Method and system for estimating state of health of power battery and medium

Abstract

The application provides a power battery state of health estimation method, system and medium, the method comprises: acquiring real-time operation data of a vehicle power battery system, the real-time operation data at least includes attenuation characteristics generated in a narrow segment of the power battery system charging process; input the real-time operation data of the vehicle power battery system into the pre-established relationship model of the attenuation characteristic parameters and the state of health of the power battery charging process, the state of health includes the capacity and the remaining service life of the power battery; based on the attenuation characteristics and the relationship model, the state of health estimation of the power battery is completed on the vehicle, and the capacity and the remaining life prediction are carried out. The application realizes the estimation of SOH and the prediction of the remaining life on the actual vehicle, saves a lot of time, effort and money; realizes that the estimation of SOH can be completed only by using part of the charging segment, improves the feasibility and efficiency of the user to estimate SOH; realizes the extraction and analysis of the attenuation factor, and provides user use guidance.

Description

Technical Field

[0001] This invention relates to the field of power battery technology, and in particular to a method, system and medium for estimating the health status of a power battery. Background Technology

[0002] As power batteries age with increasing years of use and mileage, their capacity decreases, their state of health (SOH) declines, and their performance irreversibly deteriorates, affecting user experience. However, neither users nor automakers can directly and accurately estimate the degree of battery aging, causing difficulties in maintenance, troubleshooting, and other aspects.

[0003] Existing technologies can collect relatively accurate battery capacity data from the actual operation of electric vehicles; simultaneously, different battery degradation models can be established using big data to estimate the current health status. However, they cannot predict the remaining battery life and are also difficult to use to guide user behavior, specifically as follows:

[0004] 1. Traditional SOH estimation and remaining life prediction for actual vehicles often require sending the vehicle to a repair center, which takes a lot of time, effort and money, greatly increasing the maintenance costs for users and manufacturers.

[0005] 2. For lithium iron phosphate batteries, during the charging process, the relationship between the cell voltage and the state of charge (SOC) exhibits a plateau period (e.g., ...). Figure 1 As shown in the figure, SOH estimation during the plateau period has a large error. Based on the previously proposed capacity calculation method, data collection needs to start before entering the plateau period (i.e., when SOC is below 30%) and stop after the end of the plateau period (i.e., when SOC reaches 100%). Due to different user habits, especially for conservative users, who often do not let the vehicle battery SOC drop below 30% before charging, this method often cannot cover enough usage scenarios and cannot be practically applied in many cases.

[0006] 3. User habits often have a significant impact on battery life. Currently, user behavior analysis cannot guide users to use electric vehicles correctly, thus affecting battery life. Summary of the Invention

[0007] The main objective of this invention is to provide a method, system, and medium for estimating the state of health (SOH) of a power battery, aiming to achieve SOH estimation and remaining life prediction in actual vehicles, saving significant time, effort, and money; enabling SOH estimation using only a portion of the charging segment (e.g., the segment from 50% to 70% SOC), improving the feasibility and efficiency of SOH estimation for users; and enabling the extraction and analysis of degradation factors, providing user guidance, such as providing optimal charging ranges.

[0008] To achieve the above objectives, this invention proposes a method for estimating the health status of a power battery, the method comprising the following steps:

[0009] Step S10: Obtain real-time operating data of the vehicle power battery system, wherein the real-time operating data includes at least the attenuation characteristics generated during a narrow segment of the charging process of the power battery system;

[0010] Step S20: Input the real-time operating data of the vehicle power battery system into a pre-established relationship model between the attenuation characteristic parameters of the power battery charging process and the health status, wherein the health status includes the capacity and remaining service life of the power battery.

[0011] Step S30: Based on the degradation characteristics and the relationship model, the health status of the power battery on the vehicle is estimated, and the capacity and remaining lifespan are predicted.

[0012] A further technical solution of the present invention is that, before obtaining the real-time operating data of the vehicle power battery system in step S10, the following is included:

[0013] A model is pre-established to show the relationship between the degradation characteristic parameters of the power battery during the charging process and its health status.

[0014] A further technical solution of the present invention is that the step of pre-establishing the relationship model between the degradation characteristic parameters of the power battery charging process and the health status includes:

[0015] The range of state of charge covered by the power battery system during charging is calculated from historical data. The attenuation characteristic parameters of the charging segment in this range are extracted and machine learning is performed to establish a relational model for this range.

[0016] A further technical solution of the present invention is that the real-time operating data also includes the degradation factor of the power battery, and the step S10 is followed by:

[0017] Correlation analysis was performed between the attenuation factor and capacity attenuation to quantify the impact of different attenuation factors and provide users with usage guidance.

[0018] A further technical solution of the present invention is that the attenuation factor includes average daily driving mileage, average charging depth, average idle time, and user behavior.

[0019] A further technical solution of the present invention is that the attenuation characteristics include driving mileage, service life, and power consumption increment.

[0020] To achieve the above objectives, the present invention also proposes a power battery health status estimation system, the system including a memory, a processor, and a power battery health status estimation program stored on the processor, wherein the power battery health status estimation program is executed by the processor to perform the steps of the method described above.

[0021] To achieve the above objectives, the present invention also proposes a computer-readable storage medium storing a power battery health status estimation program, wherein the power battery health status estimation program is executed by a processor to perform the steps of the method described above.

[0022] The beneficial effects of the power battery health status estimation method, system, and medium of the present invention are:

[0023] (1) This invention enables the estimation of SOH and prediction of remaining life on actual vehicles, saving a lot of time, effort and money.

[0024] (2) The present invention enables the estimation of SOH using only a portion of the charging segment (e.g., the segment from SOC to 70%), thereby improving the feasibility and efficiency of SOH estimation for users.

[0025] (3) This invention realizes the extraction and analysis of the attenuation factor and provides user guidance, such as providing the optimal charging range. Attached Figure Description

[0026] Figure 1 This is a schematic diagram showing the relationship between single-cell voltage and SOC during the charging process of a lithium iron phosphate battery;

[0027] Figure 2 This is a flowchart illustrating a preferred embodiment of the power battery health status estimation method of the present invention;

[0028] Figure 3 This is a schematic diagram of SOH estimation and remaining lifetime prediction based on the charging narrow-range attenuation characteristics;

[0029] Figure 4 This is a schematic diagram illustrating the extraction of attenuation characteristics and attenuation factors;

[0030] Figure 5 This is a schematic diagram of attenuation analysis based on the attenuation factor.

[0031] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. Detailed Implementation

[0032] It should be understood that the specific embodiments described herein are for illustrative purposes only and are not intended to limit the scope of the invention.

[0033] This invention proposes a method for estimating the health status of a power battery. The main technical solution adopted in this invention is as follows:

[0034] (1) By collecting and processing big data on vehicle operation, the State of Health (SOH) and various related degradation features extracted during the charging process (such as mileage, years of use, and energy increment) can be obtained. Based on big data, machine learning can be used to establish a relationship model between the degradation feature parameters of the charging process and the SOH. For the same type of vehicle, the SOH can be estimated directly on the vehicle based on the new degradation feature parameters collected during the charging process, and the remaining life can be predicted.

[0035] (2) For power batteries, such as lithium iron phosphate batteries, in the process of machine learning, only a narrow segment of the charging process is selected, and the decay characteristics generated by the segment (e.g., the increase in single cell capacity) are used to find the relationship between the decay characteristics and the capacity, and to establish an effective SOH estimation model.

[0036] (3) By collecting and processing big data on vehicle operation, we can obtain SOH and various degradation factors related to battery aging (such as average daily mileage, average charging depth, etc.). Analyzing the relationship between different degradation factors and battery aging can provide users with usage guidance.

[0037] Specifically, such as Figure 2 As shown, a preferred embodiment of the power battery health status estimation method of the present invention includes the following steps:

[0038] Step S10: Obtain real-time operating data of the vehicle's power battery system, wherein the real-time operating data includes at least the attenuation characteristics generated during a narrow segment of the power battery system's charging process.

[0039] Step S20: Input the real-time operating data of the vehicle power battery system into a pre-established relationship model between the degradation characteristic parameters of the power battery charging process and the health status. The health status includes the capacity and remaining service life of the power battery.

[0040] Step S30: Based on the degradation characteristics and the relationship model, the health status of the power battery on the vehicle is estimated, and the capacity and remaining lifespan are predicted.

[0041] Furthermore, in this embodiment, step S10, before acquiring the real-time operating data of the vehicle's power battery system, includes:

[0042] A model is pre-established to show the relationship between the degradation characteristic parameters of the power battery during the charging process and its health status.

[0043] Furthermore, in this embodiment, the step of pre-establishing the relationship model between the degradation characteristic parameters of the power battery charging process and its health status includes:

[0044] The range of state of charge covered by the power battery system during charging is calculated from historical data. The attenuation characteristic parameters of the charging segment in this range are extracted and machine learning is performed to establish a relational model for this range.

[0045] Furthermore, in this embodiment, the real-time operating data also includes the degradation factor of the power battery, and after step S10, the following is also included:

[0046] Correlation analysis was performed between the attenuation factor and capacity attenuation to quantify the impact of different attenuation factors and provide users with usage guidance.

[0047] Furthermore, in this embodiment, the attenuation factor includes average daily mileage, average charging depth, average idle time, and user behavior.

[0048] Furthermore, in this embodiment, the attenuation characteristics include driving mileage, years of use, and battery charge increment.

[0049] The following combination Figures 3 to 5 The present invention will further elaborate on the method for estimating the health status of power batteries.

[0050] like Figure 3 As shown, the SOH estimation and remaining lifetime prediction based on the charging narrow-range attenuation characteristics in this invention specifically adopt the following scheme.

[0051] (1) The vehicle power battery system collects signals such as time, current, temperature, SOC, and cell voltage through sensors and uploads them to the big data cloud computing platform;

[0052] (2) The historical data is cleaned and preprocessed, and the historical capacity information is calculated by the ampere-hour integration method. Through the analysis of the historical data, multiple battery degradation characteristics are calculated, such as: capacity degradation rate, average daily driving mileage, average charging depth, average charging frequency, etc.

[0053] (3) For multidimensional decay features, unsupervised learning algorithms, such as K-means, are used to cluster them to obtain multiple decay pattern classifications;

[0054] (4) For different attenuation mode types, health features and calculated capacity are extracted based on charging data in narrow range. Artificial intelligence algorithms such as Gaussian process regression and convolutional neural network are used to establish SOH estimation models respectively, and finally form a library of SOH estimation models under multiple attenuation modes.

[0055] (5) For the vehicle under test, firstly, based on its historical operating data, the attenuation characteristics are identified and the attenuation mode is classified; secondly, based on the narrow interval, the health characteristics are extracted, and the model under the attenuation mode is used to estimate the SOH and predict the remaining life.

[0056] like Figures 4 to 5 As shown, the attenuation analysis process based on the attenuation factor in this invention is as follows:

[0057] (1) The vehicle power battery system collects signals such as time, current, temperature, SOC, and cell voltage through sensors and uploads them to the big data cloud computing platform;

[0058] (2) The historical data is cleaned and preprocessed, and the historical capacity information is calculated by the ampere-hour integral method. By analyzing user behavior (including driving behavior, charging behavior, and idle behavior), multiple battery degradation factors are calculated, such as capacity degradation rate, average daily driving mileage, average charging depth, and average charging frequency.

[0059] (3) Calculate the historical capacity decay of the vehicle and analyze the correlation between the decay and the multidimensional decay factor;

[0060] (4) Based on the correlation between different attenuation factors and capacity attenuation, the attenuation factors that contribute significantly to capacity attenuation of the vehicle are obtained, the degree of influence of different user behaviors on capacity attenuation is clarified, and vehicle usage suggestions are provided to users.

[0061] The power battery state of health estimation method of this invention can calculate battery capacity and degradation characteristic parameters from historical data through statistical analysis of big data from real vehicle operation, and establish a machine learning-based SOH estimation model. Based on this model, accurate SOH estimation and remaining life prediction can be achieved for any vehicle of the same type.

[0062] The power battery health state estimation method of this invention can identify the range of SOC (State of Charge) covered by most users during charging based on big data statistics of user charging habits. For example, during the charging process of most users, the SOC will cover a segment from 50% to 70% (e.g., ...). Figure 1 (As shown in the area enclosed by the dashed line), the attenuation characteristic parameters of this charging segment are extracted, machine learning is performed, and a SOH estimation model for this segment is established to ensure that most users can estimate the SOH based on the charging process.

[0063] The power battery health status estimation method of this invention can calculate the degradation factors such as average daily mileage, average charging depth, and average resting time from historical data through statistical analysis of big data from actual vehicle operation. The degradation factors are then correlated with capacity degradation to further quantify the impact of different degradation factors and provide users with usage guidance.

[0064] The technical innovation of this invention's method for estimating the health status of power batteries lies in:

[0065] (1) A method for estimating the capacity of a real vehicle power battery and a method for predicting its remaining life based on its attenuation characteristics are proposed.

[0066] (2) Establish independent SOH estimation models for different types of vehicles to improve the estimation accuracy of SOH for a single vehicle.

[0067] (3) Extract different attenuation factors related to use, further analyze the impact of attenuation factors on battery aging, and provide users with usage guidance.

[0068] The beneficial effects of the power battery health status estimation method of the present invention are:

[0069] (1) This invention enables the estimation of SOH and prediction of remaining life on actual vehicles, saving a lot of time, effort and money.

[0070] (2) The present invention enables the estimation of SOH using only a portion of the charging segment (e.g., the segment from SOC to 70%), thereby improving the feasibility and efficiency of SOH estimation for users.

[0071] (3) This invention realizes the extraction and analysis of the attenuation factor and provides user guidance, such as providing the optimal charging range.

[0072] To achieve the above objectives, the present invention also proposes a power battery health status estimation system, the system including a memory, a processor, and a power battery health status estimation program stored on the processor. The power battery health status estimation program is executed by the processor to perform the steps of the method described above, which will not be repeated here.

[0073] To achieve the above objectives, the present invention also proposes a computer-readable storage medium storing a power battery health status estimation program. When the power battery health status estimation program is run by a processor, the steps of the method described above are executed, which will not be repeated here.

[0074] The above description is only a preferred embodiment of the present invention and does not limit the patent scope of the present invention. Any equivalent structural or procedural changes made based on the content of the present invention specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of the present invention.

Claims

1. A method for estimating the state of health of a power battery, characterized in that, The method comprises the following steps: a relationship model between a degradation characteristic parameter of a power battery charging process and a state of health is pre-established; In step S10, real-time operation data of a vehicle power battery system is acquired, wherein the real-time operation data at least comprises a degradation characteristic generated by a narrow segment in the charging process of the power battery system; wherein the narrow segment is a segment with a SOC from 50% to 70%, and the degradation characteristic comprises an increment of electric quantity; the power battery is a lithium iron phosphate battery; In step S20, the real-time operation data of the vehicle power battery system is input into the pre-established relationship model between the degradation characteristic parameter of the power battery charging process and the state of health, and the state of health comprises a capacity and a remaining service life of the power battery; In step S30, a state of health estimation of the power battery is completed on the vehicle based on the degradation characteristic and the relationship model, and a capacity and a remaining life prediction are performed.

2. The power cell state of health estimation method of claim 1, wherein, The step of pre-establishing the relationship model between the degradation characteristic parameter of the power battery charging process and the state of health comprises: An interval covered by a state of charge during charging of the power battery system is calculated from historical data, a degradation characteristic parameter of a charging segment in the interval is extracted for machine learning, and a relationship model of the interval is established.

3. The power cell state of health estimation method of claim 1, wherein, The real-time operation data further comprises a degradation factor of the power battery, and the step S10 further comprises: The degradation factor and capacity degradation are correlated for correlation analysis, an influence degree of different degradation factors is quantified, and use guidance is provided for a user.

4. The power cell state of health estimation method of claim 3, wherein, The degradation factor comprises a daily average driving distance, an average charging depth, and an average standing time.

5. A power cell state of health estimation system, characterized by, The system comprises a memory, a processor, and a power battery state of health estimation program stored on the processor, and the power battery state of health estimation program performs the steps of the method of any one of claims 1 to 4 when executed by the processor.

6. A computer-readable storage medium, characterized in that, The computer readable storage medium stores a power battery state of health estimation program, and the power battery state of health estimation program performs the steps of the method of any one of claims 1 to 4 when executed by the processor.

Images