Power battery consistency safety status assessment method

By extracting the historical charging data of new energy vehicles and unsupervised training, a power battery consistency safety status evaluation model was established, and the complexity of power battery status evaluation of new energy vehicles was solved, and rapid safety status quantification and risk warning were achieved. .

CN114879049BActive Publication Date: 2024-11-26CHINA AUTOMOTIVE ENG RES INST
View PDF 0 Cites 3 Cited by

Patent Information

Application Number
CN202210664642.6
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2022-06-13
Publication Date
2024-11-26
Estimated Expiration
2042-06-13

AI Technical Summary

Technical Problem

The state evaluation of power batteries in new energy vehicles is difficult to achieve through complex electrochemical mechanisms, and a method that does not consider coupling is needed to quickly quantify the safety status of the battery system in order to promptly detect potential risks.

Method used

By obtaining the charging data fragments in the historical running data, calculating the standard deviation and variance entropy characteristics of the single voltage, building a feature matrix and performing unsupervised training, establishing a consistent safety state quantization calculation model and a state evaluation alarm level model, and outputting a safety assessment Score and grade results.

Benefits of technology

It realizes rapid quantification and evaluation of the safety status of the power battery system, timely warning of potential risks, and avoids abnormal statuses from evolved into serious accidents.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN114879049B_ABST
    Figure CN114879049B_ABST
Patent Text Reader

Abstract

The present invention relates to the technical field of power battery evaluation, and specifically to a method for evaluating the consistent safety state of a power battery, including: extracting multiple charging data fragments that can reflect the state of a vehicle; calculating the standard deviation characteristics and variance entropy consistency characteristics of the single cell voltage of each charging data fragment to obtain a characteristic value; obtaining the number of times the charging process of the vehicle covers a target preset interval throughout its life cycle, combining the number of times covered with a first correction number and a second correction number to obtain a reference sample and an evaluation sample respectively; constructing the characteristic value into a characteristic matrix, and performing unsupervised training after processing to divide it into two categories to obtain a confusion matrix; constructing a quantitative calculation model for the consistent safety state of a power battery; constructing a state evaluation alarm level model, and outputting a level result to characterize the safety state of the power battery. The present invention can timely warn of potential risks of vehicles and prevent abnormal vehicle states from evolving into more serious accident risks.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of power battery evaluation, and in particular to a method for evaluating the consistent safety state of a power battery. Background Art

[0002] Compared with existing fuel-powered vehicles, new energy vehicles have a short development history. The development of various technologies of new energy vehicles is not as mature as that of fuel-powered vehicles, and the exploration and research of new energy vehicles and their power batteries is far less thorough than that of fuel-powered vehicles. The actual operation of new energy vehicles has the characteristics of changeable environment, complex scenarios, multi-dimensional, redundant, heterogeneous and strongly coupled data, which brings great challenges to the exploration and mining of the safety status of power vehicles contained in the operation big data, making it difficult to achieve the evaluation of the power battery status of new energy vehicles in strongly coupled complex systems through mechanism analysis.

[0003] Since new energy vehicles will upload operating data according to standard GB-32960 during their service, and voltage and other characterization signals are a characterization of the comprehensive state of the power battery, the degree of difference in the power battery's state during service can be further quantified by analyzing and mining the safety features of the power vehicle's historical operating data.

[0004] Therefore, in order to timely discover safety problems of power vehicles, there is an urgent need for a method that can describe the overall signal safety level of the power battery system by directly calculating the data characteristics of the signal without considering the complex and coupled electrochemical mechanisms, so as to quantify the safety status of the battery system and realize the rapid quantification and evaluation of the battery safety status, thereby guiding the power vehicle to provide safety warnings. Summary of the invention

[0005] The present invention aims to provide a method for evaluating the consistent safety status of a power battery, so as to quickly quantify and evaluate the safety status of the power battery.

[0006] The power battery consistency safety status assessment method in this scheme includes the following steps:

[0007] Step 1: Obtain historical operating data of the vehicle to be evaluated, and extract multiple charging data segments that can reflect the vehicle status from the historical operating data;

[0008] Step 2: Calculate the standard deviation feature and variance entropy consistency feature of the cell voltage of each charging data segment to obtain the feature value;

[0009] Step 3, obtaining the number of times the charging process of the vehicle covers the target preset interval throughout its life cycle, multiplying the number of times covered by the first parameter to obtain the first correction number, using the characteristic value of the first correction number in the charging data segment as the reference sample, multiplying the number of times covered by the second parameter to obtain the second correction number, using the characteristic value of the second correction number in the charging data segment as the evaluation sample;

[0010] Step 4: construct the eigenvalues ​​into a feature matrix, delete the eigenvalues ​​greater than the 99th percentile in the feature matrix, perform unsupervised training on the feature matrix using a preset algorithm, and divide it into two categories to obtain a confusion matrix;

[0011] Step 5: construct a power battery consistency safety state quantitative calculation model based on the confusion matrix, and calculate the consistency state safety assessment score through the consistency safety state quantitative calculation model;

[0012] Step six: construct a state assessment alarm level model based on the consistent state safety assessment score, and output the level result to characterize the safety state of the power battery with the level result.

[0013] The beneficial effects of this program are:

[0014] By slicing the charging data in the historical operation data of new energy vehicles, we obtain charging data fragments, complete charging feature extraction, and then combine the consistency state safety assessment method to obtain the assessment score, and finally output the state assessment alarm level to timely warn of potential risks of the vehicle and prevent the abnormal state of the vehicle from evolving into a more serious accident risk.

[0015] Furthermore, in the step 1, the battery OCV curve of the vehicle to be evaluated is obtained, and the charging data segment is extracted from the SOC interval on the OCV curve.

[0016] The beneficial effect is that the safety status of the vehicle can be accurately reflected while retaining the corresponding characteristics.

[0017] Furthermore, in step 2, the calculation formula of the standard deviation feature is:

[0018]

[0019] The calculation formula of variance entropy consistency feature is:

[0020]

[0021] The beneficial effects are: through the calculation of standard deviation characteristics, the discreteness of data can be analyzed quickly and accurately, and through the definition of variance entropy used to describe the degree of system chaos, the voltage difference between each cell of the power battery system can be described, that is, the consistency is good or bad. The smaller the variance entropy, the smaller the cell voltage difference, the better the consistency, and the safer the battery system.

[0022] Furthermore, in the step three, the number of coverages is represented as n, the first parameter is represented as α, the second parameter is represented as β, the first correction number is n*α, and the second correction number is n*β.

[0023] The beneficial effect is that by representing each data in characters, the calculation of the quantization process is made smoother.

[0024] Furthermore, the value range of the first parameter α is: (0.05, 0.5), and the value range of the second parameter is: (0.01, 0.2).

[0025] The beneficial effect is that by limiting the value range of the first parameter and the second parameter, the number of features in the data set can be reduced, the amount of calculation can be reduced, and the first parameter and the second parameter are taken in different ranges, which can increase the difference between the reference sample and the evaluation data.

[0026] Furthermore, in step 4, the feature matrix is ​​represented as X, and the confusion matrix obtained after unsupervised training of the feature matrix is:

[0027]

[0028] The beneficial effect is that by calculating the confusion matrix, the feature matrix can be simply classified, thereby improving the processing speed of historical operation data.

[0029] Furthermore, in step 5, the consistent safety state quantitative calculation model is:

[0030]

[0031] Among them, S represents the consistency status security assessment score of the evaluation sample, #(X 0_1 )、#(X0)、#(X 1_1 ), #(X1) respectively represent (X 0_1 )、X0、X 1_1 , the number of X1 samples, represents the mean of the variance characteristics in the reference sample, represents the mean of the variance entropy consistency feature in the reference sample, represents the variance characteristic mean in the evaluation sample, represents the variance entropy consistency feature mean in the evaluation sample, μ and Represents weight, and the value range of weight is [0,1].

[0032] The beneficial effects are: through the quantification of the number of samples, the consistency safety degree of the reference data and the data to be evaluated can be quantitatively described; through the quantification of the variance feature mean and the variance entropy consistency feature mean, the safety degree of the two segments of data can be measured, thereby quantifying the safety assessment score and improving the accuracy of the assessment.

[0033] Further, in step 6, the state assessment alarm level model is:

[0034]

[0035] Among them, alarmLevel represents the alarm level obtained by evaluating the status of the data to be evaluated. Level 0, 1, 2, and 3 alarms correspond to no, low, medium, and high risks, respectively. r = 1-S represents the vehicle risk. r1, r2, and r3 correspond to the risk alarm thresholds, respectively, and r1 <r2<r3。

[0036] The beneficial effect is that by quantitatively evaluating the consistent safety status of the power battery, risk factors can be more intuitively discovered from a large amount of historical operating data analysis. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 It is a flowchart of an embodiment of a method for evaluating a consistent safety state of a power battery according to the present invention;

[0038] Figure 2 It is an OCV curve diagram in an embodiment of a method for evaluating a consistent safety state of a power battery of the present invention;

[0039] Figure 3 This is a voltage data diagram of a single charging data segment in an embodiment of the power battery consistency safety state assessment method of the present invention. DETAILED DESCRIPTION

[0040] The following is further explained in detail through specific implementation methods.

[0041] Example

[0042] Power battery consistency safety status assessment method, such as Figure 1 As shown, the following steps are included:

[0043] Step 1: Obtain the historical operation data of the vehicle to be evaluated, and obtain the battery OCV curve of the vehicle to be evaluated. The OCV curve is a reference tool for power batteries. Each battery is given by the manufacturer when it is launched on the market. Extract multiple charging data segments that can reflect the vehicle status from the historical operation data. For example, take 100 charging data segments as an example. The voltage data of a single charging data segment is as follows: Figure 3As shown in FIG. 1 , the preset time period when the vehicle just starts to be put into service is defined as the reference state of the vehicle safety state, that is, it is assumed that all new vehicles are good, and the charging data fragment is extracted for the SOC interval on the OCV curve, that is, the extracted charging data fragment contains the charging data from 85% to 95% of the power. The purpose of selecting the SOC interval according to the OCV curve is: within this SOC interval, the voltage has a sudden change adjustment, which is equivalent to giving a typical excitation, and then the signal change based on this excitation can show the battery state, such as Figure 2 As shown in the box in .

[0044] Step 2: Calculate the standard deviation feature and variance entropy consistency feature of the single cell voltage of each charging data segment to obtain two eigenvalues. Then, after calculating 100 charging data segments, a 100×2 eigenvalue matrix can be obtained. The single cell voltage is the voltage value of the single cell, and the single cell voltage is represented by C. The calculation formula of the standard deviation feature is:

[0045] Among them, x ij is the single cell voltage value of the jth battery cell at the i-th moment, x i is the average value of all monomer voltages at time i;

[0046] The calculation formula of variance entropy consistency feature is:

[0047] Among them, E j is the square of the cell voltage.

[0048] Step three, obtain the number of coverage times of the charging process that covers the target preset interval during the entire life cycle of the vehicle, the coverage number is denoted as n, multiply the coverage number by the first parameter to obtain the first correction number, the first parameter is denoted as α, the value range of the first parameter α is: (0.05, 0.5), the first correction number is n*α, the characteristic value of the first correction number in the charging data segment is used as the reference sample, the coverage number is multiplied by the second parameter to obtain the second correction number, the second parameter is denoted as β, the value range of the second parameter is: (0.01, 0.2), the second correction number is n*β, and the characteristic value of the second correction number in the charging data segment is used as the evaluation sample.

[0049] The "before" in step three refers to the acquisition of the charging data fragments according to the starting point of time, and the "after" refers to the acquisition of the charging data fragments from the end point of time. For example, the acquired charging data fragments are 100, that is, the data obtained from 100 charging times, and the data fragments are arranged in chronological order. The first correction number can be set to 20%, and the second correction number can be set to 10%. The data fragments of the first correction number are the first 20 of the 100 data fragments as the control group, and the data fragments of the second correction number are the last 10 of the 100 data fragments as the evaluation samples.

[0050] Step 4: construct the eigenvalues ​​into a feature matrix, represented as X, delete the eigenvalues ​​greater than the 99th percentile in the feature matrix, perform unsupervised training on the feature matrix using a preset algorithm, the preset algorithm is the existing K-Means algorithm, divide X into two categories, namely category 0 and category 1, and obtain the confusion matrix, which is:

[0051] Where X0 is the reference sample and X1 is the evaluation sample.

[0052] Among them, X 0_0 It means that the reference sample group 0 is divided into the 0th category, X 0_1 It means that the reference sample group 0 is divided into the first category, X 1_0 Indicates that one group of the evaluation samples is divided into class 0, X 1_1 Indicates that a group of evaluation samples is classified into the first category.

[0053] Step 5: Based on the reference samples and evaluation samples in the confusion matrix, a power battery consistency safety state quantitative calculation model is constructed according to the confusion matrix. The consistency state safety assessment score is calculated by the consistency safety state quantitative calculation model. The consistency safety state quantitative calculation model is:

[0054]

[0055] Among them, S represents the consistency status security assessment score of the evaluation sample, #(X 0_1 )、#(X0)、#(X 1_1 ), #(X1) respectively represent (X 0_1 )、X0、X 1_1 , the number of X1 samples, represents the mean of the variance characteristics in the reference sample, represents the mean of the variance entropy consistency feature in the reference sample, represents the variance characteristic mean in the evaluation sample, represents the variance entropy consistency feature mean in the evaluation sample, μ and Represents the weight, and the value range of the weight is [0, 1].

[0056] Initially, the state of the vehicle battery is good. Equivalent to taking the two eigenvalue of the reference sample as the initial origin, the closer the two eigenvalues calculated from the evaluation sample are to the initial origin, the closer the state of the vehicle battery is to the initial state, and the better the safety of the vehicle. As the vehicle is driven and used, the vehicle's battery will become worse and worse. Use the data during the subsequent vehicle use process as the evaluation sample for classification, and then calculate the corresponding score. That is, the higher the proportion of classifying the first group of evaluation samples into the 0th group represented by (X 0_1 ), the closer the evaluation sample is to the reference sample, the smaller the difference and separation degree between the evaluation sample and the reference sample, and the higher the safety of the vehicle.

[0057] Step Six, construct a state evaluation alarm level model based on the consistency state safety evaluation score, that is, subtract the consistency state safety evaluation score from the total number to obtain the vehicle risk, and output the level result. That is, the vehicle risk is expressed as r = 1 - S, and use the level result to characterize the safety state of the power battery. The state evaluation alarm level model is:

[0058]

[0059] Among them, alarmLevel represents the alarm level obtained by evaluating the state of the data to be evaluated. The 0, 1, 2, and 3-level alarms correspond to no, low, medium, and high risks respectively. r = 1 - S represents the vehicle risk. r1, r2, and r3 correspond to the risk alarm thresholds respectively, and r1 < r2 < r3. The risk alarm thresholds are set according to actual needs, such as 40, 60, 80.

[0060] In this embodiment, by slicing the historical charging data of new energy vehicles, charging data segments are obtained, charging feature extraction is completed, and then combined with the consistency state safety evaluation method to obtain the evaluation score, and finally the state evaluation alarm level is output to timely warn of potential vehicle risks and prevent the abnormal state of the vehicle from evolving into a more serious accident risk.

[0061] The above are only embodiments of the present invention. Common general knowledge such as specific structures and characteristics in the solutions is not described in detail here. It should be noted that for those skilled in the art, without departing from the structure of the present invention, several deformations and improvements can be made, which should also be regarded as the protection scope of the present invention, and these will not affect the implementation effect of the present invention and the practicality of the patent. The protection scope required by this application should be subject to the content of its claims, and the specific implementation manners in the specification and the like can be used to explain the content of the claims.

Claims

1. A method for evaluating the consistency safety status of a power battery, characterized in that: The following steps are involved: Step 1: Obtain historical operating data of the vehicle to be evaluated, and extract multiple charging data segments that can reflect the vehicle status from the historical operating data; Step 2: Calculate the standard deviation feature and variance entropy consistency feature of the cell voltage of each charging data segment to obtain the feature value; Step 3, obtaining the number of times the charging process of the vehicle covers the target preset interval throughout its life cycle, multiplying the number of times covered by the first parameter to obtain the first correction number, using the characteristic value of the first correction number in the charging data segment as the reference sample, multiplying the number of times covered by the second parameter to obtain the second correction number, using the characteristic value of the second correction number in the charging data segment as the evaluation sample; Step 4: construct the eigenvalues ​​into a feature matrix, delete the eigenvalues ​​greater than the 99th percentile in the feature matrix, perform unsupervised training on the feature matrix using a preset algorithm, and divide it into two categories to obtain a confusion matrix; Step 5: construct a power battery consistency safety state quantitative calculation model based on the confusion matrix, and calculate the consistency state safety assessment score through the consistency safety state quantitative calculation model; Step six: construct a state assessment alarm level model based on the consistent state safety assessment score, and output the level result to characterize the safety state of the power battery with the level result.

2. The method for evaluating the consistency safety status of a power battery according to claim 1, characterized in that: In the step 1, the battery OCV curve of the vehicle to be evaluated is obtained, and a charging data segment is extracted from the SOC interval on the OCV curve.

3. The method for evaluating the consistency safety status of a power battery according to claim 2, characterized in that: In step 2, the calculation formula of the standard deviation feature is: The calculation formula of variance entropy consistency feature is:

4. The method for evaluating the consistency safety status of a power battery according to claim 3, characterized in that: In the step three, the number of coverages is represented as n, the first parameter is represented as α, the second parameter is represented as β, the first correction number is n*α, and the second correction number is n*β.

5. The method for evaluating the consistency safety status of a power battery according to claim 4, characterized in that: The value range of the first parameter α is: (0.05, 0.5), and the value range of the second parameter is: (0.01, 0.2).

6. The method for evaluating the consistency safety status of a power battery according to claim 5, characterized in that: In step 4, the feature matrix is ​​represented as X, and the confusion matrix obtained after unsupervised training of the feature matrix is:

7. The method for evaluating the consistency safety status of a power battery according to claim 5, characterized in that: In step 5, the consistent safety state quantitative calculation model is: Among them, S represents the consistency status security assessment score of the evaluation sample, #(X 0_1 )、#(X0)、#(X 1_1 ), #(X1) respectively represent (X 0_1 )、X0、X 1_1 , the number of X1 samples, represents the mean of the variance characteristics in the reference sample, represents the mean of the variance entropy consistency feature in the reference sample, represents the variance characteristic mean in the evaluation sample, represents the variance entropy consistency feature mean in the evaluation sample, μ and Represents weight, and the value range of weight is [0,1].

8. The method for evaluating the consistency safety status of a power battery according to claim 5, characterized in that: In step 6, the state assessment alarm level model is: Among them, alarmLevel represents the alarm level obtained by evaluating the status of the data to be evaluated. Level 0, 1, 2, and 3 alarms correspond to no, low, medium, and high risks, respectively. r = 1-S represents the vehicle risk. r1, r2, and r3 correspond to the risk alarm thresholds, respectively, and r1 <r2<r3。

Citation Information

Cited By

  • Power battery safety state detection method, electronic equipment, storage medium and device

    CN117962609A

  • Power battery safety state detection method, electronic device, storage medium and apparatus

    CN117962609B

  • Method for rapidly evaluating attenuation level and consistency of lithium ion storage battery

    CN122043234A