Method for constructing and calculating a power storage battery soh calculation model

Abstract

The embodiment of the application discloses a power storage battery SOH calculation model construction method and a calculation method. The model construction method comprises the following steps: obtaining running data of an automobile in a running process and health degree SOH data of a power storage battery, wherein the running data corresponds to multiple data types; processing the running data and the SOH data by using a gradient boosting tree to obtain an influence weight of each data type on SOH, wherein the influence weight represents the importance of each data type on SOH; and performing linear regression analysis on the running data and the SOH data according to the influence weight to obtain a linear regression model of SOH. The embodiment improves the accuracy of SOH calculation.

Description

TECHNICAL FIELD

[0001] The embodiment of the present application relates to the field of battery performance estimation, in particular to a power storage battery SOH calculation model construction method and calculation method. BACKGROUND

[0002] The power storage battery health degree (SOH) of a new energy vehicle reflects the actual performance in the use process of the battery and the performance in the brand new state, and is an important index for measuring the performance of the battery. Accurate prediction of the SOH value of the battery can reasonably determine the replacement time and the cruising range of the battery, and is beneficial to driving safety and recycling of the battery.

[0003] In the prior art, the SOH calculation of the battery is mainly based on the method of establishing an equivalent battery circuit model, and various circuit components (such as resistors, capacitors, voltage sources, etc.) are selected to form an equivalent circuit to describe the external characteristics of the battery. However, the actual driving conditions of the vehicle are very complex, and a simple equivalent circuit cannot cover most of the actual conditions, resulting in that the model constructed cannot accurately predict the SOH of the battery. SUMMARY

[0004] The embodiment of the present application provides a power storage battery SOH calculation model construction method and calculation method, which improves the accuracy of SOH calculation.

[0005] In a first aspect, the embodiment of the present application provides a power storage battery SOH calculation model construction method, comprising:

[0006] Obtaining running data of a vehicle in a running process and health degree SOH data of a power storage battery, wherein the running data corresponds to multiple data types;

[0007] Processing the running data and the SOH data by using a gradient boosting tree to obtain an influence weight of each data type on SOH, wherein the influence weight represents the importance of each data type on SOH;

[0008] According to the influence weight, performing linear regression analysis on the running data and the SOH data to obtain a linear regression model of SOH.

[0009] In a second aspect, the embodiment of the present application provides a power storage battery SOH calculation method, comprising:

[0010] Obtaining a linear regression model of the health degree SOH of the power storage battery, wherein the linear regression model is constructed by using the method in the above embodiment;

[0011] Substituting the running data of the vehicle in a to-be-tested working condition into the linear regression model to obtain SOH data corresponding to the to-be-tested working condition.

[0012] In a third aspect, an electronic device is provided, and the electronic device includes:

[0013] one or more processors;

[0014] a memory for storing one or more programs,

[0015] When the one or more programs are executed by the one or more processors, the one or more processors implement the power storage battery SOH calculation model construction method or the power storage battery SOH calculation method according to any of the embodiments.

[0016] In a fourth aspect, a computer readable storage medium is provided, and the computer readable storage medium stores a computer program, and the computer program is executed by a processor to implement the power storage battery SOH calculation model construction method or the power storage battery SOH calculation method according to any of the embodiments.

[0017] The embodiments of the present application fuse gradient boosting tree and regression analysis, and estimate regression coefficients under the guidance of influence weights, so as to take into account the action mechanisms of the influence factors reflected by the two algorithms, and construct a more accurate calculation model. BRIEF DESCRIPTION OF DRAWINGS

[0018] In order to more clearly illustrate the specific embodiments of the present application or the technical solutions in the prior art, the drawings needed in the specific embodiments or the prior art description will be briefly introduced as follows. Obviously, the drawings in the following description are some embodiments of the present application, and those skilled in the art can obtain other drawings according to these drawings without creative labor.

[0019] Figure 1 is a flowchart of a power storage battery SOH calculation model construction method provided by the embodiments of the present application.

[0020] Figure 2 is a flowchart of a power storage battery SOH calculation method provided by the embodiments of the present application.

[0021] Figure 3 is a structural schematic diagram of an electronic device provided by the embodiments of the present application. DETAILED DESCRIPTION

[0022] In order to make the objects, technical solutions and advantages of the present application clearer, the technical solutions of the present application will be described clearly and completely. Obviously, the described embodiments are only some of the embodiments of the present application, not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0023] In the description of the present application, it should be noted that the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer" and the like indicate the orientation or positional relationship shown in the drawings, and are only for the convenience of describing the present application and simplifying the description, and do not indicate or imply that the devices or elements referred to must have a particular orientation, be constructed and operated in a particular orientation, and therefore cannot be understood as limiting the present application. In addition, the terms "first", "second", "third" are only for descriptive purposes and cannot be understood as indicating or implying relative importance.

[0024] In the description of the present application, it should be noted that, unless otherwise explicitly specified and limited, the terms "mounting", "connection", "connection" should be understood broadly, for example, it can be fixedly connected, or it can be detachably connected, or integrally connected; it can be mechanically connected, or it can be electrically connected; it can be directly connected, or it can be indirectly connected through an intermediate medium; it can be the communication inside two elements. For those skilled in the art, the specific meaning of the above terms in the present application can be understood according to the specific circumstances.

[0025] Figure 1 is a flowchart of a power storage battery SOH calculation model construction method provided by an embodiment of the present application. The method is applicable to the case of constructing a relationship model between automobile running data and battery SOH data, and is executed by an electronic device. As shown in Figure 1 , the method specifically comprises:

[0026] S110, acquiring running data of the automobile in the running process and SOH data of the power storage battery, wherein the running data corresponds to a plurality of data types.

[0027] The plurality of data types include: automobile mileage, battery use time, battery charging time, average charging temperature, average charging voltage difference, etc. It can be seen that these data types are all influencing factors of SOH, for example, the average charging temperature affects the efficiency and aging speed of the battery, and the battery charging time affects the service life of the battery. These types of running data will be used as a data source for constructing the SOH calculation model. Optionally, the running data uploaded by the new energy automobile through the T-BOX is acquired from the new energy automobile remote monitoring platform. The running data reflects the performance of the automobile under actual working conditions, and is more in line with the actual situation.

[0028] It should be explained that the running data in the present embodiment specifically refers to the running data value. For a data type, it can be one running data value at one time, such as the automobile mileage value, or one running data value in one period, such as the average charging voltage difference value.

[0029] In addition to obtaining the vehicle operation data, the SOH data corresponding to the operation data is also obtained as another data source for constructing the SOH calculation model. The SOH of the battery reflects the comparison between the actual performance in the use process of the battery and the performance in the brand-new state, and is defined as follows:

[0030] SOH = (C act / C nom )*100% (1)

[0031] wherein C act represents the actual capacity of the battery, and C nom represents the rated capacity of the battery.

[0032] Optionally, after obtaining the operation data of the vehicle in the operation process, the simulated discharge of the power storage battery under the operation data is performed; and the SOH data corresponding to the operation data is obtained according to the simulated discharge amount. Specifically, the simulated discharge can be controlled to be performed in a simulation platform by a vehicle model, or to be performed in a laboratory environment by a battery entity, and the simulated discharge amount obtained is the actual capacity in formula (1).

[0033] Similarly, the SOH data in this embodiment specifically refers to the SOH value, which can correspond to one SOH value at one time or one SOH value in one period, and the specific corresponding relationship matches the corresponding relationship of the operation data.

[0034] S120, processing the operation data and the SOH data by using a gradient boosting tree to obtain an influence weight of each data type on the SOH, wherein the influence weight represents the importance of each data type on the SOH.

[0035] This step uses the GBDT (Gradient Boosting Decision Tree) algorithm to quantitatively analyze the importance of each data type on the SOH. Specifically, the SOH data and the operation data are taken as inputs of the GBDT, and the GBDT calculates the influence weight of each data type on the SOH to reflect the importance of each data type in a quantitative numerical result, thereby preparing for the linear regression analysis in the next step.

[0036] S130, performing linear regression analysis on the operation data and the SOH data according to the influence weight to obtain a linear regression model of the SOH.

[0037] The basic principle of the regression analysis is to regard one variable as a dependent variable and one or more other variables as independent variables among the related variables, to establish a linear or nonlinear quantitative relationship between the variables, and to perform statistical analysis by using sample data.

[0038] The embodiment takes SOH as the dependent variable type Y, takes multiple data types affecting SOH as k independent variable types, and adopts a linear regression model as the basic form of the SOH calculation model.

[0039] Y = β0+1X1+2X2+…+ k X k (2)

[0040] wherein the linear coefficients β1, β2… k and the constant term β0 are unknown constants, collectively referred to as regression coefficients in the linear regression model. Based on the above model, the SOH data and the operation data are taken as sample data to estimate the regression coefficients, and the SOH calculation model is obtained.

[0041] It can be seen that, similar to the influence weights obtained in S120, the regression coefficients in formula (2) also reflect the influence of X1,2… k on SOH. That is, both the gradient boosting tree and the regression analysis algorithm can fit the action mechanism of each influencing factor on SOH. The difference is that the influence weights obtained by the gradient boosting tree are not limited to the linear form, but the calculation is complex and no specific model formula is formed. The regression coefficients in the regression analysis reflect the linear relationship, but it is beneficial to form a specific model formula and the calculation is simple.

[0042] Therefore, the embodiment combines the two algorithms, estimates the regression coefficients under the guidance of the influence weights, takes into account the action mechanism of the influencing factors reflected by the two algorithms, and constructs a more accurate calculation model.

[0043] The following describes the process of regression analysis according to the influence weights through two implementation manners.

[0044] In the first implementation manner, first, the independent variable types of the linear regression model of SOH are selected from the multiple data types.

[0045] Generally, there are many influencing factors of SOH, and there are also many types of operation data that can be obtained. If all the data types that can be obtained are taken as the independent variable types of the linear regression model, not only the calculation time of the regression analysis will be increased, but also unnecessary influencing factors may be introduced, leading to difficulty in model convergence or inaccuracy. Therefore, this step first selects a part of the important data types as the independent variable types from the multiple data types. The specific selection manner will be described in detail in subsequent embodiments.

[0046] After the independent variable types are selected, the influence weights corresponding to the independent variable types are taken as the linear coefficients of the independent variable types in the linear regression model.

[0047] In order to keep the influence weight obtained by the gradient boosting tree, the embodiment directly takes the influence weight as the linear coefficient in the linear regression model of each variable type. That is, the influence weight of X1 is directly taken as the value of β1 in formula (2), the influence weight of X2 is directly taken as the value of β2 in formula (2), and so on. At this time, only the value of the constant term β0 in formula (2) is to be determined.

[0048] After determining the linear coefficient, the SOH data and the operation data corresponding to the variable type are substituted into the linear regression model to perform linear regression analysis to determine the constant term in the linear regression model.

[0049] The step adopts linear regression analysis to determine the constant term β0, on the one hand, to obtain a complete linear regression model, and on the other hand, to make the linear regression model further adapt to the collected SOH data and operation data, reflect the actual effect of the operation data on the SOH data, and obtain a prediction result in line with the actual situation.

[0050] In the second implementation, first, the variable types of the linear regression model of SOH are selected from the plurality of data types.

[0051] The specific selection method is the same as that of the first implementation, which will be described in detail in subsequent embodiments.

[0052] After selecting the variable types, a loss function for linear regression analysis is constructed according to the influence weight corresponding to the variable types.

[0053] The loss function is used to estimate the degree of inconsistency between the predicted value of the linear regression model and the true value, and the optimal linear regression model can be determined by minimizing the loss function. In the embodiment, the Y value calculated by formula (2) is the predicted value, and the SOH data obtained in step S110 is the true value. If the loss function adopts the form of mean square error, the traditional loss function is represented as:

[0054]

[0055] wherein, SOH m represents the true value of the mth sample, Y m represents the predicted value of the mth sample, and m represents the sample quantity. In the embodiment, a group of operation data corresponding to the variable types and an SOH data constitute a sample.

[0056] Based on the above loss function, the embodiment structures the influence weight obtained by the gradient boosting tree to construct the following loss function:

[0057]

[0058] wherein, Xm represents the mth sample of X j The corresponding operating data in the predicted value Y m The components; Ym represents the mth sample of Y m Loss in other dimensions except X j Constrained by X j The importance of SOH α j .

[0059] Specifically, the Loss term is constrained as a weight When X j The importance of SOH α j The smaller, The greater the importance of The weight in the loss function The greater the impact on the linear regression model. In this way, while ensuring the accuracy of the prediction result, the importance of the independent variable type to SOH is also reflected, so that the final linear regression model is consistent with the data law reflected by α1, α2…α k .

[0060] If the loss function adopts the form of mean absolute error, the traditional loss function is represented as:

[0061]

[0062] The embodiment constructs a new loss function:

[0063]

[0064] Optionally, if α j All are ratios between 0 and 1, then In formula (4) and formula (6) can be replaced by (1-α j ).

[0065] After constructing the loss function, according to the loss function, the operating data and the SOH data are subjected to linear regression analysis to obtain regression coefficients β0, β1, β2…β k , so as to determine the linear regression model of SOH.

[0066] In the above loss function, both the consistency between the true value and the predicted value and the importance of each independent variable type to SOH are reflected. Through minimization of the loss function, the prediction result can be consistent with the true data, and the regression model can be consistent with the data law reflected by α1, α2…α k , making the prediction result more scientific.

[0067] The two embodiments above combine gradient boosting trees and regression analysis from different perspectives to determine a linear regression model of SOH. In the first embodiment, the influence weight of GBDT is explicitly applied in the regression model, and the regression of the constant term is used to adapt to the sample data. The whole construction process is simple in operation and can also ensure the accuracy of prediction. In the second embodiment, the influence weight of GBDT is implicitly applied in the regression model, and the loss function is used to constrain the importance of the independent variable type to SOH. The constraint mechanism is more flexible, and the sample data can also be adapted to obtain accurate prediction results.

[0068] Based on the above and the following embodiments, the present embodiment adds pre-processing to the sample data. Optionally, after the TBOX collects the original data of the vehicle during operation, the original data is cleaned, and the cleaned data is used as the operation data in S110.

[0069] Specifically, the original data uploaded by TOX often has problems such as noise, missing, abnormality, inconsistency. Data cleaning mainly deletes irrelevant data, repeated data, smooth noise data, handles missing values, abnormal value processing, deletes records and data interpolation, etc. to extract effective charging data.

[0070] In addition, since X1, X2…X k and Y have different variable types and different data dimensions, they are not directly comparable. In order to eliminate the influence of different dimensions on the model, optionally, the values of each variable are set to normalized data for the independent variable type.

[0071] Specifically, for any variable type, the variable data is adjusted as follows:

[0072]

[0073] wherein V represents any one of X1, X2…X k and Y, V max and V min respectively represent the maximum and minimum values of V in the data sample; V and respectively represent the set maximum and minimum normalized values. These two normalized values can be set according to actual needs, for example, to form a percentage result.

[0074] For example, the variable type of X1 is the mileage of the vehicle, then:

[0075]

[0076] For example, the variable type of Y is SOH, then:

[0077]

[0078] By normalizing the sample data of each variable type, on the one hand, the influence of different dimensions on the data can be eliminated, and the contribution of different data types to SOH can be unified to a comparable numerical dimension, which is conducive to data fusion; on the other hand, the data range [V min , V max ] of a certain variable type in the sample can be reduced or expanded to a new range , so that the sample is more evenly distributed in the new range, and overfitting phenomenon is avoided.

[0079] Based on the above and the following embodiments, this embodiment refines the selection process of the independent variable type. Specifically, from the plurality of data types corresponding to the operating data, a part of the types are selected as the independent variable types in the SOH linear regression model. This embodiment gives two selection methods:

[0080] Method one, from the plurality of data types, select the data types with an impact weight greater than a set threshold as the independent variable types of the SOH linear regression model.

[0081] The impact weight as a numerical reflection of the importance of the independent variable can be used as a basis for selecting the independent variable type. Select the data type with an impact weight greater than a set threshold (for example, 0.2) as the independent variable type; the data type with an impact weight less than the set threshold can not be reflected in the linear regression model.

[0082] Method two, according to the correlation coefficient matrix of the plurality of data types and SOH, select the independent variable types of the SOH linear regression model.

[0083] The correlation coefficient is used to reflect the closeness of the correlation between variables. The matrix composed of the correlation coefficients between each variable type is called the correlation coefficient matrix. Through the correlation coefficient matrix between each operating data type and SOH, the independent variable types in the linear regression model can be selected.

[0084] Specifically, according to the operating data and SOH data obtained in S110, the correlation between each operating data type and SOH is analyzed to obtain the correlation coefficient matrix as shown in Table 1:

[0085] Table 1

[0086]

[0087] Wherein, the rows and columns of the correlation coefficient matrix are SOH and each operation data type, and the value corresponding to a row and a column is the correlation coefficient between the variable type of the row and the variable type of the column, for example, the correlation coefficient between the automobile mileage and SOH is -0.0443.

[0088] According to the above correlation coefficient matrix, a plurality of data types with an absolute value of the correlation coefficient with SOH greater than a second threshold (such as 0.2) are selected as a plurality of independent variable types of the linear regression model of SOH. As shown in Table 1, the plurality of data types with an absolute value of the correlation coefficient with SOH greater than 0.2 include the automobile mileage, the battery use time, the battery charging time and the charging average temperature, and then the four data types are selected as the independent variable types of the linear regression model.

[0089] Optionally, after the plurality of independent variable types are determined through the second threshold, the method further includes: if the absolute value of the correlation coefficient between any two independent variable types in the plurality of independent variable types is greater than a third threshold (such as 0.5), the any one of the two independent variable types is removed.

[0090] Still taking Table 1 as an example, the correlation coefficient between the independent variable types of the automobile mileage and the battery use time is 0.698, which is greater than 0.5, indicating that there is a high linear correlation between the two independent variable types, and then in the linear regression model, the influence of the two independent variable types on SOH can be simultaneously reflected by adjusting the linear coefficient of any one of the two independent variable types. Therefore, any one of the two independent variable types is removed, the model complexity is reduced, and the influence of the linear relationship between the independent variables on the model is eliminated.

[0091] It should be noted that the method one and the method two give two selection methods of the independent variable types. In actual application, the method one can be used alone, or the method two can be used alone, or the method one can be used after the method two, or the method two can be used after the method one, and the embodiment does not make a specific limitation on this.

[0092] Figure 2 is a flowchart of a power storage battery SOH calculation method provided by an embodiment of the application. The method is suitable for calculating the battery SOH through the automobile operation data, and is executed by an electronic device. As shown in Figure 2 , the method specifically includes:

[0093] S210, a linear regression model of the power storage battery SOH is acquired, wherein the linear regression model is constructed by using the method in any one of the above embodiments.

[0094] S220, the operation data of the automobile under the to-be-tested working condition is substituted into the linear regression model, and the SOH data corresponding to the to-be-tested working condition is obtained.

[0095] The to-be-tested working condition is a moment or a period, and the SOH value corresponding to the working condition is predicted by using the running data of the working condition.

[0096] The SOH calculation method provided in the embodiment is implemented based on any of the above embodiments and has the technical effects of any of the above embodiments.

[0097] Optionally, after obtaining the SOH data corresponding to the to-be-tested working condition, at least one of the following steps is further included: determining the endurance time of the battery according to the SOH data, and determining the scrapping time of the battery according to the SOH data.

[0098] In a specific embodiment, the current capacity of the battery is calculated from the current SOH data according to the definition of SOH, the discharge speed of the battery in the current working condition is obtained, the ratio of the current capacity to the discharge speed is calculated, and the endurance time of the battery is obtained. Calculating the endurance time of the battery is beneficial to timely reminding the user to charge, and avoiding the situation that the vehicle is powered off during driving.

[0099] In another specific embodiment, the corresponding relationship between the SOH data of the battery after each charging and the final scrapping time is analyzed in a big data manner, and the scrapping time of the battery is determined according to the corresponding relationship. Calculating the scrapping time of the battery is beneficial to timely reminding the user to replace the battery, and avoiding the additional environmental burden caused by battery failure or low battery performance.

[0100] Figure 3 A structural schematic diagram of an electronic device provided in an embodiment of the present application is shown in FIG. 1. Figure 3 As shown in FIG. 1, the device includes a processor 60, a memory 61, an input device 62, and an output device 63; the number of processors 60 in the device can be one or more, Figure 3 and one processor 60 is taken as an example; the processor 60, the memory 61, the input device 62, and the output device 63 in the device can be connected through a bus or other means, Figure 3 and a connection through a bus is taken as an example.

[0101] The memory 61 is a computer readable storage medium, which can be used to store software programs, computer executable programs, and modules, such as program instructions / modules corresponding to the power storage battery SOH calculation model construction method or the power storage battery SOH calculation method in the embodiment of the present application. The processor 60 executes the software programs, instructions, and modules stored in the memory 61, thereby performing various functional applications and data processing of the device, that is, implementing the power storage battery SOH calculation model construction method or the power storage battery SOH calculation method.

[0102] The memory 61 can include a program storage area and a data storage area, where the program storage area can store an operating system, application programs required by at least one function, and the data storage area can store data created according to the use of the terminal, etc. In addition, the memory 61 can include a high-speed random access memory, and can also include a non-volatile memory such as at least one magnetic disk storage device, a flash memory device, or other non-volatile solid-state storage device. In some examples, the memory 61 can further include a memory disposed remotely with respect to the processor 60, which can be connected to the device through a network. Examples of the above network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and a combination thereof.

[0103] The input device 62 can be used to receive input digital or character information, and to generate key signal inputs related to user settings of the device and function controls. The output device 63 can include a display device such as a display screen.

[0104] The embodiment of the present application also provides a computer readable storage medium, which stores a computer program, and the program is executed by a processor to implement the power storage battery SOH calculation model construction method or the power storage battery SOH calculation method of any embodiment.

[0105] The computer storage medium of the embodiment of the present application can adopt any combination of one or more computer readable media. The computer readable medium can be a computer readable signal medium or a computer readable storage medium. The computer readable storage medium may, for example, but is not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or apparatus, or any combination thereof. More specific examples (non-exhaustive list) of the computer readable storage medium include: an electrical connection having one or more wires, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In this document, the computer readable storage medium can be any tangible medium that contains or stores a program that can be used by or in connection with an instruction execution system, device or apparatus.

[0106] The computer readable signal medium can include a data signal propagated in a baseband or as a part of a carrier wave, where the computer readable program code is carried. Such propagated data signals can take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. The computer readable signal medium can also be any computer readable medium that is not a computer readable storage medium and that can transmit, propagate or transport program for use by or in connection with an instruction execution system, device or apparatus.

[0107] The program code embodied on the computer readable media can be transmitted using any appropriate medium, including but not limited to wireless, wire line, optical fiber cable, RF, etc., or any suitable combination of the foregoing.

[0108] Computer program code for carrying out operations of the present application can be written in any combination of one or more programming languages, including an object oriented programming language such as Java, Smalltalk, C++ or the like and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code can execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the latter scenario, the remote computer can be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or the connection can be made to an external computer (for example, through the Internet using an Internet Service Provider).

[0109] Finally, it should be noted that the above-mentioned embodiments are merely used to illustrate the technical solutions of the present application, instead of limiting the present application; although the present application has been described in detail with reference to the above-mentioned embodiments, those skilled in the art should understand that the technical solutions recorded in the above-mentioned embodiments can be modified or equivalent replacements can be made to some or all of the technical features; and the modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the technical solutions of the embodiments of the present application.

Claims

1. A method of constructing a power storage battery SOH calculation model, characterized by, The method comprises the following steps: obtaining running data of a vehicle during operation and state of health (SOH) data of a power storage battery, wherein the running data corresponds to multiple data types; processing the running data and the SOH data by using a gradient boosting tree to obtain an influence weight of each data type on SOH, wherein the influence weight represents the importance of each data type on SOH; performing linear regression analysis on the running data and the SOH data according to the influence weight to obtain a linear regression model of SOH; the step of performing linear regression analysis on the running data and the SOH data according to the influence weight to obtain a linear regression model of SOH comprises the following steps: selecting an independent variable type of the linear regression model of SOH from the multiple data types; constructing a loss function for linear regression analysis according to the influence weight corresponding to the independent variable type; and performing linear regression analysis on the running data and the SOH data according to the loss function to obtain the linear regression model of SOH. The loss function L2 is as follows: wherein SOH m represents the true value of the mth sample, Y m represents the predicted value of the mth sample, m represents the sample number; represents the component of the mth sample X j corresponding to the operating data in the predicted value Y m ; represents the Y m loss in other dimensions except X j , is constrained by the importance of X j to SOH α j .

2. The method of claim 1, wherein, the multiple data types comprise vehicle mileage, battery use time, battery charging time, average charging temperature and average charging voltage difference.

3. The method of claim 1, wherein, The method comprises the following steps: obtaining running data of a vehicle during operation; performing simulated discharge of a power storage battery under the running data; obtaining SOH data corresponding to the running data according to the simulated discharge amount.

4. The method of claim 1, wherein, The step of performing linear regression analysis on the running data and the SOH data according to the influence weight to obtain a linear regression model of SOH comprises the following steps: selecting an independent variable type of the linear regression model of SOH from the multiple data types; taking the influence weight corresponding to the independent variable type as a linear coefficient of the independent variable type in the linear regression model; substituting the SOH data and the running data corresponding to the independent variable type into the linear regression model to perform linear regression analysis and determine a constant term in the linear regression model.

5. The method according to claim 4 or 1, characterized in that, The step of selecting an independent variable type of the linear regression model of SOH from the multiple data types comprises the following steps: selecting a data type with an influence weight greater than a set threshold value from the multiple data types as the independent variable type of the linear regression model of SOH; and / or selecting an independent variable type of the linear regression model of SOH according to a correlation coefficient matrix of the multiple data types and SOH. The step of selecting an independent variable type of the linear regression model of SOH according to a correlation coefficient matrix of the multiple data types and SOH comprises the following steps:

6. The method of claim 5, wherein, selecting multiple data types with a correlation coefficient absolute value greater than a second threshold value from the correlation coefficient matrix of the multiple data types and SOH as multiple independent variable types of the linear regression model of SOH; and if there are two independent variable types with a correlation coefficient absolute value greater than a third threshold value, eliminating any one of the two independent variable types. The method comprises the following steps:

7. A method of calculating the state of health (SOH) of a power storage battery, characterized by, ​ obtaining a linear regression model of state of health (SOH) of the power battery, wherein the linear regression model is constructed by the method according to any one of claims 1-6; substituting the running data of the automobile in the to-be-tested working condition into the linear regression model to obtain SOH data corresponding to the to-be-tested working condition.

8. An electronic device, comprising: comprising: a memory, a processor, and a computer program stored in the memory and executable by the processor, wherein the processor implements the construction method according to any one of claims 1-6 or the calculation method according to claim 7 when executing the computer program.

9. A computer-readable storage medium, characterized in that, The computer readable storage medium stores computer instructions for causing the computer to execute the construction method according to any one of claims 1-6 or the calculation method according to claim 7.

Images