Battery SOH evaluation method, apparatus and device, and storage medium

Abstract

The invention discloses a battery SOH assessment method, device and equipment and a storage medium, and relates to the technical field of battery SOH assessment, and the method comprises the steps: obtaining battery SOH assessment working condition data; according to the SOH evaluation working condition data of the battery, calculating the charging capacity of the battery under each evaluation working condition by adopting an ampere-hour integral method; correcting the initial capacity under each evaluation working condition by adopting a nonlinear regression method to obtain the corrected initial capacity; the initial capacity refers to the battery capacity when the battery leaves a factory; calculating the SOH of the battery under each evaluation working condition according to the corrected initial capacity and the charging capacity of the battery; and fusing all the battery SOHs by adopting a Gaussian process regression method to obtain the comprehensive SOH of the battery. According to the invention, the evaluation precision of the SOH of the battery can be effectively improved.

Description

Technical Field

[0001] The present application relates to the technical field of battery SOH evaluation, and particularly to a battery SOH evaluation method, device, equipment and storage medium. Background Art

[0002] SOH (State Of Health) is an important indicator reflecting battery performance and remaining service life. SOH evaluation is an important means to master the decline state of battery charging capacity. Therefore, it is very necessary to accurately evaluate battery SOH. Battery SOH is usually closely related to external factors such as battery calendar life, cumulative charge and discharge cycle times, battery usage conditions, and internal factors such as the battery's own performance.

[0003] Currently, new energy vehicles can obtain accurate battery charging capacity and battery SOH evaluation only under full charge and full discharge conditions. However, the actual operating conditions of new energy vehicles are complex and diverse, making it difficult to meet the evaluation conditions of full charge and full discharge. Therefore, the problem of inaccurate battery SOH evaluation often occurs. Moreover, when evaluating battery SOH relying on actual operating condition data, affected by factors such as SOC (State Of Charge) error, temperature, and charging rate, there is a problem of low evaluation accuracy of battery SOH. Summary of the Invention

[0004] The purpose of the present application is to provide a battery SOH evaluation method, device, equipment and storage medium, which can effectively improve the evaluation accuracy of battery SOH.

[0005] To achieve the above purpose, the present application provides the following solutions.

[0006] In the first aspect, the present application provides a battery SOH evaluation method, and the battery SOH evaluation method specifically includes the following steps.

[0007] Obtain battery SOH evaluation condition data.

[0008] According to the battery SOH evaluation condition data, use the ampere-hour integration method to calculate the battery charging capacity under each evaluation condition.

[0009] Adopt a non-linear regression method to correct the initial capacity under each evaluation condition to obtain the corrected initial capacity; the initial capacity refers to the battery capacity when the battery leaves the factory.

[0010] According to the corrected initial capacity and the battery charging capacity, calculate the battery SOH under each evaluation condition.

[0011] Adopt the Gaussian process regression method to fuse all the battery SOHs to obtain the battery comprehensive SOH.

[0012] Optionally, obtain the battery SOH evaluation operating condition data, which specifically includes the following steps.

[0013] Obtain the battery charging data; the charging data fields of the battery charging data include a charging status field, a charging current field, an SOC field, a vehicle status field, and a message time field.

[0014] According to the charging status field and the charging current field, perform data screening on the battery charging data to obtain the battery SOH evaluation operating condition data.

[0015] Optionally, according to the charging status field and the charging current field, perform data screening on the battery charging data to obtain the battery SOH evaluation operating condition data, which specifically includes the following steps.

[0016] According to the charging status field, select the battery charging data of the complete charging process from any SOC full charge to 100% SOC from the battery charging data.

[0017] According to the charging current field, select the battery charging data with a charging current greater than the threshold from the battery charging data.

[0018] According to the magnitude of the charging current, divide the battery charging data with a charging current greater than the threshold into multiple constant current charging intervals.

[0019] Take the battery charging data corresponding to each constant current charging interval as a battery SOH evaluation operating condition data segment, and obtain multiple such battery SOH evaluation operating condition data segments.

[0020] Take the multiple battery SOH evaluation operating condition data segments and the battery charging data of the complete charging process as the battery SOH evaluation operating condition data.

[0021] Optionally, the calculation formula for the battery charging capacity under each evaluation operating condition by the ampere-hour integration method is as follows.

[0022] ; Wherein, is the battery charging capacity, is the charging current, is the charging time.

[0023] Optionally, adopt a non-linear regression method to correct the initial capacity under each evaluation operating condition to obtain the corrected initial capacity, which specifically includes the following steps.

[0024] Construct a non-linear regression model; the non-linear regression model refers to the regression model of the initial capacity of the battery varying with the charging rate and temperature.

[0025] With the goal of minimizing the sum of squared residuals, the gradient descent method, Levenberg-Marquardt algorithm, Powell algorithm, BFGS algorithm, or SLSQP algorithm is used to estimate the parameters of the non-linear regression model to correct the initial capacity under each evaluation condition and obtain the corrected initial capacity.

[0026] Optionally, the battery SOH under each evaluation condition is calculated using the following formula.

[0027] ; Where represents the battery health state value, is the battery charging capacity, is the corrected initial capacity, is the remaining charge at the start of charging, is the remaining charge at the end of charging.

[0028] Optionally, the Gaussian process regression method is used to fuse all the battery SOH values to obtain the comprehensive battery SOH, which specifically includes the following steps.

[0029] Based on the Matern kernel function, a Gaussian process regression model is constructed.

[0030] With the negative log marginal likelihood function as the objective function, the hyperparameters of the Gaussian process regression model are solved by minimizing the objective function.

[0031] The quasi-Newton method, genetic algorithm, or particle swarm optimization algorithm is used to train the hyperparameters of the Gaussian process regression model, and the Gaussian process regression model corresponding to the best root mean square error is selected as the final Gaussian process regression model for fusing the battery SOH.

[0032] All the battery SOH values are substituted into the final Gaussian process regression model to obtain the comprehensive battery SOH.

[0033] In a second aspect, the present application provides a battery SOH evaluation device for implementing the battery SOH evaluation method described in the first aspect. The battery SOH evaluation device specifically includes the following modules.

[0034] A battery SOH evaluation condition data acquisition module for acquiring battery SOH evaluation condition data.

[0035] A battery charging capacity calculation module for calculating the battery charging capacity under each evaluation condition using the ampere-hour integration method according to the battery SOH evaluation condition data.

[0036] An initial capacity correction module, which is used to correct the initial capacity under each evaluation condition by using a non - linear regression method to obtain the corrected initial capacity; the initial capacity refers to the battery capacity when the battery leaves the factory.

[0037] A battery SOH calculation module, which is used to calculate the battery SOH under each evaluation condition according to the corrected initial capacity and the battery charging capacity.

[0038] A battery SOH fusion module, which is used to fuse all the battery SOHs by using the Gaussian process regression method to obtain the comprehensive battery SOH.

[0039] In a third aspect, the present application provides a computer device, including: a memory, a processor, and a computer program stored on the memory and executable on the processor, where the processor executes the computer program to implement the battery SOH evaluation method described in the first aspect.

[0040] In a fourth aspect, the present application provides a computer - readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, it implements the battery SOH evaluation method described in the first aspect.

[0041] According to the specific embodiments provided by the present application, the present application has the following technical effects: The present application provides a battery SOH evaluation method, device, equipment, and storage medium, which combines the correction of the initial capacity and the fusion of the battery SOH. First, the initial capacity under each evaluation condition is corrected by using a non - linear regression method, thereby eliminating the influence of factors such as SOC error, temperature, and charging rate on the battery SOH evaluation. By calculating the battery SOH under each evaluation condition with the corrected initial capacity and the battery charging capacity, a more accurate battery SOH can be obtained, improving the evaluation accuracy of the battery SOH. Second, by using the principle of Gaussian process regression fusion, all the battery SOHs are fused by the Gaussian process regression method, so that a more accurate and reliable comprehensive battery SOH can be obtained, effectively improving the evaluation accuracy of the battery SOH. Description of the Drawings

[0042] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required for use in the embodiments. Obviously, the drawings in the following description are only some embodiments of the present application, and for those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0043] Figure 1 It is an application environment diagram of a battery SOH evaluation method provided by an embodiment of the present application.

[0044] Figure 2 Schematic flowchart of a battery SOH evaluation method provided by an embodiment of the present application.

[0045] Figure 3 Schematic diagram of the fitting standard deviation of the Gaussian process regression model provided by an embodiment of the present application.

[0046] Figure 4 Curve graph before and after battery SOH fusion of a new energy vehicle provided by an embodiment of the present application.

[0047] Figure 5 Schematic structural diagram of a battery SOH evaluation device provided by an embodiment of the present application.

[0048] Figure 6 Schematic structural diagram of a computer device provided by an embodiment of the present application. Detailed implementation manners

[0049] Next, the technical solutions in the embodiments of the present application will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present application.

[0050] To make the above objects, features, and advantages of the present application more obvious and understandable, the present application will be further described in detail below in conjunction with the accompanying drawings and specific implementation manners.

[0051] The battery SOH evaluation method provided by the embodiments of the present application can be applied to, for example Figure 1In the application environment shown. Among them, the terminal 102 communicates with the server 104 through the network. The data storage system can store the data that the server 104 needs to process. The data storage system can be set separately, integrated on the server 104, placed on the cloud or other servers. The terminal 102 can send the battery SOH evaluation working condition data to the server 104. After the server 104 receives the battery SOH evaluation working condition data, for the battery SOH evaluation working condition data, the server 104 uses the ampere-hour integration method to calculate the battery charging capacity under each evaluation working condition; uses the non-linear regression method to correct the initial capacity under each evaluation working condition to obtain the corrected initial capacity; the initial capacity refers to the battery capacity when the battery leaves the factory; according to the corrected initial capacity and the battery charging capacity, calculate the battery SOH under each evaluation working condition; use the Gaussian process regression method to fuse all the battery SOHs to obtain the comprehensive battery SOH. The server 104 can feedback the obtained comprehensive battery SOH to the terminal 102. In addition, in some embodiments, the battery SOH evaluation method can also be implemented separately by the server 104 or the terminal 102. For example, the terminal 102 can directly perform the battery SOH evaluation process on the battery SOH evaluation working condition data, or the server 104 can obtain the battery SOH evaluation working condition data from the data storage system and perform the battery SOH evaluation process on the battery SOH evaluation working condition data.

[0052] Among them, the terminal 102 can be, but is not limited to, various desktop computers, laptop computers, smart phones, tablet computers, Internet of Things devices and portable wearable devices. The Internet of Things devices can be smart speakers, smart TVs, smart air conditioners, smart in-vehicle devices, etc. The portable wearable devices can be smart watches, smart bracelets, head-mounted devices, etc. The server 104 can be implemented by an independent server or a server cluster composed of multiple servers, and can also be a cloud server.

[0053] In an exemplary embodiment, as Figure 2 shown, a battery SOH evaluation method is provided. This method is executed by a computer device, and can be specifically executed alone by a computer device such as a terminal or a server, or jointly executed by a terminal and a server. In the embodiments of the present application, taking this method applied to Figure 1 the server 104 in

[0054] Step S1, obtain battery SOH evaluation working condition data.

[0055] In this embodiment, step S1 of obtaining the battery SOH evaluation working condition data specifically includes the following steps.

[0056] Step S11: Obtain battery charging data; the charging data fields of the battery charging data include a charging status field, a charging current field, an SOC field, a vehicle status field, a message time field, etc.

[0057] Step S12: According to the charging status field and the charging current field, perform data screening on the battery charging data to obtain battery SOH evaluation working condition data.

[0058] In this embodiment, step S12 performs data screening on the battery charging data according to the charging status field and the charging current field to obtain battery SOH evaluation working condition data, which specifically includes the following steps.

[0059] Step S121: According to the charging status field, select the battery charging data of a complete charging process from any SOC full charge to 100% SOC from the battery charging data.

[0060] Step S122: According to the charging current field, select the battery charging data with a charging current greater than the threshold from the battery charging data.

[0061] Step S123: Divide the battery charging data with a charging current greater than the threshold into multiple constant current charging intervals according to the magnitude of the charging current.

[0062] Step S124: Take the battery charging data corresponding to each constant current charging interval as a battery SOH evaluation working condition data segment to obtain multiple battery SOH evaluation working condition data segments.

[0063] Step S125: Take multiple battery SOH evaluation working condition data segments and the battery charging data of the complete charging process as battery SOH evaluation working condition data.

[0064] Step S2: According to the battery SOH evaluation working condition data, use the ampere-hour integration method to calculate the battery charging capacity under each evaluation working condition.

[0065] Step S3: Use the nonlinear regression method to correct the initial capacity under each evaluation working condition to obtain the corrected initial capacity; the initial capacity refers to the battery capacity when the battery leaves the factory.

[0066] In this embodiment, step S3 uses the nonlinear regression method to correct the initial capacity under each evaluation working condition to obtain the corrected initial capacity, which specifically includes the following steps.

[0067] Step S31: Construct a nonlinear regression model. Among them, the nonlinear regression model refers to a regression model in which the initial capacity of the battery changes with the charging rate and temperature.

[0068] Step S32: Using optimization algorithms such as the Gradient Descent method, the Levenberg-Marquardt (LM) algorithm, the Powell algorithm, the BFGS (Broyden Fletcher Goldfarb Shanno) algorithm, or the SLSQP (Sequential Least Squares Programming) algorithm with the goal of minimizing the sum of squared residuals, perform parameter estimation on the non-linear regression model to correct the initial capacity under each evaluation condition and obtain the corrected initial capacity.

[0069] Step S4: Calculate the battery SOH under each evaluation condition based on the corrected initial capacity and the battery charging capacity. This battery SOH is a value representing the battery health, which characterizes the health of the battery.

[0070] Step S5: Use the Gaussian process regression method to fuse all the battery SOH values to obtain the comprehensive battery SOH. This comprehensive battery SOH, as the final battery SOH evaluation result, is also a value representing the battery health, which more comprehensively and accurately reflects the health of the battery.

[0071] In this embodiment, Step S5 uses the Gaussian process regression method to fuse all the battery SOH values to obtain the comprehensive battery SOH, which specifically includes the following steps.

[0072] Step S51: Based on the Matern kernel function, construct a Gaussian process regression model.

[0073] Step S52: Using the negative log marginal likelihood function as the objective function, solve the hyperparameters of the Gaussian process regression model by minimizing the objective function.

[0074] Step S53: Use the Quasi-Newton Methods, the Genetic Algorithm (GA), or the Particle Swarm Optimization (PSO) algorithm to train the hyperparameters of the Gaussian process regression model and select the Gaussian process regression model corresponding to the best root mean square error ( ) as the final Gaussian process regression model for fusing the battery SOH.

[0075] Step S54: Substitute all the battery SOH values into the final Gaussian process regression model to obtain the comprehensive battery SOH.

[0076] To make the technical solution of this embodiment clearer, the following will take the form of examples to elaborate on the specific implementation process of the technical solution of this embodiment, including the following implementation steps.

[0077] Step 1: Obtain battery SOH evaluation working condition data.

[0078] In this embodiment, battery charging data is obtained, and battery SOH evaluation working condition data is filtered according to the charge state field and the charge current field.

[0079] In this embodiment, battery SOH evaluation working condition data is first filtered. Select the battery charging data of the complete charging process from any SOC full charge to 100% SOC, denoted as evaluation condition 1. Select the battery charging data with a charging current greater than 0.3C, and divide it into multiple constant current charging intervals according to the current magnitude. The data of each constant current charging interval is denoted as a battery SOH evaluation working condition data segment, denoted as evaluation condition 2. The charging data fields include the charge current field, the SOC field, the vehicle state field, the charge state field, the message time field, etc.

[0080] Step 2: According to the battery SOH evaluation working condition data, use the ampere-hour integration method to calculate the battery charging capacity under each evaluation condition.

[0081] In this embodiment, the ampere-hour integration method is used to calculate the battery charging capacity of all evaluation conditions 1 and evaluation condition 2 respectively , and the calculation formula is as follows.

[0082] .

[0083] Among them, is the battery charging capacity, is the charging current (A), is the charging time (h).

[0084] Step 3: Use the nonlinear regression method to correct the initial capacity under each evaluation condition to obtain the corrected initial capacity.

[0085] In this embodiment, a nonlinear regression model of the initial capacity of the battery at the time of factory with respect to the charging rate and temperature is constructed, and an optimization algorithm is used for parameter estimation, so as to obtain the corresponding initial capacity under each evaluation condition.

[0086] In this embodiment, when correcting the initial capacity under each evaluation condition When considering that the battery charging capacity may fluctuate slightly due to the influence of the charging rate C and temperature T, generally, the higher the charging rate, the smaller the capacity; the higher the charging temperature, the larger the capacity. To make the battery SOH calculation more accurate, the initial capacity data of the battery at different charging rates and temperatures are used, and a regression formula for the battery initial capacity, charging rate, and temperature is obtained by fitting based on the regression algorithm. The regression calculation correction process is as follows.

[0087] (1) Construct a non - linear regression model for the initial capacity of the battery at the time of factory shipment varying with the charging rate and temperature. Non - linear regression models such as the multivariate non - linear polynomial model, multivariate exponential regression model, and multivariate logarithmic regression model can be used to construct the correction regression formula for the initial capacity. of the initial capacity.

[0088] It should be noted that for the correction regression formula of the initial capacity , among the multivariate non - linear polynomial, polynomial with exponent, and polynomial with logarithm, a correction regression formula model with better fitting effect can be selected, and whether the correction regression formula is reasonably designed can be judged through the regression effect.

[0089] (2) Parameter estimation of the non - linear regression model. Parameter estimation is carried out with the goal of minimizing the sum of squared residuals, and algorithms such as the gradient descent method, Levenberg - Marquardt algorithm, Powell algorithm, BFGS algorithm, or SLSQP algorithm are used to solve the parameter values of the non - linear regression model.

[0090] (3) Effect test of the non - linear regression model. The coefficient of determination ( ) and root mean square error are used to evaluate the goodness of fit and error of the non - linear regression model. Among them, the closer it is to 1, the smaller it is, the better the regression effect of the non - linear regression model.

[0091]

[0092] Among them, is the actual output value, is the predicted output value, is the mean of the actual output values.

[0093] (4) According to the charging rate and temperature of the battery SOH evaluation working condition data, calculate the initial capacity under this evaluation working condition.

[0094] In this embodiment, the initial capacity of the battery under the condition of room temperature and 1C charging rate is obtained, the 1C charging current is obtained, and the ratio of the constant - current charging current to the 1C charging current is recorded as the charging rate Statistically calculate the maximum value of the probe temperature within this evaluation condition, and record it as the temperature under this evaluation condition. The charging rate under this evaluation condition , temperature are substituted into the modified regression formula of the non-linear regression model to obtain the modified initial capacity value under this evaluation condition.

[0095] For example, the initial capacity data of a certain battery pack is shown in Table 1.

[0096] Table 1 Initial capacity data of a certain battery pack

[0097] For the initial capacity data of the battery pack in Table 1, the constructed non-linear regression model is a binary non-linear polynomial regression model, and the expression of this binary non-linear polynomial regression model is as follows.

[0098]

[0099] Among them, is the modified initial capacity, a, b, c, d are parameters to be estimated, C is the charging rate, and T is the temperature.

[0100] Then set the objective function for parameter estimation to use , and use the Levenberg-Marquardt algorithm for iterative optimization to obtain the modified regression formula, which is expressed as the following formula.

[0101]

[0102] Then it can be obtained that the of the above non-linear regression model is 0.993, is 0.633, and the fitting effect of the non-linear regression model is good.

[0103] If the current under this evaluation condition is 150 A and the maximum value of the probe temperature during charging is 30 °C, then the charging rate = 0.75, the temperature = 30 °C, and substituting them into the modified regression formula to calculate the modified initial capacity under this evaluation condition = 200.2 Ah.

[0104] Step 4: Calculate the battery SOH under each evaluation condition according to the battery charging capacity and the modified initial capacity.

[0105] In this embodiment, when calculating the battery SOH under each evaluation condition, the initial remaining charge at the start of charging and the remaining charge at the end of charging , and combined with the battery charging capacity and the corrected initial capacity , calculate the battery SOH under this evaluation condition, and the formula is as follows.

[0106] .

[0107] Among them, represents the battery health state value, is the battery charging capacity, is the corrected initial capacity, is the remaining power at the start of charging, is the remaining power at the end of charging.

[0108] Step 5: Use the Gaussian process regression method to fuse the battery SOH under all evaluation conditions to obtain the comprehensive battery SOH.

[0109] In step 5 of this embodiment, it is mainly to fuse the battery SOH under multiple conditions to obtain the comprehensive battery SOH. Use the Matern kernel function to construct a Gaussian process regression model, use the Gaussian process regression model as the SOH fusion model, and use an optimization algorithm to train the hyperparameters of the Gaussian process regression model. According to the trained Gaussian process regression model, fuse the battery SOH under multiple conditions to obtain the comprehensive battery SOH. The specific steps are as follows.

[0110] (1) Use the Gaussian process regression method to fuse the battery SOH for the non-linear regression ability of low-dimensional and small-sample problems. For the one-dimensional time series data of the battery SOH corresponding to evaluation condition 2, use a sliding window to intercept 4 adjacent battery SOH data as the input x of the Gaussian process regression model. Set the SOH corresponding to evaluation condition 1 as the output y of the Gaussian process regression model. To ensure that the input and output dimensions are consistent, when the input data corresponding to evaluation condition 2 is missing, the last 1 SOH data of evaluation condition 1 can be used to fill the missing data; when the output data corresponding to evaluation condition 1 is missing, the mean value of the last 10 SOH data of evaluation condition 2 can be used to fill the missing data.

[0111] (2) According to the prior Gaussian process regression kernel function, use the set input data and output data to train the Gaussian process regression model. Use the Matern kernel function, which is expressed as follows.

[0112]

[0113] Among them, represents the calculation result of the Matern kernel function, represents the th model input vector, represents the th model input vector, , respectively represent the standard deviation and characteristic length of the kernel function, = are hyperparameters to be trained, represents the input variable and the Euclidean distance between, = .

[0114] (3) Using the negative log marginal likelihood function as the objective function, solve for the hyperparameters by minimizing the objective function. The objective function formula is as follows.

[0115]

[0116] Among them, represents the negative log marginal likelihood function; represents the covariance matrix of the training dataset , calculated from the kernel function formula; represents the determinant of; is the number of samples in the training dataset.

[0117] (4) Use optimization algorithms such as quasi-Newton method, genetic algorithm, particle swarm optimization algorithm, etc. to solve for the hyperparameters of the kernel function, complete the training of the Gaussian process regression model, and obtain the mean function and covariance function.

[0118] (5) Use the root mean square error to evaluate the training effect of the Gaussian regression process model, and select the Gaussian regression process model corresponding to the best root mean square error as the final Gaussian regression process model for subsequent battery SOH fusion.

[0119] (6) When the real vehicle data accumulates 4 battery SOHs of evaluation condition 2, an input vector is obtained, and substituting it into the final Gaussian process regression model, the output result of the mean function can be obtained, and the output result of this mean function is the final battery SOH evaluation result, that is, the comprehensive battery SOH.

[0120] In this embodiment, the battery SOH data of a new energy vehicle is fused by the Gaussian process regression method, and the optimized hyperparameters of the kernel function are =[1.5, 6.38], and the root mean square error of the Gaussian process regression model training is 1.7606500392020368e-14. Figure 3 is the schematic diagram of the fitting standard deviation of the Gaussian process regression model, Figure 4 is the curve graph of the battery SOH of this new energy vehicle before and after fusion.

[0121] In this embodiment, considering that there are deviations in the initial capacity of the battery under different charging rates and different temperature conditions, the existing technologies usually use a fixed initial capacity value of 1C or 0.33C at room temperature to calculate the battery SOH, which increases the calculation error of the battery SOH. Based on this, this embodiment corrects the influence of factors such as temperature and charging rate on the evaluation model based on the battery SOH evaluation working condition data. Since the charging rate and temperature are used to correct the initial capacity of the battery at the time of factory, the initial capacity value of this evaluation working condition is used under different charging rates and different temperature evaluation working conditions, thereby improving the evaluation accuracy of the battery SOH.

[0122] In addition, most of the existing technologies at present rely on experimental working conditions or need experimental data to train the model before they can be applied to actual engineering, which has great limitations. And this embodiment uses the battery charging data collected during the actual use of the vehicle and its battery SOH evaluation working condition data, which can be widely applied to the mainstream pure electric vehicle models on the market and has strong applicability to actual vehicles.

[0123] In an exemplary embodiment, a battery SOH evaluation device is provided, as Figure 5 shown, and specifically includes the following modules.

[0124] The battery SOH evaluation working condition data acquisition module M1 is used to acquire battery SOH evaluation working condition data.

[0125] The battery charging capacity calculation module M2 is used to calculate the battery charging capacity under each evaluation working condition by using the ampere-hour integration method according to the battery SOH evaluation working condition data.

[0126] The initial capacity correction module M3 is used to correct the initial capacity under each evaluation working condition by using the non-linear regression method to obtain the corrected initial capacity; the initial capacity refers to the battery capacity when the battery leaves the factory.

[0127] The battery SOH calculation module M4 is used to calculate the battery SOH under each evaluation working condition according to the corrected initial capacity and the battery charging capacity.

[0128] The battery SOH fusion module M5 is used to fuse all the battery SOH by using the Gaussian process regression method to obtain the comprehensive battery SOH.

[0129] In an exemplary embodiment, a computer device is provided. The computer device can be a server or a terminal, and its internal structure diagram can be as Figure 6As shown in the figure. The computer device includes a processor, a memory, an input / output interface (Input / Output, abbreviated as I / O), and a communication interface. Among them, the processor, the memory, and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store battery charging data and its battery SOH evaluation working condition data. The input / output interface of the computer device is used to exchange information between the processor and external devices. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, it implements a method for evaluating battery SOH.

[0130] Those skilled in the art can understand that Figure 6 the structure shown in the figure is only a block diagram of some structures related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0131] In an exemplary embodiment, a computer device is provided, including a memory and a processor. A computer program is stored in the memory, and when the processor executes the computer program, the steps in the above method embodiments are implemented.

[0132] In an exemplary embodiment, a computer-readable storage medium is provided, storing a computer program, and when the computer program is executed by the processor, the steps in the above method embodiments are implemented.

[0133] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above methods. Among them, any reference to a memory, database, or other medium used in the embodiments provided in the present application can include at least one of non-volatile and volatile memories. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc.

[0134] The technical features of the above embodiments can be combined arbitrarily. For the sake of brevity of description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered as the scope recorded in this specification.

[0135] Specific examples are used in this article to elaborate on the principles and implementation manners of the present application. The description of the above embodiments is only used to help understand the method and its core idea of the present application; at the same time, for those of ordinary skill in the art, according to the idea of the present application, there will be changes in the specific implementation manners and application scopes. In summary, the content of this specification should not be construed as a limitation to the present application.

Claims

1. A battery SOH evaluation method, characterized in that: The battery SOH evaluation method comprises: Obtain battery SOH assessment operating condition data; According to the battery SOH evaluation condition data, the battery charging capacity under each evaluation condition is calculated using the ampere-hour integration method; The nonlinear regression method is used to correct the initial capacity under each evaluation condition to obtain the corrected initial capacity; the initial capacity refers to the battery capacity when the battery leaves the factory; Calculating the battery SOH under each evaluation condition according to the corrected initial capacity and the battery charging capacity; The Gaussian process regression method is used to merge all the battery SOHs to obtain the battery comprehensive SOH.

2. The battery SOH evaluation method according to claim 1, characterized in that: Obtain battery SOH assessment operating condition data, including: Acquire battery charging data; the charging data field of the battery charging data includes a charging state field, a charging current field, a SOC field, a vehicle state field and a message time field; The battery charging data is screened according to the charging state field and the charging current field to obtain battery SOH evaluation operating condition data.

3. The battery SOH evaluation method according to claim 2, characterized in that: According to the charging state field and the charging current field, the battery charging data is screened to obtain the battery SOH evaluation condition data, specifically including: According to the charging state field, selecting battery charging data of a complete charging process from full charging of any SOC to 100% SOC from the battery charging data; According to the charging current field, selecting battery charging data having a charging current greater than a threshold from the battery charging data; According to the size of the charging current, the battery charging data with the charging current greater than the threshold is divided into a plurality of constant current charging intervals; Taking the battery charging data corresponding to each constant current charging interval as a battery SOH evaluation operating condition data segment, to obtain a plurality of battery SOH evaluation operating condition data segments; The plurality of battery SOH evaluation operating condition data segments and the battery charging data of the complete charging process are used as battery SOH evaluation operating condition data.

4. The battery SOH evaluation method according to claim 1, characterized in that: The calculation formula for calculating the battery charging capacity under each evaluation condition using the ampere-hour integration method is: ； in, For battery charging capacity, is the charging current, For charging time.

5. The battery SOH evaluation method according to claim 1, characterized in that: The nonlinear regression method is used to correct the initial capacity under each evaluation condition to obtain the corrected initial capacity, including: Constructing a nonlinear regression model; the nonlinear regression model refers to a regression model in which the initial capacity of the battery changes with the charging rate and temperature; With the goal of minimizing the residual sum of squares, the nonlinear regression model is estimated by using a gradient descent method, a Levenberg-Marquardt algorithm, a Powell algorithm, a BFGS algorithm or an SLSQP algorithm to correct the initial capacity under each evaluation condition and obtain a corrected initial capacity.

6. The battery SOH evaluation method according to claim 1, characterized in that: The battery SOH under each evaluation condition is calculated using the following formula: ； in, Indicates the battery health status value, For battery charging capacity, is the corrected initial capacity, The remaining power at the beginning of charging. The remaining power is charged.

7. The battery SOH evaluation method according to claim 1, characterized in that: The Gaussian process regression method is used to merge all the battery SOHs to obtain the battery comprehensive SOH, which specifically includes: Based on the Matern kernel function, a Gaussian process regression model is constructed; Taking the negative log marginal likelihood function as the objective function, solving the hyperparameters of the Gaussian process regression model by minimizing the objective function; Using a quasi-Newton method, a genetic algorithm or a particle swarm optimization algorithm, the hyperparameters of the Gaussian process regression model are trained, and the Gaussian process regression model corresponding to the best root mean square error is selected as the final Gaussian process regression model for the fusion battery SOH; Substitute all the battery SOHs into the final Gaussian process regression model to obtain the battery comprehensive SOH.

8. A battery SOH evaluation device, characterized in that: The battery SOH evaluation device is used to implement the battery SOH evaluation method according to any one of claims 1 to 7, and the battery SOH evaluation device comprises: A battery SOH evaluation operating condition data acquisition module is used to acquire battery SOH evaluation operating condition data; The battery charging capacity calculation module is used to calculate the battery charging capacity under each evaluation condition based on the battery SOH evaluation condition data and using the ampere-hour integration method; An initial capacity correction module is used to correct the initial capacity under each evaluation condition by using a nonlinear regression method to obtain a corrected initial capacity; the initial capacity refers to the battery capacity when the battery leaves the factory; A battery SOH calculation module, used to calculate the battery SOH under various evaluation conditions according to the corrected initial capacity and the battery charging capacity; The battery SOH fusion module is used to fuse all the battery SOHs using the Gaussian process regression method to obtain the battery comprehensive SOH.

9. A computer device comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the battery SOH evaluation method according to any one of claims 1 to 7.

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 battery SOH evaluation method according to any one of claims 1 to 7 is implemented.

Images