Method and device for determining SOC of battery, automobile and storage medium

By using equivalent circuit model and pre-trained SOC error estimation model in electric vehicles, combined with current eigenvalue and SOC estimation value for compensation, the problem of insufficient SOC estimation accuracy in electric vehicles is solved, and higher estimation accuracy is achieved .

CN118938023APending Publication Date: 2024-11-12GUANGZHOU XIAOPENG MOTORS TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411165736.4
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-22
Publication Date
2024-11-12

AI Technical Summary

Technical Problem

In the prior art, the battery SOC estimation accuracy in electric vehicles is not high, especially the current factors are difficult to quantify and match, resulting in insufficient SOC estimation accuracy.

Method used

The preset estimation algorithm based on the equivalent circuit model is used to obtain the current characteristic value and SOC estimation value of the power battery, and compensate through the pre-trained SOC error estimation model to improve the accuracy of the SOC output value.

Benefits of technology

The battery SOC estimation accuracy is improved, the current factor affects SOC estimation, and the accuracy of the estimation value is improved.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118938023A_ABST
    Figure CN118938023A_ABST
Patent Text Reader

Abstract

The invention relates to a method and device for determining the SOC of a battery, an automobile and a storage medium. The method comprises the following steps: acquiring a first current characteristic value of a power battery in a vehicle driving process; a first SOC estimated value, determined according to a preset estimation algorithm, of the power battery in the vehicle running process is obtained, the preset estimation algorithm is determined based on an equivalent circuit model of the power battery, and consideration factors of the equivalent circuit model in the parameter identification stage do not contain current parameters; inputting the first current characteristic value and the first SOC estimation value into a pre-trained SOC error estimation model to obtain an SOC compensation value; and determining an SOC output value according to the first SOC estimation value and the SOC compensation value. According to the scheme provided by the invention, the SOC estimation precision of the battery can be improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the technical field of electric vehicles, and in particular to a method, device, vehicle and storage medium for determining a battery SOC. Background Art

[0002] With the rapid development of electric vehicles in recent years, the accuracy of the remaining mileage estimation of electric vehicles has received more and more attention. The remaining mileage estimation requires the estimation of the battery's SOC (State of Charge). Generally speaking, lithium-ion batteries will show different voltage and internal resistance characteristics under different SOC, temperature, charge and discharge current, and aging degree, and the changes in voltage and internal resistance will show a certain regular distribution with the changes in various influencing factors. In practice, electric vehicles need to estimate the battery's SOC in real time during use. They often establish an equivalent circuit model for the battery to match its external characteristics, and use a state estimation algorithm to estimate the SOC on this basis.

[0003] Among them, the factors that affect the model parameters of the battery mainly include SOC, temperature, charge and discharge current, and aging degree. Considering the time segments of seconds and minutes, the changes in SOC, temperature, and aging degree can be considered to be limited, while the current change can reach thousands of amperes, and the current change will also affect the estimation of SOC with a lag. In the laboratory, the temperature, aging degree, and SOC factors of the test can be well matched with the actual vehicle working conditions, but the current factor is difficult to quantify and match. The relevant technology generally adopts the SOC estimation method of the equivalent circuit model. This method deals with the influence of the current factor on the battery equivalent circuit model. Generally, in the modeling stage of the equivalent circuit model, the current is used as a variable to carry out a series of tests, and the equivalent circuit model parameters are extracted. The obtained parameter matrix completely includes multiple dimensions such as SOC, temperature, and current. However, this technology requires a lot of testing in the early stage, and the parameter matrix actually used is complex and computationally intensive. Moreover, the SOC estimation method using the equivalent circuit model is not very accurate at present and needs to be further improved. Summary of the invention

[0004] In order to solve or partially solve the problems existing in the related art, the present application provides a method, device, vehicle and storage medium for determining battery SOC, which can improve the battery SOC estimation accuracy.

[0005] A first aspect of the present application provides a method for determining a battery SOC, comprising: Acquire a first current characteristic value of a power battery during vehicle driving; Obtaining a first SOC estimation value of the power battery during vehicle driving determined according to a preset estimation algorithm, wherein the preset estimation algorithm is determined based on an equivalent circuit model of the power battery, and the equivalent circuit model does not include current parameters as a factor considered in a parameter identification stage; Inputting the first current characteristic value and the first SOC estimation value into a pre-trained SOC error estimation model to obtain an SOC compensation value; An SOC output value is determined according to the first SOC estimation value and the SOC compensation value.

[0006] In one embodiment, determining the SOC output value according to the first SOC estimation value and the SOC compensation value includes: The first SOC estimation value is added to the SOC compensation value to obtain an SOC output value.

[0007] In one embodiment, obtaining a first current characteristic value of a power battery during vehicle driving includes: A first current characteristic value of the power battery calculated according to the real-time current during the driving of the vehicle is obtained.

[0008] In one embodiment, the pre-trained SOC error estimation model is pre-trained in the following manner: Acquire a second current characteristic value in a power battery test environment; Obtaining a second SOC estimation value in a power battery test environment determined according to a preset estimation algorithm; Determining an SOC error value according to the actual SOC value of the power battery and the second SOC estimated value; Taking the second current characteristic value and the second SOC estimation value as characteristic quantities, taking the SOC error value as a label, and establishing a data pair set of the label and the characteristic quantity; Based on the data pair set, a preset algorithm is used to train the relationship between the label and the feature quantity to obtain the pre-trained SOC error estimation model.

[0009] In one embodiment, the preset algorithm includes a regression algorithm.

[0010] In one embodiment, obtaining the second current characteristic value in the power battery test environment includes: A second current characteristic value of the power battery is obtained by calculating the real-time current under the operating condition test of the power battery.

[0011] In one embodiment, the second current characteristic value of the power battery calculated according to the real-time current includes: Performing a standard HPPC test on the equivalent circuit model of the power battery to extract a time constant representing the charge transfer process; A second current characteristic value of the power battery is calculated based on the real-time current and the time constant.

[0012] In one embodiment, the using a preset algorithm to train the relationship between the label and the feature quantity includes: Using a regression algorithm to train the relationship between the label and the feature quantity to obtain a potential relationship between the label and the feature quantity; The potential relationship is quantified to obtain the relationship between the label and the feature quantity.

[0013] A second aspect of the present application provides a device for determining a battery SOC, comprising: A first acquisition module, used to acquire a first current characteristic value of a power battery during the driving process of the vehicle; A second acquisition module is used to acquire a first SOC estimation value of the power battery during vehicle driving determined according to a preset estimation algorithm, wherein the preset estimation algorithm is determined based on an equivalent circuit model of the power battery, and the equivalent circuit model does not include current parameters as a factor considered in a parameter identification stage; A model processing module, configured to input the first current characteristic value and the first SOC estimation value into a pre-trained SOC error estimation model to obtain an SOC compensation value; The compensation output module is used to determine an SOC output value according to the first SOC estimation value and the SOC compensation value.

[0014] In one embodiment, the device further comprises: A model training module is used to obtain a second current characteristic value in a power battery test environment; obtain a second SOC estimation value in the power battery test environment determined according to a preset estimation algorithm; determine an SOC error value according to the SOC true value of the power battery and the second SOC estimation value; use the second current characteristic value and the second SOC estimation value as feature quantities, and use the SOC error value as a label to establish a data pair set of the label and the feature quantity; based on the data pair set, use a preset algorithm to train the relationship between the label and the feature quantity to obtain the pre-trained SOC error estimation model.

[0015] A third aspect of the present application provides a car, comprising: Processor; and The memory stores executable codes thereon, and when the executable codes are executed by the processor, the processor is caused to execute the method as described above.

[0016] A fourth aspect of the present application provides a computer-readable storage medium having executable code stored thereon. When the executable code is executed by a processor of an electronic device, the processor is caused to execute the method as described above.

[0017] The technical solution provided by this application may have the following beneficial effects: The technical solution of the present application obtains a first current characteristic value of a power battery during vehicle driving; A first SOC estimation value of a power battery during vehicle driving is obtained according to a preset estimation algorithm, wherein the preset estimation algorithm is determined based on an equivalent circuit model of the power battery, and the equivalent circuit model does not include current parameters in the consideration of the parameter identification stage; the first current characteristic value and the first SOC estimation value are input into a pre-trained SOC error estimation model to obtain an SOC compensation value; an SOC output value is determined according to the first SOC estimation value and the SOC compensation value, and the SOC output value is output as the final SOC estimation value to improve the accuracy of the conventional SOC estimation value of the battery based on the equivalent circuit model.

[0018] It should be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] The above and other objects, features and advantages of the present application will become more apparent by describing in more detail exemplary embodiments of the present application in conjunction with the accompanying drawings, wherein the same reference numerals generally represent the same components in the exemplary embodiments of the present application.

[0020] Figure 1 is a flow chart of a method for determining a battery SOC according to an embodiment of the present application; Figure 2 is a schematic diagram of a process of pre-training a SOC error estimation model according to an embodiment of the present application; Figure 3 is another flow chart of a method for determining a battery SOC shown in an embodiment of the present application; Figure 4 is a schematic diagram of the process of the power battery in the working condition test stage shown in the embodiment of the present application; Figure 5 is a schematic structural diagram of a device for determining a battery SOC according to an embodiment of the present application; Figure 6 It is a schematic diagram of the structure of an electronic device shown in an embodiment of the present application. DETAILED DESCRIPTION

[0021] The embodiments of the present application will be described in more detail below with reference to the accompanying drawings. Although the embodiments of the present application are shown in the accompanying drawings, it should be understood that the present application can be implemented in various forms and should not be limited by the embodiments described herein. On the contrary, these embodiments are provided to make the present application more thorough and complete, and to fully convey the scope of the present application to those skilled in the art.

[0022] The terms used in this application are for the purpose of describing specific embodiments only and are not intended to limit this application. The singular forms of "a", "said" and "the" used in this application and the appended claims are also intended to include plural forms unless the context clearly indicates other meanings. It should also be understood that the term "and / or" used herein refers to and includes any or all possible combinations of one or more associated listed items.

[0023] It should be understood that although the terms "first", "second", "third", etc. may be used in this application to describe various information, this information should not be limited to these terms. These terms are only used to distinguish the same type of information from each other. For example, without departing from the scope of this application, the first information may also be referred to as the second information, and similarly, the second information may also be referred to as the first information. Thus, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of this application, the meaning of "multiple" is two or more, unless otherwise clearly and specifically defined.

[0024] In related technologies, electric vehicles need to estimate the SOC of the battery in real time during use. This is often done by building an equivalent circuit model for the battery to match its external characteristics, and then using a state estimation algorithm to estimate the SOC. However, the SOC estimation method using the equivalent circuit model is not very accurate at present and needs to be further improved.

[0025] In view of the above problems, an embodiment of the present application provides a method for determining a battery SOC, which can improve the estimation accuracy of the battery SOC.

[0026] The technical solution of the embodiments of the present application is described in detail below with reference to the accompanying drawings.

[0027] Figure 1 It is a flow chart of a method for determining battery SOC shown in an embodiment of the present application.

[0028] like Figure 1 As shown, an embodiment of the present application provides a method for determining a battery SOC. The battery in the method may be a power battery of an electric vehicle. The method for determining a battery SOC mainly includes steps S101 to S104, which are described as follows.

[0029] Step S101: obtaining a first current characteristic value of a power battery during vehicle driving.

[0030] Wherein, step S101 may further include: obtaining a first current characteristic value of the power battery calculated according to the real-time current during the driving of the vehicle. In this embodiment, the power battery is arranged on the electric vehicle, and during the driving of the electric vehicle, the real-time current I1(t) may be filtered according to a predetermined time constant tau to obtain the first current characteristic value I1_feature(t) of the power battery.

[0031] Among them, the predetermined time constant tau can be obtained by performing a standard HPPC (Hybrid Pulse Power Characterization) test on the power battery in a laboratory test environment, and the power battery in the test environment and the power battery during vehicle driving can be the same or two with the same specifications.

[0032] Step S102: Obtain a first SOC estimation value of the power battery during vehicle driving determined according to a preset estimation algorithm, wherein the preset estimation algorithm is determined based on an equivalent circuit model of the power battery, and the equivalent circuit model does not include current parameters in the consideration factors during the parameter identification stage.

[0033] Among them, in step S102, the step of establishing a preset estimation algorithm may be: establishing an equivalent circuit model of the power battery, wherein parameter factors such as SOC, temperature and aging degree are considered in the parameter identification stage of the equivalent circuit model of the power battery, that is, the current parameter is not considered; on this basis, an SOC estimation algorithm based on the equivalent circuit model of the power battery is established, that is, the preset estimation algorithm is obtained.

[0034] In this embodiment, the influence of factors such as SOC, temperature and aging degree are considered in the parameter identification stage of the equivalent circuit model of the power battery, and on this basis, an SOC estimation algorithm based on the equivalent circuit model is established, and the first SOC estimation value SOC1_est(t) can be output in real time. Among them, the preset estimation algorithm involved in determining the first SOC estimation value of the power battery based on the equivalent circuit model of the power battery in step S102 is a conventional SOC estimation algorithm, which is not particularly limited and can be an open circuit method, an ampere-hour integration method, or other existing algorithms.

[0035] Among them, the power battery is in a charging state or a discharging state, the first current characteristic value and the first SOC estimation value are dynamically changing, and the first current characteristic value obtained in step S101 and the first SOC estimation value obtained in step S102 in this embodiment correspond to each other, and can be data obtained at the same time or in the same acquisition cycle.

[0036] Step S103: inputting the first current characteristic value and the first SOC estimation value into a pre-trained SOC error estimation model to obtain an SOC compensation value.

[0037] In this embodiment, the first current characteristic value I1_feature(t) and the first SOC estimation value SOC1_est(t) of the power battery are input into a pre-trained SOC error estimation model, and the SOC compensation value Err_SOC1(t) of the power battery is obtained according to the relationship Err_SOC(t)=f(I_feature(t), SOC_est(t)) in the SOC error estimation model.

[0038] Step S104: Determine an SOC output value according to the first SOC estimation value and the SOC compensation value.

[0039] Specifically, step S104 may be: adding the first SOC estimated value to the SOC compensation value to obtain the SOC output value. In this embodiment, the SOC real value of the power battery may be equal to the first SOC estimated value after the power battery compensation, so the first SOC estimated value after the power battery compensation is used as the SOC output value SOC1_out (t) with higher accuracy output in real time during driving of the electric vehicle, that is: SOC1_out(t)=SOC1_est(t)+Err_SOC1(t), Wherein, SOC1_out(t) is the SOC output value, SOC1_est(t) is the first SOC estimation value, and Err_SOC1(t) is the SOC compensation value.

[0040] like Figure 2 As shown, as a preferred solution of the embodiment of the present application, the pre-trained SOC error estimation model is pre-trained in the following manner: Step S201: obtaining a second current characteristic value in a power battery test environment; Step S202: obtaining a second SOC estimation value in a power battery test environment determined according to a preset estimation algorithm; Step S203: determining an SOC error value according to the actual SOC value of the power battery and the second SOC estimated value; Step S204: taking the second current characteristic value and the second SOC estimation value as characteristic quantities, taking the SOC error value as a label, and establishing a data pair set of the label and the characteristic quantity; Step S205: Based on the data pair set, a preset algorithm is used to train the relationship between the label and the feature quantity to obtain a pre-trained SOC error estimation model.

[0041] Further, step S201 may include: under the working condition test of the power battery, calculating a second current characteristic value of the power battery according to the real-time current.

[0042] In step S201, the second current characteristic value of the power battery calculated based on the real-time current may further include: performing a standard HPPC test on the equivalent circuit model of the power battery to extract the time constant that characterizes the charge transfer process; and calculating the second current characteristic value of the power battery based on the real-time current and the time constant. In particular, under the operating condition test of the power battery, the establishment method of the equivalent circuit model of the power battery can refer to the establishment method of the power battery in the application stage, which will not be described in detail here. In addition, the time constant extracted from the power battery under the operating condition test can be used as the time constant used by the power battery in the application stage.

[0043] Among them, the full name of the HPPC test is "Hybrid Pulse Power Characterization Test", which is a test method used to evaluate the performance of power batteries and is widely used in battery research and application fields. The standard HPPC (Hybrid Pulse Power Characterization) test refers to the HPPC test performed in accordance with certain specifications, processes and requirements to ensure that the test results are comparable, accurate and reliable. For example, in the performance evaluation of electric vehicle power batteries, standard HPPC tests are usually performed based on relevant industry standards or standard processes formulated within the company to understand the battery's power output characteristics, internal resistance changes and other performance parameters under different states (such as different states of charge, temperature, etc.).

[0044] In this embodiment, under the working condition test of the power battery, the power battery is fitted into an N-order RC equivalent circuit model, where N is a positive integer, and a standard HPPC test is performed on the N-order RC equivalent circuit model to extract the time constant tau characterizing the charge transfer process. The real-time current I2(t) of the power battery is filtered according to the time constant tau, wherein the value of the key parameter filtering time is the same as the value of the time constant tau, thereby obtaining the second current characteristic value I2_feature(t) of the power battery.

[0045] It should be noted that in actual filtering processing, if a filtering time is set, and this filtering time is numerically the same as the time constant of the filtering circuit, then in this specific case, their values ​​can be the same, but the concepts are essentially different. For example, in an RC low-pass filter circuit, if the filtering time is set to 5 times the time constant τ, then the filtering time is 5τ, which is different from the simple time constant τ.

[0046] Preferably, the preset algorithm includes a regression algorithm. In theory, the preset algorithm can also be a machine learning algorithm, but machine learning algorithms require a larger data set size and a larger amount of calculation. In contrast, a regression algorithm requires a smaller data set and thus a smaller amount of calculation.

[0047] In step S205, a preset algorithm is used to train the relationship between the label and the feature quantity, including: using a regression algorithm to train the relationship between the label and the feature quantity to obtain the potential relationship between the label and the feature quantity; quantifying the potential relationship to obtain the relationship between the label and the feature quantity. Among them, the regression algorithm may include a polynomial regression algorithm. In this embodiment, after obtaining a set of data pairs, each set of data pairs includes a feature quantity and a corresponding label, and a polynomial regression model is used to train the obtained data pair set. Compared with the use of a machine learning algorithm to train the relationship between the feature quantity and the corresponding label, which requires a larger data set specification and a larger amount of calculation, this embodiment can find the potential relationship between the feature quantity and the corresponding label through a limited number of operating condition tests. However, using a regression algorithm alone to obtain the potential relationship between the feature quantity and the corresponding label is relatively abstract to a certain extent. In order to support the prediction of the actual SOC error, it is necessary to further quantify the potential relationship to obtain the relationship between the feature quantity and the corresponding label, that is, to obtain a pre-trained SOC error estimation model.

[0048] Among them, the working condition test includes working condition actual measurement and simulation test. In the working condition test stage of the power battery, the real SOC value of the power battery is known, and the SOC error value corresponds to the SOC compensation value. Therefore, the SOC error value Err_SOC2 (t) is equal to the real SOC value SOC2_real (t) minus the second SOC estimated value SOC2_est (t), and the SOC error value Err_SOC2 (t) is used as a label, and correspondingly, the second current characteristic value I2_feature (t) and the second SOC estimated value SOC2_est (t) are used as feature quantities, and then a data pair set of labels and feature quantities can be established. In this embodiment, the SOC error estimation model to be trained can be trained using a polynomial regression algorithm based on the data pair set of the power battery to obtain a pre-trained SOC error estimation model. After obtaining the pre-trained SOC error estimation model, the corresponding SOC compensation value Err_SOC1(t) can be predicted according to the first current characteristic value I1_feature(t) of the power battery and the first SOC estimation value SOC1_est(t) corresponding to the first current characteristic value I1_feature(t) during vehicle driving, and the first SOC estimation value SOC1_est(t) can be compensated by the predicted SOC compensation value Err_SOC1(t) to obtain a more accurate SOC output value SOC1_out(t).

[0049] Among them, the way of calculating the current characteristic value in step S101 and step S102 is basically the same, the difference is that the time constant used by the power battery during vehicle driving is derived from the time constant extracted in step S102. The way of calculating the SOC estimation value in step S102 and step S202 is also basically the same. Then, using the pre-trained SOC error estimation model to predict the first SOC estimation value of the power battery, the prediction result is more accurate.

[0050] The present application is described in detail below with reference to an embodiment.

[0051] like Figure 3 As shown, the embodiment of the present application can be essentially divided into two stages. The first stage can be the working condition test stage of the power battery, and the real SOC value of the first stage is known; the second stage can be the application stage of the power battery in the electric vehicle, and the real SOC value of the second stage is unknown. In some specific scenarios, the power battery can be installed in the electric vehicle after the working condition test, so the power batteries in different stages can be the same, or different power batteries with the same specifications.

[0052] like Figure 4As shown, during the working condition test phase, the power battery can obtain the second current characteristic value I2_feature(t), the second SOC estimation value SOC2_est(t) and the SOC error value Err_SOC2(t), etc. The number of power batteries is not limited to the same one, and can be multiple power batteries of the same specifications. At this time, the power battery has not been installed on the electric vehicle and is still in the laboratory test phase. The power battery is tested under working conditions. First, an N-order RC equivalent circuit model is established. The parameter factors of SOC, temperature and aging degree are considered in the parameter identification stage of the equivalent circuit model of the power battery, that is, the current parameter is not considered. On the one hand, a SOC estimation algorithm based on the equivalent circuit model of the power battery is established on this basis to output the second SOC estimation value SOC2_est(t) of the power battery in real time. On the other hand, a standard HPPC test is performed on the N-order RC equivalent circuit model to extract the time constant that characterizes the charge transfer process. The real-time current I2(t) of the power battery is filtered according to the time constant. The key parameter filtering time takes the same value as the time constant, and then the second current characteristic value I2_feature(t) of the power battery is obtained. Then, the SOC error value Err_SOC2(t) can be determined based on the actual measurement or simulation test of the power battery under working conditions. The error between the second SOC estimation value SOC2_est(t) and the real value SOC2_real(t) of the power battery can be solved to obtain the corresponding SOC error value Err_SOC2(t) of the power battery. Finally, by taking the second current characteristic value I2_feature(t) and the second SOC estimation value SOC2_est(t) as feature quantities and the SOC error value Err_SOC2(t) as a label, a series of data pairs of feature quantities and labels can be obtained, and then a data pair set of the polynomial regression model is formed. The polynomial regression algorithm is used to obtain the potential relationship between the feature quantities and the labels, and then quantified into a specific relationship Err_SOC(t)=f(I_feature(t), SOC_est(t)), which is the pre-trained SOC error estimation model.

[0053] The power battery is installed on the electric vehicle. When the electric vehicle is driving, the first current characteristic value I1_feature(t) can be calculated according to the real-time current I1(t), and the first OSC estimation value SOC1_est(t) of the power battery can be solved according to the existing SOC estimation algorithm. The existing SOC estimation algorithm can be the SOC estimation algorithm of the power battery in the working condition test phase, and then combined with the pre-trained SOC error estimation model Err_SOC(t)=f(I_feature(t), SOC_est(t)) to obtain the compensation value Err_SOC1(t). Finally, the output value SOC1_out(t) is output according to the compensation value Err_SOC1(t) of the power battery and the first OSC estimation value SOC1_est(t), wherein SOC1_out(t)=SOC1_est(t)+Err_SOC1(t).

[0054] In summary, the embodiment of the present application adopts a combination of an equivalent circuit model and data drive. First, the influence of SOC, temperature, and aging factors are considered in the equivalent circuit model parameter identification stage, and on this basis, an SOC estimation algorithm based on an equivalent circuit model is established to obtain the SOC estimation value of the power battery, and the current characteristic value is obtained in combination with the equivalent circuit model. Secondly, the first SOC estimation value and the first current characteristic value of the power battery are input into the pre-trained SOC error estimation model to predict the SOC compensation value of the power battery, and the SOC error estimation model uses data drive quantification to obtain the relationship between the second current characteristic value, the second SOC estimation value, and the SOC compensation value. Finally, the first SOC estimation value is compensated in real time by the predicted SOC compensation value of the power battery, thereby obtaining a more accurate SOC output value. According to the method for determining the battery SOC of the embodiment of the present application, the estimation accuracy of the battery SOC can be improved to 2~3%, and is not limited to the degree of change of the current, which effectively reduces the influence of the current factor on the battery SOC estimation.

[0055] Corresponding to the aforementioned application function implementation method embodiment, the present application also provides a device for determining battery SOC, an electronic device and corresponding embodiments.

[0056] Figure 5 It is a schematic diagram of the structure of a device for determining battery SOC shown in an embodiment of the present application.

[0057] like Figure 5 As shown, an embodiment of the present application provides a device for determining a battery SOC, including: A first acquisition module 501 is used to acquire a first current characteristic value of a power battery during vehicle driving; A second acquisition module 502 is used to acquire a first SOC estimation value of the power battery during vehicle driving determined according to a preset estimation algorithm, wherein the preset estimation algorithm is determined based on an equivalent circuit model of the power battery, and the equivalent circuit model does not include current parameters as a factor considered in a parameter identification stage; A model processing module 503 is used to input the first current characteristic value and the first SOC estimation value into a pre-trained SOC error estimation model to obtain an SOC compensation value; The compensation output module 504 is used to determine the SOC output value according to the first SOC estimation value and the SOC compensation value.

[0058] Further, the compensation output module 504 may include an adder, and the adder is used to add the first SOC estimation value and the SOC compensation value to obtain the SOC output value.

[0059] Furthermore, the first acquisition module 501 may include a first calculation unit, and the first calculation unit is used to obtain a first current characteristic value of the power battery calculated according to the real-time current during the driving process of the vehicle.

[0060] Furthermore, the device also includes: A model training module is used to obtain a second current characteristic value in a power battery test environment; obtain a second SOC estimation value in the power battery test environment determined according to a preset estimation algorithm; determine an SOC error value according to the SOC true value of the power battery and the second SOC estimation value; use the second current characteristic value and the second SOC estimation value as feature quantities, and use the SOC error value as a label to establish a data pair set of labels and feature quantities; based on the data pair set, use a preset algorithm to train the relationship between the label and the feature quantity to obtain a pre-trained SOC error estimation model.

[0061] Furthermore, the model training module may include a second calculation unit, which is used to calculate a second current characteristic value of the power battery based on the real-time current under the working condition test of the power battery.

[0062] The second calculation unit can be further used to perform a standard HPPC test on the equivalent circuit model of the power battery to extract a time constant characterizing the charge transfer process; and calculate a second current characteristic value of the power battery based on the real-time current and the time constant.

[0063] Furthermore, the preset algorithm may include a regression algorithm.

[0064] Furthermore, the model training module may include a training unit and a quantification unit; the training unit is used to train the relationship between the label and the feature quantity using a regression algorithm to obtain the potential relationship between the label and the feature quantity; the quantification unit is used to quantify the potential relationship to obtain the relationship between the label and the feature quantity.

[0065] Regarding the device in the above embodiment, the specific manner in which each module performs the operation has been described in detail in the embodiment of the method, and will not be elaborated again here.

[0066] Figure 6 It is a schematic diagram of the structure of a car shown in an embodiment of the present application.

[0067] See also Figure 6 , the car 600 includes a memory 601 and a processor 602 .

[0068] The processor 602 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or the processor may be any conventional processor, etc.

[0069] The memory 601 may include various types of storage units, such as system memory, read-only memory (ROM), and permanent storage devices. Among them, ROM can store static data or instructions required by the processor 602 or other modules of the computer. The permanent storage device may be a readable and writable storage device. The permanent storage device may be a non-volatile storage device that does not lose the stored instructions and data even after the computer is powered off. In some embodiments, the permanent storage device uses a large-capacity storage device (such as a magnetic or optical disk, flash memory) as a permanent storage device. In some other embodiments, the permanent storage device may be a removable storage device (such as a floppy disk, an optical drive). The system memory may be a readable and writable storage device or a volatile readable and writable storage device, such as a dynamic random access memory. The system memory may store some or all instructions and data required by the processor at run time. In addition, the memory 601 may include any combination of computer-readable storage media, including various types of semiconductor storage chips (such as DRAM, SRAM, SDRAM, flash memory, programmable read-only memory), and disks and / or optical disks may also be used. In some embodiments, the memory 601 may include a readable and / or writable removable storage device, such as a laser disc (CD), a read-only digital versatile disc (such as a DVD-ROM, a double-layer DVD-ROM), a read-only Blu-ray disc, an ultra-density optical disc, a flash memory card (such as an SD card, a mini SD card, a Micro-SD card, etc.), a magnetic floppy disk, etc. Computer-readable storage media do not include carrier waves and transient electronic signals transmitted wirelessly or wired.

[0070] The memory 601 stores executable codes, and when the executable codes are processed by the processor 602 , the processor 602 can execute part or all of the above-mentioned methods.

[0071] In addition, the method according to the present application may also be implemented as a computer program or a computer program product, which includes computer program code instructions for executing some or all of the steps in the above method of the present application.

[0072] Alternatively, the present application can also be implemented as a computer-readable storage medium (or non-transitory machine-readable storage medium or machine-readable storage medium) on which executable code (or computer program or computer instruction code) is stored. When the executable code (or computer program or computer instruction code) is executed by a processor of an electronic device (or server, etc.), the processor executes part or all of the steps of the above-mentioned method according to the present application.

[0073] The embodiments of the present application have been described above, and the above description is exemplary, not exhaustive, and is not limited to the disclosed embodiments. Many modifications and changes will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described embodiments. The selection of terms used herein is intended to best explain the principles of the embodiments, practical applications, or improvements to the technology in the market, or to enable other persons of ordinary skill in the art to understand the embodiments disclosed herein.

Claims

1. A method for determining battery SOC, characterized in that: include: Acquire a first current characteristic value of a power battery during vehicle driving; Obtaining a first SOC estimation value of the power battery during vehicle driving determined according to a preset estimation algorithm, wherein the preset estimation algorithm is determined based on an equivalent circuit model of the power battery, and the equivalent circuit model does not include current parameters as a factor considered in a parameter identification stage; Inputting the first current characteristic value and the first SOC estimation value into a pre-trained SOC error estimation model to obtain an SOC compensation value; An SOC output value is determined according to the first SOC estimation value and the SOC compensation value.

2. The method according to claim 1, characterized in that The determining the SOC output value according to the first SOC estimation value and the SOC compensation value includes: The first SOC estimation value is added to the SOC compensation value to obtain an SOC output value.

3. The method according to claim 1, characterized in that The obtaining of a first current characteristic value of a power battery during vehicle driving includes: A first current characteristic value of the power battery calculated according to the real-time current during the driving of the vehicle is obtained.

4. The method according to claim 1, characterized in that The pre-trained SOC error estimation model is pre-trained in the following manner: Acquire a second current characteristic value in a power battery test environment; Obtaining a second SOC estimation value in a power battery test environment determined according to a preset estimation algorithm; Determining an SOC error value according to the actual SOC value of the power battery and the second SOC estimated value; Taking the second current characteristic value and the second SOC estimation value as characteristic quantities, taking the SOC error value as a label, and establishing a data pair set of the label and the characteristic quantity; Based on the data pair set, a preset algorithm is used to train the relationship between the label and the feature quantity to obtain the pre-trained SOC error estimation model.

5. The method according to claim 4, characterized in that The preset algorithm includes a regression algorithm; or, The obtaining of the second current characteristic value in the power battery test environment includes: A second current characteristic value of the power battery is obtained by calculating the real-time current under the operating condition test of the power battery.

6. The method according to claim 5, characterized in that The second current characteristic value of the power battery calculated according to the real-time current includes: Performing a standard HPPC test on the equivalent circuit model of the power battery to extract a time constant representing the charge transfer process; A second current characteristic value of the power battery is calculated based on the real-time current and the time constant.

7. The method according to claim 4, characterized in that The using a preset algorithm to train the relationship between the label and the feature quantity includes: Using a regression algorithm to train the relationship between the label and the feature quantity to obtain a potential relationship between the label and the feature quantity; The potential relationship is quantified to obtain the relationship between the label and the feature quantity.

8. A device for determining a battery SOC, characterized in that: include: A first acquisition module, used to acquire a first current characteristic value of a power battery during the driving process of the vehicle; A second acquisition module is used to acquire a first SOC estimation value of the power battery during vehicle driving determined according to a preset estimation algorithm, wherein the preset estimation algorithm is determined based on an equivalent circuit model of the power battery, and the equivalent circuit model does not include current parameters as a factor considered in a parameter identification stage; A model processing module, configured to input the first current characteristic value and the first SOC estimation value into a pre-trained SOC error estimation model to obtain an SOC compensation value; The compensation output module is used to determine an SOC output value according to the first SOC estimation value and the SOC compensation value.

9. The device according to claim 8, characterized in that The device also includes: A model training module is used to obtain a second current characteristic value in a power battery test environment; obtain a second SOC estimation value in the power battery test environment determined according to a preset estimation algorithm; determine an SOC error value according to the SOC true value of the power battery and the second SOC estimation value; use the second current characteristic value and the second SOC estimation value as feature quantities, and use the SOC error value as a label to establish a data pair set of the label and the feature quantity; based on the data pair set, use a preset algorithm to train the relationship between the label and the feature quantity to obtain the pre-trained SOC error estimation model.

10. An automobile, characterized in that: include: processor; as well as A memory having executable codes stored thereon, which, when executed by the processor, causes the processor to execute the method according to any one of claims 1 to 7.

11. A computer-readable storage medium, characterized in that: An executable code is stored thereon, and when the executable code is executed by a processor of an electronic device, the processor is caused to execute the method as claimed in any one of claims 1 to 7.