CN112630662A - Power battery SOH estimation method based on data driving and multi-parameter fusion - Google Patents
Power battery SOH estimation method based on data driving and multi-parameter fusion Download PDFInfo
- Publication number
- CN112630662A CN112630662A CN202011604636.9A CN202011604636A CN112630662A CN 112630662 A CN112630662 A CN 112630662A CN 202011604636 A CN202011604636 A CN 202011604636A CN 112630662 A CN112630662 A CN 112630662A
- Authority
- CN
- China
- Prior art keywords
- battery
- charging
- soh
- constant
- fusion
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
- G01R31/36—Arrangements for testing, measuring or monitoring the electrical condition of accumulators or electric batteries, e.g. capacity or state of charge [SoC]
- G01R31/367—Software therefor, e.g. for battery testing using modelling or look-up tables
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
- G01R31/36—Arrangements for testing, measuring or monitoring the electrical condition of accumulators or electric batteries, e.g. capacity or state of charge [SoC]
- G01R31/392—Determining battery ageing or deterioration, e.g. state of health
Abstract
The invention relates to a power battery SOH estimation method based on data driving and multi-parameter fusion, which carries out comprehensive evaluation on the battery SOH by extracting five characteristic parameters of the battery related to battery performance attenuation in a charging state; and analyzing the correlation between the parameters and the actual capacity of the battery; in order to solve the problem of operation efficiency caused by multiple parameters and eliminate data redundancy, a multidimensional scaling Method (MDS) is selected to reduce the dimension of the selected parameters, and fused one-dimensional characteristic parameters are obtained and used as comprehensive health factors; and finally, training the SOH estimation model by using machine learning. The multi-parameter fusion estimation model provided by the invention solves the problems that the single parameter representation battery health state estimation precision is low and the model is only limited to one battery type. According to the method, effective health factors are selected to fully represent the influence factors of the battery SOH, a complex battery model is not required to be constructed, the operation speed and the estimation precision can be improved, the performance is better compared with that of other models, and the overall operation speed and the use efficiency of the BMS are improved.
Description
Technical Field
The invention belongs to the field of real-time estimation of power battery state parameters, and particularly relates to a power battery SOH estimation method based on data driving and multi-parameter fusion.
Background
The state of health (SOH) of a battery indicates the degree of deviation of a characteristic parameter of the battery from a nominal parameter, and is generally measured by the decrease in capacity of the battery, the increase and change in internal resistance, and the like, from the viewpoint of external characteristics of the battery. Meanwhile, the method is also an important basis for battery SOC estimation, SOP estimation, balanced battery management and service life management. The accurate estimation of the SOH provides safety guarantee for a Battery Management System (BMS) to monitor a control system and a use state of a power battery.
The method for calculating the SOH of the battery at the present stage mainly comprises two methods, namely a battery model-based method and a data-driven method, wherein the method for establishing the battery model is based on the internal mechanism change of the power battery, the power battery model is established and comprises an equivalent circuit model, a decline model and an electrochemical model, and the health condition and the residual service life are estimated by identifying the capacity, the resistance and other electrochemical parameters of the battery through the model. The method based on data driving is characterized in that health factors representing health states are extracted through measurable data in the charging and discharging processes of the lithium ion battery, and the estimated value of the SOH of the battery is estimated by utilizing algorithms such as a neural network method, a support vector machine, a correlation vector machine and Gaussian process regression. The data driving method does not need to solve the working principle of the battery and establish a complex battery model, and is more suitable for estimating the SOH of the battery in the actual operation process. Meanwhile, in practical application, the electric automobile almost has no stable continuous constant-current discharge working condition, and the voltage and other parameters are not in a simple and predictable state due to the phenomena of braking energy recovery and the like, and compared with the situation that the battery is selected to be fixed in the charging working condition rule, the performance state of the battery can be reflected more correctly, so that the change of the health performance of the battery is reflected.
However, if the estimation method using the battery model is limited to the same type of battery, and for the data-driven model, if a single health factor is selected, the estimation result is inaccurate, and if multiple health factors are selected at the same time, the estimation calculation amount is large, which causes the BMS to bear a high-strength load, resulting in a low calculation speed, and the like.
Disclosure of Invention
The invention aims to solve the related problems in the prior art, and provides a power battery SOH estimation method based on data driving and multi-parameter fusion.
In order to achieve the purpose, the invention adopts the following technical scheme, which specifically comprises the following steps:
a SOH estimation method of a power battery based on data driving and multi-parameter fusion is characterized in that relevant data of the battery in a charging state are used for extracting characteristic parameters related to battery performance degradation, and the extracted characteristic parameters are analyzed by a Pearson and Spearman rank correlation analysis method; performing data dimensionality reduction on the extracted multiple feature vectors by using a multidimensional scaling Method (MDS), and finally fusing a health factor capable of representing the battery health state; performing regression model training by adopting a data-driven algorithm, estimating the SOH of the test battery by using the regression model obtained by training, and embedding the regression model into an on-board battery management system to perform SOH estimation on the battery; the method specifically comprises the following steps:
step 1: extracting characteristic parameters of the battery, which can represent battery degradation under the condition of data characteristic parameters such as current, voltage, temperature and the like in a charging state;
step 2: and verifying the correlation between the extracted characteristic parameters and the battery capacity by using a Pearson and Spearman rank correlation analysis method.
And step 3: based on multi-dimensional scaling method MDS algorithm to selected multipleCarrying out dimension reduction fusion on the group characteristic parameters, eliminating parameter value redundancy, and acquiring new health factor HI after fusion1；
And 4, step 4: training a battery SOH estimation model by using a machine learning algorithm;
and 5: and the power Battery SOH estimation model obtained by training is embedded into a Battery management system BMS.
Preferably, in the above power battery SOH estimation method based on data driving and multi-parameter fusion, the step 1 of extracting the data in the charging state takes into account that in practical application, the electric vehicle hardly has a stable continuous constant current discharge condition, and the voltage and other parameters are not in a simple and predictable state due to the existence of phenomena such as energy recovery, and compared with the method that the battery is selected to be fixed in the charging condition rule, the performance state of the battery can be reflected more correctly, so that the change of the health performance of the battery is reflected. Based on the above principle, the present invention selects the current, voltage and temperature variation of the lithium ion battery in different cycle periods in the charging state as shown in fig. 2, fig. 3 and fig. 4. The characteristic parameters comprise: the charging method comprises the steps of constant-time constant-current charging average voltage increasing value, constant-current charging to charging cut-off voltage time, constant-time constant-voltage charging average current decreasing value, constant-current charging current cut-off time and charging temperature peak value reaching time.
Preferably, in the above method, the time-equal constant-current charging average voltage rise value in step 1 is expressed as: [ V ]1，V2，...，Vn]The time from constant current charging to charging cut-off voltage is represented as: [ T ]V1，TV2，...，TVn](ii) a The average current drop value of the constant-time constant-voltage charging is represented as: [ I ] of1，I2，...，In]；
The cutoff time of the constant current charging current is represented as: [ T ]I1，TI2，...，TIn](ii) a The time for the charging temperature to reach the peak value is represented as: [ T ]t1，Tt2，...，Ttn]。
Preferably, in the method for estimating the SOH of the power battery based on data driving and multi-parameter fusion, in step 2, correlation is performed on the characteristic parameters extracted in step 1 and the battery capacity by using Pearson and Spearman rank correlation analysis, and abnormal values of data are removed. The expressions are respectively:
(1) the Pearson correlation analysis method can quantitatively reflect the linear relation between two groups of data, the Pearson correlation coefficient is r, and two groups of variables S are assumed to be equal (S)1,s2,…,sn),Y＝(y1，y2，…，yn) Then Pearson correlation coefficient can be calculated by the following formula:
the value range of the correlation coefficient is r epsilon < -1, + 1), when r is +/-1, the linear relation is formed between two data sequences, and one group of data can be directly used for describing the other group of data. When r is 0, it means that there is no relation between the two data sequences. When r is between 0 and +/-1, it indicates that there is a relationship between the two data sequences, but not a linear relationship, and a certain transformation needs to be performed on the two data sequences to make r approach +/-1, so that the two data sequences reach an approximate linear relationship.
(2) Spearman rank correlation analysis may reflect the degree of monotonicity between the two sets of data. The Spearman rank correlation coefficient is a calculation formula shown as the following formula:
Ridenotes siRank in S, QiDenotes yiRank in Y, where rank represents the ordinal number where the values of the variables are arranged from large to small. Spearman rank correlation coefficient range is rs∈[-1，+1]If r issWhen ± 1, it means strict monotony between the two sets of variables.
Preferably, in the above power battery SOH estimation method based on data driving and multi-parameter fusion, in step 3, dimension reduction fusion is performed on the selected five sets of characteristic parameters based on the multi-dimensional scaling MDS algorithmEliminating parameter value redundancy and obtaining new health factor (HI) after fusion1). The expression of the matrix Z obtained by dimensionality reduction is as follows:
where Λ ═ diag (λ ″)1,λ2,…,λn') is the diagonal matrix formed by the maximum features of the feature vector after dimension reduction, and let V' be the corresponding feature vector matrix.
Preferably, in the method for estimating the SOH of the power battery based on data driving and multi-parameter fusion, the machine learning algorithm is used as the SVR in the step 4, and the SVR has a smaller storage space than other machine learning models, and can process high-dimensional small sample data well.
Preferably, in the above support vector regression model, the model input is the health factor HI obtained based on MDS method multiparameter fusion in step 31And the actual value of the actual capacity of the battery; the output is a predicted value of the battery capacity fade.
Preferably, in the above support vector regression model, the support vector regression model uses a Radial Basis Function (RBF) which is expressed by the following formula:
σ is a parameter of the kernel function, xiAre the elements in the matrix variables for the kernel function.
Preferably, in the above method for estimating the SOH data driven by multi-parameter fusion of power batteries, the SOH estimation model of the power batteries obtained by training in step 5 is embedded in the BMS, and the training model is not limited to the same type of battery model.
Compared with the prior art, the invention has the following technical effects:
the SOH estimation method of the power battery based on data driving and multi-parameter fusion can be suitable for batteries of different types, a data set of the SOH estimation method selects multiple groups of data of the battery during constant current charging to serve as characteristic parameters for estimating the health state of the battery together, the multiple groups of characteristic parameters are fused to obtain one group of vectors serving as health factors (HI), and compared with the SOH estimation method in which all data are selected to be driven simultaneously, high-intensity operation load is caused, operation speed is reduced, and related chips in the system are damaged for the BMS.
Drawings
FIG. 1 is a flow chart of SOH data-driven estimation for multi-parameter fusion of a power battery;
fig. 2 is a power battery BMS control system architecture diagram; (ii) a
FIG. 3 is a graph of battery constant current charging voltage;
FIG. 4 is a graph of battery constant current charging current;
fig. 5 is a graph of constant current charging temperature variation of a battery.
Detailed Description
The technical features of the present invention will be described in further detail below with reference to the accompanying drawings.
The invention provides a power battery SOH estimation method based on data driving and multi-parameter fusion, and a flow chart of the method is shown in figure 1, and the method specifically comprises the following steps:
step 1: extracting characteristic parameters of the battery which can represent battery degradation in a charging state;
considering that in practical application, an electric automobile hardly has a stable continuous constant-current discharge working condition, and parameters such as voltage are not in a simple and predictable state due to phenomena such as energy recovery, in contrast, the performance state of the battery can be reflected more correctly by selecting the battery to be fixed in the charging working condition rule, so that the change of the health performance of the battery is reflected. Based on the above principle, the invention selects the current, voltage and temperature changes of the lithium ion battery in different cycle periods in the charging state as shown in fig. 3, 4 and 5, and extracts the parameters capable of reflecting the performance degradation of the battery. The characteristic parameters extracted by the lithium ion battery charging current, voltage and temperature data include: the charging method comprises the steps of constant-time constant-current charging average voltage increasing value, constant-current charging to charging cut-off voltage time, constant-time constant-voltage charging average current decreasing, constant-current charging current cut-off time and charging temperature peak value reaching time.
(1) Constant-time constant-current charging average voltage rise value:
as shown in FIG. 2, the charging voltage of the battery is divided into three regions, wherein in the gentle region, the voltage has obvious stable variation trend under different cycles, and a reasonable charging time interval [ T ] is selectedi,Tj]Voltage data [ V ] of1，V2，...，Vn]As one of the feature vectors.
(2) Time from constant current charging to charge cutoff voltage:
as shown in FIG. 2, the time to reach the charge cutoff voltage is significantly reduced with the increase of the charge cycle, and the same charge initiation voltage to cutoff voltage time T is selectedV1，TV2，...，TVn]As one of the feature vectors.
(3) The average current reduction value of constant-voltage charging in equal time:
as shown in fig. 3, the current at different cycles has a significant tendency to decay during the constant voltage charging phase. Selecting current attenuation values [ I ] for equal time segments1，I2，...，In]As one of the feature vectors.
(4) Constant current charging current off-time:
as shown in fig. 3, when the constant-current charging reaches the upper limit of the cutoff charging voltage, the constant-current charging is ended. With the increase of charging cycle, the constant current charging time has obvious decline trend, and the decay time [ T ] is selectedI1，TI2，...，TIn]As one of the feature vectors.
(5) Charging temperature peak time:
as shown in fig. 4, during the charging process of the battery, the internal temperature of the battery has a significant rising trend along with the charging time, and the time when the temperature of the battery reaches the peak value has a significant declining trend along with the increase of the charging cycle, and the battery is selectedSelecting its peak time Tt1，Tt2，…，Ttn]As one of the feature vectors.
Step 2: analyzing the correlation of the characteristic parameters;
pearson and Spearman rank correlation analysis quantitatively reflects the linear relationship and degree of monotonicity between the two sets of data.
(1) The Pearson correlation analysis method can quantitatively reflect the linear relation between two groups of data, the Pearson correlation coefficient is r, and two groups of variables S are assumed to be equal (S)1,s2,…,sn),Y＝(y1，y2，…，yn) Then the Pearson correlation coefficient can be calculated by the formula:
the value range of the correlation coefficient is r epsilon < -1, + 1), when r is +/-1, the linear relation is formed between two data sequences, and one group of data can be directly used for describing the other group of data. When r is 0, it means that there is no relation between the two data sequences. When r is between 0 and +/-1, it indicates that there is a relationship between the two data sequences, but not a linear relationship, and a certain transformation needs to be performed on the two data sequences to make r approach +/-1, so that the two data sequences reach an approximate linear relationship.
(2) Spearman rank correlation analysis may reflect the degree of monotonicity between the two sets of data. The Spearman rank correlation coefficient is a calculation formula shown as the following formula:
Ridenotes siRank in S, QiDenotes yiRank in Y, where rank represents the ordinal number where the values of the variables are arranged from large to small. Spearman rank correlation coefficient range is rs∈[-1，+1]If r issWhen ± 1, it means strict monotony between the two sets of variables.
And step 3: acquiring a fused new health factor based on MDS;
in order to obtain the optimal health factor, the selected health factor can simultaneously represent each performance parameter of the battery, and the factors of large calculation amount, difficult data processing and analysis and the like of the estimation of a plurality of characteristic parameters are considered. Therefore, the invention utilizes the method of multidimensional scaling MDS (multidimensional scaling) to perform dimension reduction processing on the data of the multidimensional vector HIn consisting of a plurality of screened characteristic parameters, not only realizes the full guarantee of the richness of the data representing the battery decline performance, but also eliminates the redundancy among a plurality of data, and finally obtains a new fused one-dimensional characteristic parameter HI1As a health factor characterizing the state of health of the battery.
(1) The raw input variable matrix is normalized. Distance matrix D ∈ Rn×mWhere the element d is in the ith row and j columnijIs a sample ziTo zjDistance, n represents the number of characteristic parameters, m represents the number of each characteristic parameter:
wherein d isij＝||zi-zjIf the matrix Z after dimension reduction belongs to Rn′×mWherein n ' is the number of characteristic parameters after dimensionality reduction, n ' here '<n, achieving the effect of reducing the dimension.
(2) Finding a vector B ═ ZTZ, wherein B ═ BijAre such that
(3) Determining the value of the matrix B, and performing eigenvalue decomposition on the matrix B:
B＝VΛVT
wherein Λ ═ diag (λ)1,λ2,…,λn) And V is a characteristic vector matrix.
(4)Λ″＝diag(λ1,λ2,…,λn') is a diagonal matrix of n' largest features,let V "be the corresponding eigenvector matrix, and the expression of the obtained dimension reduction matrix Z is:
when n' is 1, we get as one-dimensional eigenvector: z ═ Z11 z12 … z1m]。
And 4, step 4: training a battery SOH estimation model by using a machine learning algorithm;
the support vector regression model SVR is selected, the SVR model has small storage space compared with other machine learning models, high-dimensional small sample data can be well processed, and the SVR model is very suitable for BMS and can well meet the online requirement. Through the feature selection and the correlation coefficient method analysis obtained in the step 2 and the step 3, a training sample set { (x) can be obtained1，y1)，(x2，y2)，…，(xn，yn) In which xi∈RdIs d-dimensional input feature vector, i.e. input health factor; y isie.R is the input vector xiAnd the corresponding output vector is the actual capacity of the battery.
The regression function uses the following form:
f(x)＝ωTφ(x)+b
where w and b represent the intercepts of the weight matrix and hyperplane, respectively, and φ (x) is a non-linear mapping function, which may also be referred to as a kernel function, denoted as K (x, x'i)。
Selecting a radial basis function RBF (radial basis function) of the form:
σ is a parameter of the kernel function.
Solving a nonlinear regression problem by using an epsilon-insensitivity loss function, wherein epsilon is an allowable error between a true value and an estimated value, a constant C is a penalty factor, and xii，ξi *Is a relaxation factor. The minimization of the optimization objective function is given by:
introducing Lagrange function to the above formula, solving the minimum value of Lagrange function, and constructing a dual form to obtain a nonlinear regression function:
αiand alphai *Is a lagrange multiplier.
In step 4, the x vector corresponds to the one-dimensional vector Z ═ Z obtained in step 3 after dimensionality reduction11 z12 … z1m]And the y vector is a data set of the capacity corresponding to the Z vector. Finally by the predicted capacity value Cp(i)To rated capacity CrComparing and obtaining the state of health SOH of the predicted battery:
and 5: training to obtain a model embedded BMS;
embedding the trained model into a battery BMS main control unit, acquiring the five groups of data optimized in the step 1 by using a data acquisition unit after each charging cycle is completed, finally fusing the selected characteristic parameters to obtain a comprehensive health factor representing the health state of the battery, and estimating the SOH of the current power single battery.
The data in the figures provided by the present invention are experimental data from the NASA laboratory disclosure; the method does not need a complex electrochemical model, and is a power battery SOH model with low calculation intensity and high precision based on battery charging data.
Finally, the above-mentioned embodiments are described in further detail for the purpose of illustrating the invention, and it should be understood by those skilled in the art that the above-mentioned embodiments are only illustrative of the present invention and are not intended to limit the present invention, and any modifications, equivalents, improvements and the like made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (9)
1. A power battery SOH estimation method based on data driving and multi-parameter fusion is characterized by comprising the following steps:
step 1: extracting characteristic parameters which can represent battery degradation in battery performance parameters of the battery in a charging state;
step 2: verifying and extracting correlation between the characteristic parameters and the battery capacity by using a Pearson and Spearman rank correlation analysis method;
and step 3: performing dimensionality reduction fusion on the selected multiple groups of characteristic parameters based on a multidimensional scaling MDS algorithm, eliminating parameter value redundancy, and acquiring new health factor HI after fusion1；
And 4, step 4: training a battery SOH estimation model by using a machine learning algorithm;
and 5: and training the obtained SOH estimation model of the power battery to be embedded into the BMS.
2. The method for estimating SOH of a power battery based on data driving and multi-parameter fusion as claimed in claim 1, wherein the battery performance parameters in step 1 comprise current, voltage and temperature data characteristic parameters.
3. The SOH estimation method for the power battery based on data driving and multi-parameter fusion as claimed in claim 1, wherein the characteristic parameters in step 1 comprise: the charging method comprises the steps of constant-time constant-current charging average voltage increasing value, constant-current charging to charging cut-off voltage time, constant-time constant-voltage charging average current decreasing value, constant-current charging current cut-off time and charging temperature peak value reaching time.
4. The method for estimating the SOH of the power battery based on the data driving and the multi-parameter fusion as claimed in claim 3, wherein the time-constant current charging average voltage rise value is expressed as: [ V ]1，V2，...，Vn]And the time from constant current charging to charging cut-off voltage is represented as: [ T ]V1，TV2，...，TVn]The average current drop value of the constant-time constant-voltage charging is represented as: [ I ] of1，I2，...，In]；
The cutoff time of the constant current charging current is represented as: [ T ]I1，TI2，...，TIn]And the time for the charging temperature to reach the peak value is expressed as: [ T ]t1，Tt2，...，Ttn]。
5. The method for estimating SOH of a power battery based on data driving and multi-parameter fusion as claimed in claim 4, wherein in the step 3, dimension reduction fusion is performed on the five selected sets of characteristic parameters based on a multi-dimensional scaling MDS algorithm, parameter value redundancy is eliminated, and a new health factor HI after fusion is obtained1And the expression of the matrix Z obtained by dimensionality reduction is as follows:
where Λ ═ diag (λ ″)1,λ2,L,λn′) And V is a diagonal matrix formed by the maximum features of the feature vectors after dimension reduction, and V is a feature vector matrix.
6. The method for estimating the SOH of the power battery based on the data driving and the multi-parameter fusion as claimed in claim 1, wherein the step 2 is specifically as follows:
step 2.1, Pearson' S correlation coefficient is r, assuming two sets of variables S ═ S1,s2,…,sn),Y＝(y1，y2，…，yn) Then, the Pearson correlation coefficient is calculated by the formula:
the value range of the correlation coefficient is r ∈ [ -1, +1 ];
step 2.2, the Spearman rank correlation coefficient is a calculation formula shown as the following formula:
Ridenotes siRank in S, QiDenotes yiRank in Y, where rank represents the number of places where the values of the variables are arranged from large to small, and the Spearman rank correlation coefficient range is rs∈[-1，+1]。
7. The method for estimating SOH of a power battery based on data driving and multi-parameter fusion as claimed in claim 1, wherein the step 4 utilizes a machine learning algorithm as a support vector regression model (SVR).
8. The method for estimating SOH of a power battery based on data driving and multi-parameter fusion as claimed in claim 6, wherein in the SVR, the model input quantity is the health factor HI obtained based on MDS method multi-parameter fusion in step 31And the actual value of the actual capacity of the battery; the output is a predicted value of the battery capacity fade.
9. The SOH estimation method for the power battery with data driving and multi-parameter fusion as claimed in claim 6, wherein the support vector regression model SVR adopts a radial basis kernel function RBF, and the formula is as follows:
σ is a parameter of the kernel function, xiAre the elements in the matrix variables for the kernel function.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011604636.9A CN112630662B (en) | 2020-12-30 | 2020-12-30 | Power battery SOH estimation method based on data driving and multi-parameter fusion |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011604636.9A CN112630662B (en) | 2020-12-30 | 2020-12-30 | Power battery SOH estimation method based on data driving and multi-parameter fusion |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112630662A true CN112630662A (en) | 2021-04-09 |
| CN112630662B CN112630662B (en) | 2023-03-17 |
Family
ID=75287599
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011604636.9A Active CN112630662B (en) | 2020-12-30 | 2020-12-30 | Power battery SOH estimation method based on data driving and multi-parameter fusion |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112630662B (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113567874A (en) * | 2021-06-08 | 2021-10-29 | 北京科技大学 | Monitoring and detecting method and device for safety state of power battery |
| CN113805064A (en) * | 2021-09-18 | 2021-12-17 | 北京航空航天大学 | Lithium ion battery pack health state prediction method based on deep learning |
| CN113933725A (en) * | 2021-09-08 | 2022-01-14 | 深圳大学 | Method for determining power battery charge state based on data driving |
| CN114290960A (en) * | 2021-12-30 | 2022-04-08 | 中国第一汽车股份有限公司 | Method and device for acquiring battery health degree of power battery and vehicle |
| CN114325454A (en) * | 2021-12-30 | 2022-04-12 | 东软睿驰汽车技术（沈阳）有限公司 | Method, device, equipment and medium for determining influence of multiple characteristics on battery health degree |
| CN115392134A (en) * | 2022-09-23 | 2022-11-25 | 四川新能源汽车创新中心有限公司 | Vehicle power battery comprehensive performance evaluation method and related device |
| CN116879761A (en) * | 2023-09-06 | 2023-10-13 | 杭州宇谷科技股份有限公司 | Multi-mode-based battery internal short circuit detection method, system, device and medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106908736A (en) * | 2017-03-17 | 2017-06-30 | 哈尔滨工业大学 | Based on the lithium battery method for predicting residual useful life that depth confidence net and Method Using Relevance Vector Machine are merged |
| CN108398652A (en) * | 2017-05-26 | 2018-08-14 | 北京航空航天大学 | A kind of lithium battery health state evaluation method merging deep learning based on multilayer feature |
| CN109061504A (en) * | 2018-08-28 | 2018-12-21 | 中北大学 | Same type difference lithium ion battery remaining life prediction technique and system |
| CN111337832A (en) * | 2019-12-30 | 2020-06-26 | 南京航空航天大学 | Power battery multidimensional fusion SOC and SOH online joint estimation method |
| CN111398837A (en) * | 2020-04-01 | 2020-07-10 | 重庆大学 | Vehicle battery health state estimation method based on data driving |
-
2020
- 2020-12-30 CN CN202011604636.9A patent/CN112630662B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106908736A (en) * | 2017-03-17 | 2017-06-30 | 哈尔滨工业大学 | Based on the lithium battery method for predicting residual useful life that depth confidence net and Method Using Relevance Vector Machine are merged |
| CN108398652A (en) * | 2017-05-26 | 2018-08-14 | 北京航空航天大学 | A kind of lithium battery health state evaluation method merging deep learning based on multilayer feature |
| CN109061504A (en) * | 2018-08-28 | 2018-12-21 | 中北大学 | Same type difference lithium ion battery remaining life prediction technique and system |
| CN111337832A (en) * | 2019-12-30 | 2020-06-26 | 南京航空航天大学 | Power battery multidimensional fusion SOC and SOH online joint estimation method |
| CN111398837A (en) * | 2020-04-01 | 2020-07-10 | 重庆大学 | Vehicle battery health state estimation method based on data driving |
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113567874B (en) * | 2021-06-08 | 2022-07-12 | 北京科技大学 | Monitoring and detecting method and device for safety state of power battery |
| CN113567874A (en) * | 2021-06-08 | 2021-10-29 | 北京科技大学 | Monitoring and detecting method and device for safety state of power battery |
| CN113933725B (en) * | 2021-09-08 | 2023-09-12 | 深圳大学 | Method for determining state of charge of power battery based on data driving |
| CN113933725A (en) * | 2021-09-08 | 2022-01-14 | 深圳大学 | Method for determining power battery charge state based on data driving |
| CN113805064B (en) * | 2021-09-18 | 2022-09-20 | 北京航空航天大学 | Lithium ion battery pack health state prediction method based on deep learning |
| CN113805064A (en) * | 2021-09-18 | 2021-12-17 | 北京航空航天大学 | Lithium ion battery pack health state prediction method based on deep learning |
| CN114325454A (en) * | 2021-12-30 | 2022-04-12 | 东软睿驰汽车技术（沈阳）有限公司 | Method, device, equipment and medium for determining influence of multiple characteristics on battery health degree |
| CN114290960A (en) * | 2021-12-30 | 2022-04-08 | 中国第一汽车股份有限公司 | Method and device for acquiring battery health degree of power battery and vehicle |
| CN114325454B (en) * | 2021-12-30 | 2023-07-04 | 东软睿驰汽车技术（沈阳）有限公司 | Method, device, equipment and medium for determining influence of multiple characteristics on battery health |
| CN114290960B (en) * | 2021-12-30 | 2023-10-27 | 中国第一汽车股份有限公司 | Method and device for acquiring battery health of power battery and vehicle |
| CN115392134A (en) * | 2022-09-23 | 2022-11-25 | 四川新能源汽车创新中心有限公司 | Vehicle power battery comprehensive performance evaluation method and related device |
| CN115392134B (en) * | 2022-09-23 | 2023-09-12 | 四川新能源汽车创新中心有限公司 | Vehicle power battery comprehensive performance evaluation method and related device |
| CN116879761A (en) * | 2023-09-06 | 2023-10-13 | 杭州宇谷科技股份有限公司 | Multi-mode-based battery internal short circuit detection method, system, device and medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112630662B (en) | 2023-03-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112630662B (en) | Power battery SOH estimation method based on data driving and multi-parameter fusion | |
| Tian et al. | Consistency evaluation and cluster analysis for lithium-ion battery pack in electric vehicles | |
| Wang et al. | A method for state-of-charge estimation of Li-ion batteries based on multi-model switching strategy | |
| CN110398697B (en) | Lithium ion health state estimation method based on charging process | |
| CN108549030B (en) | Lithium battery online health state rapid prediction method based on voltage key characteristics | |
| CN111537899A (en) | Method for evaluating safety of power battery by gradient utilization | |
| CN113109715B (en) | Battery health condition prediction method based on feature selection and support vector regression | |
| CN113219343A (en) | Lithium battery health state prediction method, system, equipment and medium based on elastic network | |
| CN110146822A (en) | A kind of Vehicular dynamic battery capacity On-line Estimation method based on constant-current charge process | |
| Zhou et al. | Battery health prognosis using improved temporal convolutional network modeling | |
| JP2023501995A (en) | Battery performance prediction | |
| CN113917334B (en) | Battery health state estimation method based on evolution LSTM self-encoder | |
| CN109800446B (en) | Method and device for estimating voltage inconsistency in discharging process of lithium ion battery | |
| CN109768340B (en) | Method and device for estimating voltage inconsistency in battery discharge process | |
| Huang et al. | A comprehensively optimized lithium-ion battery state-of-health estimator based on Local Coulomb Counting Curve | |
| CN111983474A (en) | Lithium ion battery life prediction method and system based on capacity decline model | |
| CN116047300A (en) | Controller for predicting characteristic parameters of battery and method thereof | |
| CN116660759A (en) | Battery life prediction method and device based on BMS battery management system | |
| Xue et al. | Online capacity estimation of lithium-ion batteries based on deep convolutional time memory network and partial charging profiles | |
| Jiang et al. | Multiple health indicators assisting data-driven prediction of the later service life for lithium-ion batteries | |
| CN116736172A (en) | Lithium battery pack health state evaluation method based on generation type countermeasure network | |
| Lv et al. | Analysis of strategies to maximize the cycle life of lithium-ion batteries based on aging trajectory prediction | |
| CN115718261A (en) | Storage battery capacity estimation method, training method, life estimation method and device | |
| CN114355218A (en) | Lithium ion battery charge state prediction method based on multi-feature quantity screening | |
| CN114966436A (en) | Lithium battery state of charge prediction method, device, equipment and readable storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |