Lithium battery health state assessment method, equipment and medium

Abstract

The invention relates to a lithium battery health state assessment method, which comprises the following steps: S1, selecting a sample data set in a database, and extracting factor data of health factors from the sample data set; s2, calculating a correlation coefficient between the factor data of the health factor and the health state; s3, comparing the correlation coefficient with a preset coefficient threshold value, and judging whether the health factor is a target health factor or not; s4, constructing a linear regression model taking the target health factor as input and the health state as output; and S5, extracting factor data of each target health factor of the lithium battery to be evaluated, inputting the factor data of each target health factor into the linear regression model, and outputting the health state of the lithium battery. The invention also relates to a device and a medium. The target health factors have strong correlation with the SOH, and the accuracy is evaluated by a linear regression model taking the target health factors as input and the health state as output.

Description

Technical Field

[0001] The present application relates to the field of battery technology, and in particular to a method, device and medium for evaluating the health status of a lithium battery. Background Art

[0002] The state of health (SOH) reflects the ability of a battery to store and supply energy relative to its initial conditions. It is usually determined based on the reduction in battery capacity or the increase in internal resistance. When the battery capacity drops to 80% of the initial value, or the internal resistance becomes 1.3 times its initial value, the fault threshold is reached. Based on this, the evaluation and prediction of lithium battery SOH helps to understand the performance changes of the battery, detect potential faults in a timely manner, and take measures to extend the service life of the battery and ensure the safe and reliable operation of new energy storage and power grids. With the large-scale application of lithium-ion batteries in new energy vehicles and power storage, higher requirements are placed on their SOH evaluation and prediction technologies.

[0003] At present, in actual operation, the capacity measurement requires a full charge / discharge cycle, the internal resistance measurement requires external equipment, and in the early stage of battery degradation, the performance degradation of the direct health indicator (DHI) is not obvious, so it is necessary to extract various indirect health factors (Refined Health Indicator, RHI) from voltage, current, temperature and other data. There are usually three main methods for identifying the health status of lithium batteries, based on voltage differential / capacity increment analysis, based on monitoring parameter changes, and based on monomer inconsistency.

[0004] However, the above methods all require traversing the charge and discharge curves or ensuring that the appropriate data segments can be selected under certain working conditions, without considering the extraction of health factors within the actual operating state of charge range of the energy storage power station. Therefore, the extracted health factors may have certain deviations from the data generated by the actual operation, resulting in inaccurate or even missing health factors, which affects the actual use effect. In addition, due to the inaccuracy of the extracted health factors, the correlation between each health factor and the health status is relatively weak, so when evaluating the health status of lithium batteries, the evaluation accuracy is low. Summary of the invention

[0005] In order to solve the above technical problems, this application provides the following technical solutions:

[0006] On the one hand, the present application provides a method for evaluating the health status of a lithium battery, the steps of which include:

[0007] Step S1: Select a sample data set in the database, and extract factor data of health factors from the sample data set;

[0008] Step S2: Calculate the correlation coefficient between the factor data of the health factor and the health status;

[0009] Step S3: Determine whether the health factor is a target health factor based on the comparison between the correlation coefficient and the preset coefficient threshold;

[0010] Step S4: construct a linear regression model with the target health factor as input and the health status as output;

[0011] Step S5: extracting the factor data of each target health factor of the lithium battery to be evaluated, inputting the factor data of each target health factor into a linear regression model, and outputting the health status of the lithium battery.

[0012] In one embodiment, selecting a sample data set in a database includes:

[0013] From the database, the changes in voltage, current, and temperature within a certain range of SOC are selected as sample data sets;

[0014] Among them, the effective charging section and the effective discharging section are screened as follows:

[0015]

[0016] In the formula, SOC(i, t) represents the state of charge of battery i at time t, and I(i, t) represents the current of battery i at time t;

[0017] Select below the lower bound lower Or above the upper bound upper The data point is taken as an outlier. After removing the outlier, the average value y of the two points before and after the outlier is used. t Fill and calculate as follows:

[0018]

[0019] In the formula, x i，t Indicates the data that battery i needs to be cleaned at time t. Arrange the data in ascending order. Q1 is the data point at 25%, Q3 is the data point at 75%, and y t Indicates the data replaced by this point after removing outliers.

[0020] In one embodiment, extracting factor data of health factors from a sample data set includes:

[0021] From the sample data set, the factor values ​​of direct health factors are extracted. The direct health factors include the internal resistance, maximum temperature, average temperature, minimum temperature, charging time, and temperature range of the battery.

[0022] In one embodiment, extracting factor data of health factors from the sample data set also includes:

[0023] From the sample data set, extract the temperature change curve and voltage change curve under each cycle.

[0024] In one embodiment, extracting factor data of health factors from the sample data set also includes:

[0025] From the sample data set, the factor data of the indirect health factors of equal discharge voltage interval time difference, equal charge voltage interval time difference, equal voltage interval discharge capacity difference and equal charge voltage interval capacity difference in the constant current charge and discharge working stage are taken.

[0026] In one embodiment, calculating the correlation coefficient between the factor data of the health factor and the health status includes:

[0027] The health factor in the sample data set has n variables, and the Spearman correlation coefficient corresponding to each variable is calculated to obtain the correlation coefficient between the factor data of the health factor and the health status.

[0028] In one embodiment, judging whether the health factor is a target health factor based on comparing the correlation coefficient with a preset coefficient threshold includes:

[0029] When the absolute value of the correlation coefficient of the health factor is greater than the preset coefficient threshold of 0.95, the health factor is the target health factor.

[0030] In one embodiment, the linear regression model formula is:

[0031] y=Xw+b

[0032] In the formula, w is the weight coefficient vector, and b is the bias term;

[0033] The loss function of the linear regression model is:

[0034]

[0035] The formula for derivation of the loss function is:

[0036]

[0037] w=(X T X+αI) -1 X T y

[0038] In the formula, α is the regularization coefficient.

[0039] On the other hand, the present application also provides an electronic device, including a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the steps of the lithium battery health status assessment method provided in the above embodiment are implemented.

[0040] On the other hand, the present application also provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the steps of the lithium battery health status assessment method provided in the above embodiment are implemented.

[0041] This application has at least the following beneficial effects:

[0042] In this application, through direct measurement of known batteries and SOH values, due to the existence of multiple linear relationships between health factors and SOH, the extracted target health factors all have a strong correlation with SOH. The linear regression model with target health factors as input and health status as output can improve the accuracy of the model in evaluating the health status of lithium batteries. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 A flowchart of a lithium battery health status assessment method provided in one embodiment of the present application.

[0044] Figure 2 A schematic diagram of a capacity increment curve provided in an embodiment of the present application.

[0045] Figure 3 A schematic diagram of an IC curve and a curve x=-1 provided in an embodiment of the present application. DETAILED DESCRIPTION

[0046] In order to make the above-mentioned purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are described in detail below in conjunction with the accompanying drawings. In the following description, many specific details are set forth to facilitate a full understanding of the present application. However, the present application can be implemented in many other ways different from those described herein, and those skilled in the art can make similar improvements without violating the connotation of the present application, so the present application is not limited by the specific embodiments disclosed below.

[0047] In the description of the present application, it should be understood that if the terms "length", "width", "thickness", "up", "down", "vertical", "inside", "outside", "clockwise", "counterclockwise", "axial", "radial", "circumferential", etc. appear, the orientation or position relationship indicated by these terms is based on the orientation or position relationship shown in the drawings, which is only for the convenience of describing the present application and simplifying the description, and does not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the present application.

[0048] In addition, if the terms "first", "second", and "third" appear, these terms are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Therefore, a feature defined as "first", "second", and "third" may explicitly or implicitly include at least one of the features. In the description of this application, if the term "plurality" appears, the meaning of "plurality" is at least two, such as two, three, etc., unless otherwise clearly and specifically defined.

[0049] In this application, unless otherwise clearly specified and limited, if the terms "installed", "connected", "connected", "fixed" and the like appear, these terms should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integrated connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium, it can be the internal connection of two elements or the interaction relationship between two elements, unless otherwise clearly defined. For ordinary technicians in this field, the specific meanings of the above terms in this application can be understood according to the specific circumstances.

[0050] In the present application, unless otherwise clearly specified and limited, if there is a description that a first feature is "above" or "below" a second feature, etc., or similar descriptions appear, it may mean that the first and second features are in direct contact, or that the first and second features are in indirect contact through an intermediate medium. Moreover, the first feature being "above", "above" and "above" the second feature may mean that the first feature is directly above or obliquely above the second feature, or simply means that the first feature is higher in level than the second feature. The first feature being "below", "below" and "below" the second feature may mean that the first feature is directly below or obliquely below the second feature, or simply means that the first feature is lower in level than the second feature.

[0051] It should be noted that if an element is referred to as being "fixed to" or "disposed on" or "set on" another element, it may be directly on the other element or there may be a central element. If an element is considered to be "connected to" another element, it may be directly connected to the other element or there may be a central element at the same time. If any, the terms "vertical", "horizontal", "upper", "lower", "left", "right" and similar expressions used in this application are for illustrative purposes only and do not represent the only implementation method.

[0052] The embodiments of the present application are described in detail below with reference to the accompanying drawings.

[0053] Reference Figure 1 As shown, in some embodiments of the present application, the present application provides a method for evaluating the health status of a lithium battery, the steps of which include:

[0054] Step S1: Determine the selected sample data set from the sample data set selected from the database, analyze the sample data set, and extract the factor value corresponding to the preset health factor.

[0055] Wherein, step S1 specifically includes:

[0056] Step S11: Batteries with more than 1000 cycles and 5 different charge and discharge strategies in the public battery data set in the database are selected as samples, and the state of charge ranges from 20% to 80% in the daily operation and maintenance of lithium battery energy storage power stations.

[0057] The database may be a Stanford-MIT dataset. Since the sample types and sample types of the Stanford-MIT dataset are relatively comprehensive, the dataset can not only obtain the capacity attenuation information of the battery during long-term operation, but also obtain the measured data during the charging / discharging process. The sample dataset extracted from the Stanford-MIT dataset can meet the needs of lithium batteries under various charging conditions, thereby improving the diversity of the dataset, and further improving the training accuracy of the lithium battery health status assessment model.

[0058] Step S12: In the database, the changes in voltage, current and temperature within the range of 20%-80% of the SOC are selected as a sample data set.

[0059] Due to human interference and equipment failure, there are outliers in the battery data. In this scheme, the interquartile range (IQR) method is used to detect outliers, and the value below the lower bound is selected. lower Or above the upper bound upper The data point is taken as an outlier. After removing the outlier, the average value y of the two points before and after the outlier is used. t Fill it.

[0060] Among them, it is necessary to extract the required HI within the appropriate charging and discharging data range, and the effective charging segment and the effective discharging segment are screened as follows:

[0061]

[0062] Where SOC(i, t) represents the state of charge of battery i at time t, and I(i, t) represents the current of battery i at time t.

[0063] The calculation of the interquartile range method is as follows:

[0064]

[0065] In the formula, x i，tIndicates the data that battery i needs to be cleaned at time t. Arrange the data in ascending order. Q1 is the data point at 25%, Q3 is the data point at 75%, and y t Indicates the data replaced by this point after removing outliers.

[0066] Step S13: extracting the factor values ​​of direct health factors from the sample data set, where the direct health factors include the internal resistance, maximum temperature, average temperature, minimum temperature, charging time, and temperature range of the battery.

[0067] It should be noted that during the continuous use of the battery, the internal resistance will gradually increase and the full charging time will gradually decrease. Both high and low temperatures will accelerate the battery life decay and aggravate the battery inconsistency. Therefore, the parameter change characteristics of the battery's internal resistance, maximum temperature, average temperature, minimum temperature, charging time, and temperature extremes can be used to identify the battery degradation state.

[0068] Step S14: extract the temperature change curve and voltage change curve under each cycle from the sample data set, determine the time corresponding to when the temperature reaches the peak value from the temperature change curve, and record it as PeakTemperatureTime, and extract the time corresponding to when the voltage reaches the valley value from the voltage change curve, and record it as CutoffVoltageTime.

[0069] The formula is as follows:

[0070] PeakTemperatureTime cycle，i =t cycle，i , T cycle，i，t >T cycle，i，n

[0071] CutoffVoltageTimecy cle，i =tcy cle，i , U cycle，i，t <U cycle，i，n

[0072] n≠t, n=1,2,3...n

[0073] Where PeakTemperatureTime cycle，i ,CutoffVoltageTime cycle， i represents the time when the temperature of battery i reaches the peak value and the time when the voltage reaches the lower limit cut-off potential in the cycle, t cycle，i represents the sampling time of the cycle of battery i, T cycle，i，t , U cycle，i，t They represent the temperature and voltage of battery i at the sampling time t of the cycle.

[0074] It should be noted that during the charging process, the battery temperature continues to rise due to the Joule heat generated by the current passing through the internal resistance. When the battery is discharging, the battery temperature will also rise due to the Joule heat generated by the current passing through the resistor and the occurrence of exothermic reactions. However, the temperature will gradually drop after the discharge is completed. As the battery ages, the time when the temperature reaches the peak value gradually shortens. The time corresponding to the peak value of the temperature is selected as the indirect health factor.

[0075] In addition, as the number of cycles increases, the capacity of the battery gradually decays, and the time it takes for the battery to reach the lower limit cutoff potential during the charge and discharge process gradually shortens. Therefore, the time corresponding to the voltage reaching the valley value is selected as the indirect health factor.

[0076] Step S15: From the sample data set, the factor values ​​of the indirect health factors of equal discharge voltage interval time difference (TIEDVD), equal charge voltage interval time difference (TIECVD), equal voltage interval discharge capacity difference (EVIDCD) and equal charge voltage interval capacity difference (CDECVI) in the constant current charge and discharge working stage are obtained.

[0077] Among them, the time difference of equal discharge voltage interval and equal charge voltage interval gradually decreases with the increase of battery charge and discharge cycles, and the discharge capacity difference of equal voltage interval and the capacity difference of equal charge voltage interval gradually decrease with the increase of battery charge and discharge cycles. Therefore, TIEDVD, TIECVD, EVIDCD, and CDECVI are selected as indirect health factors.

[0078] The calculation method of TIEDVD, TIECVD, EVIDCD, and CDECVI can be respectively the time difference when the voltage changes from 2.8V to 3.2V when I<0, the time difference when the voltage changes from 3.2V to 3.5V when I>0, the capacity difference when the voltage changes from 2.8V to 3.2V when I<0, and the capacity difference when the voltage changes from 3.2V to 3.5V when I>0. The calculation is shown in the formula:

[0079] TIECVD cycle，i =t u＝3.2，cycle，i -t u＝2.8，cycle，i , I(i, t)<0

[0080] TIECVD cycle，i =t u＝3.5，cycle，i -t u＝3.2，cycle，i , I(i, t)>0

[0081] EVIDCD cycle，i =Q u＝3.2，cycle，i -Q u＝2.8，cycle，i , I(i, t)<0

[0082] CDECVI cycle，i =Q u＝3.5，cycle，i -Qu＝3.2，cycle，i , I(i, t)>0

[0083] The above voltage value range can be adjusted according to actual needs.

[0084] Step S16: The IC curve is the gradient of the charge / discharge capacity relative to the battery voltage, as shown in the formula:

[0085]

[0086] First, extract the IC curve from the sample data set. Then, determine the voltage overlap point at each cycle number, for example, Figure 2 , 3 In the example, the coincidence point is at x = -1, the ordinates of the two intersection points of the IC curve and the curve x = -1 are y1 and y2, and the height and position of the peak of the IC curve are recorded as and The area enclosed by the IC curve and the curve x=-1 is recorded as S. The two intersection points of the IC curve and the curve x=-1 and the peak point of the IC curve are used to calculate the slope on the peak and the slope off the peak of the curve, which are recorded as slope1 and slope2.

[0087] When the curve is x=-1, the formula is as follows. When the curve x changes, just change the 1 in the formula.

[0088]

[0089] It should be noted that the conversion of the flat battery voltage platform into the peak of the IC curve describes the battery phase change characteristics during the intercalation and deintercalation of active materials. The height, area, and position of the peak of the curve change after battery aging. Therefore, the change law of the curve can be analyzed to infer the aging mechanism inside the battery. Selecting the above indirect health factors and selecting the curve X at the overlap point can ensure that the above indirect health factors can be extracted at each cycle number.

[0090] When extracting the health factor, the actual operating state of charge range is taken into account to ensure the accuracy and reliability of the health factor. When considering the actual operating state of charge range, the Stanford-MIT dataset is used for calculation with higher accuracy. Because the sampling interval of the Stanford-MIT dataset is about 1s, the laboratory data is more accurate and the sampling interval is shorter. The charge and discharge data contains more battery health state decay information, so that the voltage, current, temperature and other data within the battery charge and discharge cycle can be fully utilized to extract feature quantities within the daily charge and discharge state of charge range.

[0091] Step S2: Calculate the correlation coefficient between the factor data of the health factor and the health status. Specifically, calculate the Spearman correlation coefficient between the health factor and the health status.

[0092] Wherein, step S2 specifically includes:

[0093] Step S21: The health factor in the sample data set has n variables, and the Spearman correlation coefficient corresponding to each variable is calculated to obtain the correlation coefficient between the factor data of the health factor and the health status.

[0094] For example, suppose there are n variables X and Y in two HI sequences of a battery, sort their data values ​​from small to large, and assign rank R to the i-th variable (Xi) and R (Yi) , the Spearman correlation coefficient corresponding to each pair of variables is calculated as shown in the formula.

[0095]

[0096] The above correlation coefficient calculation formula is used to quantify the correlation between HI and SOH, and the Spearman correlation coefficient is used to evaluate the monotonic relationship between the two variables.

[0097] Step S22: 29 battery samples with a cycle life of more than 1000 times are selected from the database. According to the distribution diagram of the cycle life and charge and discharge strategy of the 29 battery samples, samples with large distribution differences under the same working conditions are extracted and regarded as abnormal samples.

[0098] In this scheme, the experiment is carried out under different working conditions using the same type of battery. Generally, the cycle life of the battery under the same working conditions should be similar. By comparing the cycle life distribution gap under the same working conditions, abnormal samples can be identified.

[0099] Step S23: Calculate the median of the Spearman correlation coefficients between HI and SOH, and determine the median as the correlation coefficient of the health factor.

[0100] For example, for 25 battery samples, since each battery sample has 19 health factors, for each health factor, the median of the correlation coefficients in the 25 battery samples is calculated as the correlation coefficient of each health factor. The correlation analysis results are shown in the following table.

[0101]

[0102] Step S3: Determine whether the health factor is a target health factor by comparing the correlation coefficient with a preset coefficient threshold. Specifically, a health factor whose correlation coefficient is higher than the preset coefficient threshold is determined as a target health factor.

[0103] In a specific embodiment, the preset coefficient threshold is 0.95. More specifically, judging whether the health factor is a target health factor based on the comparison between the correlation coefficient and the preset coefficient threshold includes:

[0104] The HI with the absolute value of the correlation coefficient greater than 0.95 was selected, and the time difference between equal discharge voltage intervals, the discharge capacity difference between equal voltage intervals, the time corresponding to the temperature reaching the peak value, and the height of the IC curve peak were used as target health factors.

[0105] When the battery model changes, the target health factor finally determined may change, that is, the above method can be used to generate a target health factor that meets the characteristics of any battery model.

[0106] Step S4: construct a linear regression model with the target health factor as input and the health status as output.

[0107] Wherein, step S4 specifically includes:

[0108] Step S41: For a given sample B = {(X1, y1), (X2, y2), ... (X n ,y n )}, where x n ∈R m is the target health factor selected above, y i is SOH, m is the data dimension of HI, and n is the battery charging and discharging data dimension. It is expected to establish a linear regression model with HI as input and SOH as output.

[0109] Since each target health factor has a strong correlation with SOH, the target health factors are likely to have multicollinearity. This can be verified by the following formula:

[0110]

[0111] in, refers to the independent variable x i The coefficient of determination of the least squares regression on other independent variables, which indicates the degree to which the independent variable can be linearly predicted by other independent variables, n is the number of samples, is the model prediction for the ith observation, x i is the true dependent variable value of the ith observation, is the dependent variable mean of all observations. After verification by the variance inflation factor indicator, there are indeed multiple linear relationships.

[0112] Since the target health factors are likely to have multicollinear relationships, the ridge regression model is considered to be a regularized linear regression method. By introducing a penalty term to the regression coefficient, it is suitable for dealing with linear regression problems with multicollinearity, high-dimensional features or noisy data, which can improve the stability and generalization ability of the model, and ultimately improve the evaluation accuracy. Therefore, the ridge regression model is used in this solution to construct the evaluation model.

[0113] Step S41: The ridge regression model is shown as follows:

[0114] y=Xw+b

[0115] Where w is the weight coefficient vector of the ridge regression model, and b is the bias term of the ridge regression model.

[0116] Among them, the process of solving the weight coefficient w and the bias term b is transformed into a least squares problem. In order to solve the problem that the output changes greatly when the input x error is very small, the ridge regression model adds a regularization term of the L2 norm of the weight coefficient w multiplied by the regularization y coefficient α to the loss function of the multivariate linear regression problem, as shown in the formula:

[0117]

[0118] The regularization coefficient α is set to 1. When the loss function is minimized, the optimal solution of the weight coefficient w and the bias term b is obtained. The derivative of the loss function is shown in the formula:

[0119]

[0120] w=(X T X+αI) -1 X T y

[0121]

[0122] In another specific embodiment, mean absolute error (MAE) and root mean square error (RMSE) can also be used as evaluation indicators of algorithm performance, as shown in the formula. MAE is the average absolute value of the difference between the predicted value and the true value, which is usually used to evaluate the overall prediction error of the model. RMSE is the root mean square of the difference between the predicted value and the true value, which gives a greater penalty for larger errors and is usually used for performance evaluation of regression problems. The smaller the values ​​of the two indicators, the higher the prediction accuracy.

[0123]

[0124] Since the acquisition of SOH requires full charge and discharge, when the acquisition of SOH is difficult and the amount of data is small, a battery health state estimation model can be established by directly measuring the known battery and the SOH value. The target health factors established in this method are strongly correlated with SOH, so the accuracy of model evaluation can be improved.

[0125] There are solutions to reduce multicollinearity in the prior art. Although the problem of multicollinearity can be reduced, the correlation between HI and SOH is reduced, so the evaluation accuracy will naturally decrease. On the basis that the target health factors are strongly correlated with SOH, various collinearity problems are considered and a ridge regression model is adopted. Ridge regression can provide robust estimates when there is multicollinearity in the processing data, thereby improving the accuracy of health status assessment. In addition, the battery material in the Stanford-MIT battery data set is lithium iron phosphate, indicating that the ridge regression method has good calculation accuracy in different battery materials.

[0126] In the process of predicting the health status of lithium batteries in energy storage power stations, the SOH values ​​and HI of some batteries at historical moments can be used to establish a SOH estimation model using the ridge regression method, so as to obtain the current moment HI of the unknown battery, and then obtain the SOH estimation value of the unknown battery at the current moment according to the mapping relationship. Simulation calculations were performed on battery samples with five different charging and discharging strategies. The calculation results of the battery SOH estimation model test set based on ridge regression had MAE within 1.5% and RMSE within 2%, indicating that this method can estimate the health status of the battery more accurately in the case of small batch data, has good generalization performance, and is suitable for monitoring and estimating the health status of the battery.

[0127] Step S5: extract the factor data of each target health factor of the lithium battery to be evaluated, input the factor data of each target health factor into the linear regression model, and output the health status of the lithium battery. Specifically, extract the factor value of each target health factor of the lithium battery to be evaluated, input the factor value of each target health factor into the ridge regression estimation model, and output the health status of the lithium battery.

[0128] In some embodiments of the present application, the present application also provides an electronic device, including a memory and a processor, the memory stores a computer program, and when the processor executes the computer program, the steps of the lithium battery health status assessment method provided in the above embodiment are implemented.

[0129] Furthermore, the present application also provides a computer-readable storage medium on which a computer management program is stored. When the computer management program is executed by a processor in an electronic device, the steps of the lithium battery health status assessment method provided in the above embodiment are implemented.

[0130] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that include computer-usable program code.

[0131] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0132] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0133] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0134] The foregoing description of specific exemplary embodiments of the present invention is for the purpose of illustration and demonstration. These descriptions are not intended to limit the present invention to the precise form disclosed, and it is clear that many changes and variations can be made based on the above teachings. The purpose of selecting and describing the exemplary embodiments is to explain the specific principles of the present invention and its practical application, so that those skilled in the art can realize and utilize various different exemplary embodiments of the present invention and various different selections and changes. The scope of the present invention is intended to be limited by the claims and their equivalents.

Claims

1. A method for evaluating the health status of a lithium battery, characterized in that: The steps include: Step S1: Select a sample data set in a database, and extract factor data of health factors from the sample data set; Step S2: Calculate the correlation coefficient between the factor data of the health factor and the health status; Step S3: judging whether the health factor is a target health factor according to comparing the correlation coefficient with a preset coefficient threshold; Step S4: constructing a linear regression model with the target health factor as input and the health status as output; Step S5: extracting the factor data of each target health factor of the lithium battery to be evaluated, inputting the factor data of each target health factor into the linear regression model, and outputting the health status of the lithium battery.

2. The lithium battery health status assessment method according to claim 1, characterized in that: The selecting of the sample data set in the database includes: From the database, the changes in voltage, current, and temperature within a certain range of SOC are selected as sample data sets; Among them, the effective charging section and the effective discharging section are screened as follows: In the formula, SOC(i, t) represents the state of charge of battery i at time t, and I(i, t) represents the current of battery t at time t; Select below the lower bound lower Or above the upper bound upper The data point is taken as an outlier. After removing the outlier, the average value y of the two points before and after the outlier is used. t Fill and calculate as follows: In the formula, x i，t Indicates the data that battery i needs to be cleaned at time t. Arrange the data in ascending order. Q1 is the data point at 25%, Q3 is the data point at 75%, and y t Indicates the data replaced by this point after removing outliers.

3. The lithium battery health status assessment method according to claim 2, characterized in that: The step of extracting factor data of health factors from the sample data set includes: From the sample data set, the factor values ​​of direct health factors are extracted. The direct health factors include the internal resistance, maximum temperature, average temperature, minimum temperature, charging time, and temperature range of the battery.

4. The lithium battery health status assessment method according to claim 3, characterized in that: The step of extracting the factor data of the health factor from the sample data set further includes: From the sample data set, extract the temperature change curve and voltage change curve under each cycle.

5. The lithium battery health status assessment method according to claim 3 or 4, characterized in that: The step of extracting the factor data of the health factor from the sample data set further includes: From the sample data set, the factor data of the indirect health factors of equal discharge voltage interval time difference, equal charge voltage interval time difference, equal voltage interval discharge capacity difference and equal charge voltage interval capacity difference in the constant current charge and discharge working stage are taken.

6. The lithium battery health status assessment method according to claim 1, characterized in that: The calculating the correlation coefficient between the factor data of the health factor and the health status comprises: The health factor in the sample data set has n variables, and the Spearman correlation coefficient corresponding to each variable is calculated to obtain the correlation coefficient between the factor data of the health factor and the health status.

7. The lithium battery health status assessment method according to claim 6, characterized in that: The step of comparing the correlation coefficient with a preset coefficient threshold to determine whether the health factor is a target health factor includes: When the absolute value of the correlation coefficient of the health factor is greater than a preset coefficient threshold of 0.95, the health factor is a target health factor.

8. The lithium battery health status assessment method according to claim 1, characterized in that: The linear regression model formula is: y=Xw+b In the formula, w is the weight coefficient vector, and b is the bias term; The loss function of the linear regression model is: The formula for derivation of the loss function is: w=(X T X+αI) -1 X T y In the formula, α is the regularization coefficient.

9. An electronic device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the lithium battery health status assessment method described in any one of claims 1-8 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the lithium battery health status assessment method described in any one of claims 1 to 8 are implemented.

Images