signal Processor (DSP), Application-Specific Integrated Circuit (ASIC), Field-Programmable Gate Array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. The general-purpose processor can be a microprocessor or any conventional processor, capable of implementing or executing the methods, steps, and logic block diagrams disclosed in the embodiments of this application.

[0087] The memory can be, but is not limited to, Random Access Memory (RAM), Read Only Memory (ROM), Programmable Read-Only Memory (PROM), Erasable Programmable Read-Only Memory (EPROM), Electrically Erasable Programmable Read-Only Memory (EEPROM), etc. The memory is used to store computer programs, and the processor can execute the computer programs accordingly after receiving execution instructions.

[0088] This application also provides a computer-readable storage medium for storing the computer program used in the aforementioned terminal device. For example, the computer-readable storage medium may include, but is not limited to, various media capable of storing program code, such as a USB flash drive, a portable hard drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0089] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can also be implemented in other ways. The apparatus embodiments described above are merely illustrative. For example, the flowcharts and block diagrams in the accompanying drawings show the architecture, functionality, and operation of possible implementations of apparatus, methods, and computer program products according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that, in alternative implementations, the functions marked in the blocks may occur in a different order than those marked in the drawings. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagram and / or flowchart, and combinations of blocks in the block diagram and / or flowchart, can be implemented using a dedicated hardware-based system that performs the specified function or action, or using a combination of dedicated hardware and computer instructions.

[0090] In addition, the functional modules or units in the various embodiments of this application can be integrated together to form an independent part, or each module can exist independently, or two or more modules can be integrated to form an independent part.

[0091] If the aforementioned functions are implemented as software functional modules and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a smartphone, personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application.

[0092] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application.

Claims

1. A risk assessment method, characterized in that, The method for assessing the risk of a single battery cell includes: The health status change of each target battery cell is obtained through the first charge and discharge operation, and the first battery cell with the first risk level is selected from all the target battery cells based on the health status change; the target battery cell is the battery cell other than the valve-opening battery cell. Obtain the self-discharge rate growth rate of each second battery cell other than the first battery cell in all the target battery cells, and further filter out the third battery cells that are at the first risk level based on the self-discharge rate growth rate. The multidimensional response parameters of each fourth battery cell (excluding the third battery cell) in the second battery cell are obtained when performing the second charge-discharge operation, and combined with the risk assessment model, the risk level of each fourth battery cell is obtained; among them, the first risk level has the highest risk.

2. The risk assessment method according to claim 1, characterized in that, The step of acquiring the change in health status of each target battery cell through the first charge-discharge operation, and selecting the first battery cell of the first risk level from all the target battery cells based on the change in health status, includes: A first charge-discharge operation is performed on each of the target battery cells to determine the change in the health status of the corresponding target battery cell based on the capacity change of each target battery cell after the first charge-discharge operation. The target battery cell whose health status change is greater than or equal to the first judgment threshold is designated as the first battery cell of the first risk level.

3. The risk assessment method according to claim 2, characterized in that, Before performing the first charge and discharge operation on each of the target battery cells, the method further includes: obtaining the initial nominal capacity of each of the target battery cells; The step of performing a first charge-discharge operation on each of the target battery cells to determine the corresponding change in health status based on the capacity change of each target battery cell after the first charge-discharge operation includes: Perform at least two first charge-discharge operations on each of the target battery cells, and obtain the average battery cell capacity based on the battery cell capacity after each first charge-discharge operation. The health status of a battery cell is obtained based on the initial nominal capacity and the average battery cell capacity of the target battery cell, and the change in health status is determined based on the health status of the battery cell.

4. The risk assessment method according to claim 1, characterized in that, The step of obtaining the self-discharge rate growth rate of each of the second battery cells (excluding the first battery cell) among all the target battery cells, and then further filtering out the third battery cells that meet the first risk level based on the self-discharge rate growth rate, includes: Perform a second charge-discharge operation and a resting operation on each of the second battery cells to obtain the self-discharge rate growth rate of each of the second battery cells relative to the initial self-discharge rate. The battery cells whose self-discharge rate growth rate is greater than or equal to the second judgment threshold are designated as the third battery cells of the first risk level.

5. The risk assessment method according to claim 4, characterized in that, The step of obtaining the self-discharge rate growth rate of each of the second battery cells relative to the initial self-discharge rate includes: The self-discharge rate of each second battery cell is calculated based on the obtained static voltage at the start and end times within a preset time before the static operation ends. The self-discharge rate growth rate of each second battery cell is determined based on the current self-discharge rate of each second battery cell, the initial self-discharge rate at the time of manufacture, and the adjustment coefficient.

6. The risk assessment method according to claim 4, characterized in that, The charging operation in the second charge-discharge operation is a constant current constant voltage charging operation; the constant current constant voltage charging operation includes a constant current charging stage and a constant voltage charging stage; The process of obtaining multidimensional response parameters of each of the fourth battery cells (excluding the third battery cell) during the second charge-discharge operation, and combining these parameters with a risk assessment model to obtain the risk level of each of the fourth battery cells, includes: The voltage standard deviation of each of the fourth battery cells during the constant current charging phase, the duration of the constant voltage charging phase, the temperature change characteristics during the charging operation of the second charge and discharge operation, and the voltage change obtained after the resting operation are extracted as the multidimensional response parameters. The multidimensional response parameters are input into the risk assessment model to obtain the risk score of each of the fourth battery cells; The risk of each of the fourth battery cells is ranked based on the risk score, and the risk level of each of the fourth battery cells is determined based on the ranking result; wherein, the risk level of the fourth battery cell includes a second risk level and a third risk level; the risk of the second risk level is higher than the risk of the third risk level.

7. The risk assessment method according to claim 6, characterized in that, The construction of the risk assessment model includes: Multiple training sample battery cells with known risk states are acquired, the second charge-discharge operation is performed, and the corresponding multidimensional response parameters are collected as input features; wherein, the known risk states are used as classification labels. The random forest algorithm is used to model the training dataset consisting of the input features and the classification labels, and the input features with importance scores higher than a preset importance threshold are selected as key features by the importance ranking method. After normalizing the importance scores of the key features, the scores are used as the weight coefficients of the corresponding key features in the risk score, so as to obtain a risk assessment model composed of the key features and their corresponding weight coefficients.

8. The risk assessment method according to claim 1, characterized in that, The target battery cell is a battery cell within a preset threshold range extending outward from the valve-opening battery cell.

9. The risk assessment method according to claim 5, characterized in that, The adjustment coefficient is dynamically adjusted based on the cumulative number of cycles, capacity decay, and ambient temperature of each of the second battery cells.

10. A risk assessment device, characterized in that, The device for assessing the risk of individual battery cells includes: The initial screening module is used to obtain the change in health status of each target battery cell through the first charge and discharge operation, and to screen out the first battery cell with the first risk level from all the target battery cells based on the change in health status; the target battery cell is a battery cell other than the valve-opening battery cell. The re-screening module is used to obtain the self-discharge rate growth rate of each second battery cell other than the first battery cell in all the target battery cells, and to screen out the third battery cells that are at the first risk level based on the self-discharge rate growth rate. The risk determination module is used to obtain the multi-dimensional response parameters of each fourth battery cell (excluding the third battery cell) in all the second battery cells when performing the second charge and discharge operation, and combine them with the risk assessment model to obtain the risk level of each fourth battery cell; wherein, the first risk level has the highest risk.

11. A terminal device, characterized in that, The terminal device includes a processor and a memory, the memory storing a computer program, and the processor executing the computer program to implement the risk assessment method according to any one of claims 1-9.

12. A computer-readable storage medium, characterized in that, It stores a computer program that, when executed on a processor, implements the risk assessment method according to any one of claims 1-9.

Images