1. A lithium battery pack residual life prediction method based on optimized variational modal decomposition is characterized by comprising the following steps: the method specifically comprises the following steps:

s1, measuring the discharge capacity data sequence of the lithium battery pack along with the charge-discharge period, and generating a charge-discharge period data sequence;

s2, processing the discharge capacity data sequence of the lithium battery pack by adopting a posterior feedback confidence coefficient method, preferably selecting the number of variable-fraction modal decomposition modal layers, and generating an intrinsic modal component data sequence;

s3, generating a discharge capacity degradation trend component data sequence and a noise component data sequence based on the eigenmode component data sequence;

s4, predicting the discharge capacity degradation trend data sequence of the lithium battery pack in the future charge-discharge period by using the particle filter optimized by the nonlinear least square method based on the discharge capacity degradation trend component data sequence and the charge-discharge period data sequence of the lithium battery pack;

s5, establishing a noise prediction model by applying Gaussian process regression based on the noise component data sequence training set and the charge-discharge period data sequence, and predicting the noise data sequence of the lithium battery pack in the future charge-discharge period;

s6, calculating a discharge capacity data sequence of the lithium battery pack in the future charge-discharge period by utilizing the discharge capacity degradation trend data sequence in the future charge-discharge period and the noise data sequence in the future charge-discharge period;

and S7, calculating the residual service life of the lithium battery pack based on the failure threshold value of the lithium battery pack.

2. The method for predicting the residual life of the lithium battery pack based on the optimized variational modal decomposition according to claim 1, wherein the method comprises the following steps: the specific content of step S1 is as follows:

the discharge capacity data sequence of the lithium battery pack along with the charge-discharge period is measured as [ C ]₁,...,C_i,...,C_n]Wherein, C_iThe discharge capacity of the lithium battery pack in the ith (i ═ 1, 2.., n) charge-discharge period is shown, and n is the number of charge-discharge periods;

the charging and discharging period data sequence of the lithium battery pack is [ T ]₁,...,T_i,...,T_n]Wherein, T_iIs the number of charge-discharge cycles corresponding to the ith charge-discharge cycle.

3. The method for predicting the residual life of the lithium battery pack based on the optimized variational modal decomposition according to claim 2, wherein the method comprises the following steps: step S2, processing the lithium battery pack discharge capacity data sequence by using the posterior feedback confidence method, preferably selecting the number of layers of the variational modal decomposition modal, specifically including the steps of:

(a) initializing a variation modal decomposition algorithm, wherein the variation modal decomposition algorithm comprises an intrinsic modal decomposition modal layer number, an iteration number lambda, an algorithm termination condition and the like, and the initial intrinsic modal decomposition modal layer number is 2;

(b) decomposing the initial discharge capacity data sequence mu into mu by using a variational modal decomposition algorithm_λ,1、μ_λ，2Verification of mu and mu_λ,1、μ_λ，2And checking whether an end condition 1 is reached, wherein μ_λ,1、μ_λ，2Two eigenmode components of μ at the λ -th iteration, respectively, and the end condition 1 is defined as: mu and mu_λ,1、μ_λ,2Maximum correlation significance level p of_λ≥0.05(ρ_λ＝max{ρ_λ,1,ρ_λ,2}), i.e., mu and mu_λ,1、μ_λ,2Less than 95% of the maximum correlation confidence;

(c) the data sequence u 'is calculated, the Pearson correlation degree of u with u' is verified, and it is checked whether the end condition 2 is reached, wherein,ε_λ，1is μ and μ at the λ -th iteration_λ,1Of the correlation coefficient of_λ,2Is μ and μ at the λ -th iteration_λ，2The end condition 2 is defined as: significant level of correlation of μ with u'. rho_λ'> 0.05, i.e., the confidence of the correlation of mu with u' is less than 95%;

(d) and if the two tests do not reach the end condition, updating the discharge capacity data sequence mu (mu is mu ') and the iteration number lambda is lambda +1, repeating the steps (b) to (c) until the algorithm end condition is met, otherwise, ending the algorithm, and outputting the number K of the mode layers after optimization when the variation mode decomposition is applied to the initial discharge capacity data sequence mu, wherein the lambda' is the iteration number at the end of the algorithm.

4. The method for predicting the residual life of the lithium battery pack based on the optimized variational modal decomposition according to claim 3, wherein the method comprises the following steps: the discharge capacity data sequence of the lithium battery pack is processed by adopting a posterior feedback confidence coefficient method, the number of variable-mode decomposition mode layers is optimized, and the generated intrinsic mode component data sequence is as follows:

k in total, wherein j is 1., K,is an intrinsic mode component IMF_jAmplitude during the ith charge-discharge cycle.

5. The method for predicting the residual life of the lithium battery pack based on the optimized variational modal decomposition according to claim 4, wherein the method comprises the following steps: the specific content of step S3 is as follows:

the discharge capacity degradation trend component data sequence is IMF₁The noise component data sequence is IMF₂,...,IMF_KAnd K-1 in total.

6. The method for predicting the residual life of the lithium battery pack based on the optimized variation modal decomposition according to claim 5, wherein the method comprises the following steps: the specific content of step S4 is as follows:

the particle filtering optimized by the nonlinear least square method is as follows: IMF (intrinsic mode function) based on discharge capacity degradation trend component of lithium battery pack₁The method comprises the following steps of optimizing an initial value of a particle filter observation equation by using a nonlinear least square method:

(1) defining an observation equation:wherein a, b, c and d are undetermined coefficients, and f (T)_i) Is the T th_iObserved values of discharge capacity in the secondary charge-discharge cycle;

(2) calculate n data points (T)_i,C_i) To discharge capacity observed value f (T)_i) Is a distance ofThe sum of squares J (a, b, c, d), note

(3) Outputting parameters a, b, c and d corresponding to the minimum J (a, b, c and d), namely the initial value of the particle filter observation equation;

the data of the discharge capacity degradation trend of the lithium battery pack in the future charge-discharge period predicted by the particle filter optimized by the nonlinear least square method isThe discharge capacity degradation trend data sequence of the lithium battery pack predicted by the particle filter optimized by the corresponding nonlinear least square method in the future charge-discharge period isWherein l is the starting point of the charge-discharge cycle number of the discharge capacity degradation trend of the lithium battery pack in the future charge-discharge cycle predicted by the particle filter optimized by the nonlinear least square method,shows a discharge capacity degradation tendency component IMF₁And predicting the discharge capacity degradation trend of the l +1 th charge-discharge period.

7. The method for predicting the residual life of the lithium battery pack based on the optimized variational modal decomposition according to claim 6, wherein the method comprises the following steps: the specific content of step S5 is as follows:

a training set of noise component data sequences asBased on a noise component data sequence training set, noise data sequences of K-1 lithium battery packs in future charge-discharge cycles predicted by applying Gaussian process regression are respectively as follows:whereinRepresenting noise components IMF_KNoise prediction value in the l +1 th charge-discharge period.

8. The method for predicting the residual life of the lithium battery pack based on the optimized variational modal decomposition according to claim 7, wherein the method comprises the following steps: the specific content of step S6 is as follows: the method for calculating the discharge capacity data sequence of the lithium battery pack in the future charge-discharge cycle by utilizing the discharge capacity degradation trend data sequence of the future charge-discharge cycle and the noise data sequence of the future charge-discharge cycle comprises the following steps:

wherein [ C_{l+1,predicted},...,C_n,predicted]And (4) obtaining a predicted discharge capacity data sequence of the lithium battery pack in the future charge-discharge period.

9. The method for predicting the residual life of the lithium battery pack based on the optimized variational modal decomposition according to claim 8, wherein the method comprises the following steps: the specific content of step S7 is as follows:

the failed capacity threshold value of the lithium battery pack is that the discharge capacity of the lithium battery pack is reduced to 70% of the nominal capacity;

the residual service life is the residual charge-discharge period number before the lithium battery pack is out of work, and RUL is ═ T_true-T_predictedL, where RUL is the residual life of the lithium battery pack, T_trueFor measuring the corresponding charge-discharge period number T when the discharge capacity of the lithium battery pack reaches the failure threshold value_predictedAnd predicting the corresponding charge-discharge cycle number when the discharge capacity of the lithium battery pack in the future charge-discharge cycle reaches the failure threshold value.

Background

The lithium battery is used as a substitute of fuel oil, and is widely assembled and used in groups on electric automobiles due to the characteristics of high specific energy, high specific power, low self-discharge rate and the like. Because its strong environmental dependence, the health status of lithium cell group constantly worsens in electric automobile use, leads to the user to feel worry to electric automobile's continuation of the journey mileage and safety problem, consequently, in order to foresee the health status of lithium cell group in advance, carries out accurate prediction to its remaining life and is indispensable.

The health condition of the lithium battery pack is generally used as a quantitative index of the capacity aging degree of the lithium battery pack, the discharge capacity data is obtained in continuous charge-discharge cycle tests, and the data obtaining process cannot avoid the influence of various factors, so that the residual life of the lithium battery pack cannot be accurately predicted.

The variation modal decomposition is a novel signal processing method, degradation trend data of the lithium battery pack can be extracted by determining the frequency center and the bandwidth of each component, noise data of different degrees are separated, the modal layer number is important for the decomposition result, and a uniform determination rule of the modal layer number is not provided. The particle filter is a sequential importance sampling method for expressing data distribution by random state particles extracted from posterior probability, and the improvement of the observation initial value of the particle filter has certain significance for improving the prediction precision of the particle filter. Gaussian process regression is a probabilistic model with generalization and resolvability and is widely applied to time series analysis problems. The number of layers of the variation modal decomposition modal is optimized, the optimized variation modal decomposition is utilized to generate an intrinsic modal function, the intrinsic modal function generates a lithium battery pack discharge capacity degradation trend component and a noise data component, the lithium battery pack discharge capacity degradation trend and the noise data are respectively predicted by adopting particle filtering and Gaussian process regression, finally, the discharge capacity and the residual life of the lithium battery pack in the future charge-discharge cycle are predicted based on the prediction result of the particle filtering and the Gaussian process regression, and the precision, the efficiency and the generalization capability of life prediction are effectively improved.

Disclosure of Invention

The invention aims to make up for the defects of the prior art, and provides a method for predicting the residual life of a lithium battery pack based on optimized variational modal decomposition, which can effectively reflect the degradation of the discharge capacity of the lithium battery pack and accurately predict the residual life of the lithium battery pack.

The invention is realized by the following technical scheme:

a lithium battery pack residual life prediction method based on optimized variational modal decomposition specifically comprises the following steps:

s1, measuring the discharge capacity data sequence of the lithium battery pack along with the charge-discharge period, and generating a charge-discharge period data sequence;

s2, processing the discharge capacity data sequence of the lithium battery pack by adopting a posterior feedback confidence coefficient method, preferably selecting the number of variable-fraction modal decomposition modal layers, and generating an intrinsic modal component data sequence;

s3, generating a discharge capacity degradation trend component data sequence and a noise component data sequence based on the eigenmode component data sequence;

s4, predicting the discharge capacity degradation trend data sequence of the lithium battery pack in the future charge-discharge period by using the particle filter optimized by the nonlinear least square method based on the discharge capacity degradation trend component data sequence and the charge-discharge period data sequence of the lithium battery pack;

s5, establishing a noise prediction model by applying Gaussian process regression based on the noise component data sequence training set and the charge-discharge period data sequence, and predicting the noise data sequence of the lithium battery pack in the future charge-discharge period;

s6, calculating a discharge capacity data sequence of the lithium battery pack in the future charge-discharge period by utilizing the discharge capacity degradation trend data sequence in the future charge-discharge period and the noise data sequence in the future charge-discharge period;

and S7, calculating the residual service life of the lithium battery pack based on the failure threshold value of the lithium battery pack.

The specific content of step S1 is as follows:

the discharge capacity data sequence of the lithium battery pack along with the charge-discharge period is measured as [ C ]₁,...,C_i,...,C_n]Wherein, C_iThe discharge capacity of the lithium battery pack in the ith (i ═ 1, 2.., n) charge-discharge period is shown, and n is the number of charge-discharge periods;

the charging and discharging period data sequence of the lithium battery pack is [ T ]₁,...,T_i,...,T_n]Wherein, T_iIs the number of charge-discharge cycles corresponding to the ith charge-discharge cycle.

Step S2, processing the lithium battery pack discharge capacity data sequence by using the posterior feedback confidence method, preferably selecting the number of layers of the variational modal decomposition modal, specifically including the steps of:

(a) initializing a variation modal decomposition algorithm, wherein the variation modal decomposition algorithm comprises an intrinsic modal decomposition modal layer number, an iteration number lambda, an algorithm termination condition and the like, and the initial intrinsic modal decomposition modal layer number is 2;

(b) decomposing the initial discharge capacity data sequence mu into mu by using a variational modal decomposition algorithm_λ,1、μ_λ，2Verification of mu and mu_λ,1、μ_λ，2And checking whether an end condition 1 is reached, wherein μ_λ,1、μ_λ，2Two eigenmode components of μ at the λ -th iteration, respectively, and the end condition 1 is defined as: mu and mu_λ,1、μ_λ,2Maximum correlation significance level p of_λ≥0.05(ρ_λ＝max{ρ_λ,1,ρ_λ,2}), i.e., mu and mu_λ,1、μ_λ,2Less than 95% of the maximum correlation confidence;

(c) the data sequence u 'is calculated, the Pearson correlation degree of u with u' is verified, and it is checked whether the end condition 2 is reached, wherein,ε_λ，1is μ and μ at the λ -th iteration_λ,1Of the correlation coefficient of_λ,2Is μ and μ at the λ -th iteration_λ，2The end condition 2 is defined as: significant level of correlation of μ with u'. rho_λ'> 0.05, i.e., the confidence of the correlation of mu with u' is less than 95%;

(d) and if the two tests do not reach the end condition, updating the discharge capacity data sequence mu (mu is mu ') and the iteration number lambda is lambda +1, repeating the steps (b) to (c) until the algorithm end condition is met, otherwise, ending the algorithm, and outputting the number K of the mode layers after optimization when the variation mode decomposition is applied to the initial discharge capacity data sequence mu, wherein the lambda' is the iteration number at the end of the algorithm.

The discharge capacity data sequence of the lithium battery pack is processed by adopting a posterior feedback confidence coefficient method, the number of variable-mode decomposition mode layers is optimized, and the generated intrinsic mode component data sequence is as follows: k in total, wherein j is 1., K,is an intrinsic mode component IMF_jAmplitude during the ith charge-discharge cycle.

The specific content of step S3 is as follows:

the discharge capacity degradation trend component data sequence is IMF₁The noise component data sequence is IMF₂,...,IMF_KAnd K-1 in total.

The specific content of step S4 is as follows:

the particle filtering optimized by the nonlinear least square method is as follows: IMF (intrinsic mode function) based on discharge capacity degradation trend component of lithium battery pack₁The method comprises the following steps of optimizing an initial value of a particle filter observation equation by using a nonlinear least square method:

(1) defining an observation equation:wherein a, b, c and d are undetermined coefficients, and f (T)_i) Is the T th_iObserved values of discharge capacity in the secondary charge-discharge cycle;

(2) calculate n data points (T)_i,C_i) To discharge capacity observed value f (T)_i) Is a distance ofThe sum of squares J (a, b, c, d), note

(3) Outputting parameters a, b, c and d corresponding to the minimum J (a, b, c and d), namely the initial value of the particle filter observation equation;

lithium battery pack future charging and discharging predicted by particle filtering optimized by nonlinear least square methodThe discharge capacity degradation trend data of the electrical cycle isThe discharge capacity degradation trend data sequence of the lithium battery pack predicted by the particle filter optimized by the corresponding nonlinear least square method in the future charge-discharge period isWherein l is the starting point of the charge-discharge cycle number of the discharge capacity degradation trend of the lithium battery pack in the future charge-discharge cycle predicted by the particle filter optimized by the nonlinear least square method,shows a discharge capacity degradation tendency component IMF₁And predicting the discharge capacity degradation trend of the l +1 th charge-discharge period.

The specific content of step S5 is as follows:

a training set of noise component data sequences asBased on a noise component data sequence training set, noise data sequences of K-1 lithium battery packs in future charge-discharge cycles predicted by applying Gaussian process regression are respectively as follows:whereinRepresenting noise components IMF_KNoise prediction value in the l +1 th charge-discharge period.

The specific content of step S6 is as follows: the method for calculating the discharge capacity data sequence of the lithium battery pack in the future charge-discharge cycle by utilizing the discharge capacity degradation trend data sequence of the future charge-discharge cycle and the noise data sequence of the future charge-discharge cycle comprises the following steps:

wherein [ C_{l+1,predicted},...,C_n,predicted]And (4) obtaining a predicted discharge capacity data sequence of the lithium battery pack in the future charge-discharge period.

The specific content of step S7 is as follows:

the failed capacity threshold value of the lithium battery pack is that the discharge capacity of the lithium battery pack is reduced to 70% of the nominal capacity;

the residual service life is the residual charge-discharge period number before the lithium battery pack is out of work, and RUL is ═ T_true-T_predictedL, where RUL is the residual life of the lithium battery pack, T_trueFor measuring the corresponding charge-discharge period number T when the discharge capacity of the lithium battery pack reaches the failure threshold value_predictedAnd predicting the corresponding charge-discharge cycle number when the discharge capacity of the lithium battery pack in the future charge-discharge cycle reaches the failure threshold value.

The invention has the advantages that: (1) the invention firstly provides a method for determining the number of layers of the variational modal decomposition modal based on a posterior feedback confidence coefficient method, so that the working time of a system is reduced, and effective data is ensured not to be lost to the maximum extent;

(2) the invention utilizes the discharge capacity degradation trend data predicted by the nonlinear two-multiplication optimized particle filter and the noise data predicted by the Gaussian process regression to predict the service life of the lithium ion battery pack, and the prediction precision is obviously higher than that of the traditional single particle filter prediction or single Gaussian process regression prediction.

Drawings

Fig. 1 is a schematic flowchart of a method for predicting remaining life of a lithium battery pack based on optimized variational modal decomposition according to an embodiment of the present invention;

fig. 2 is a discharge capacity degradation display diagram of a lithium battery pack provided by an embodiment of the present invention;

fig. 3 is a data decomposition display diagram of variation modal decomposition of a lithium battery pack data sequence according to an embodiment of the present invention;

fig. 4 is a comparison graph of the prediction result of the remaining life of the lithium battery pack based on the optimized variational modal decomposition according to the embodiment of the present invention with the prediction results of the remaining life of the lithium battery pack based on the other two methods;

Detailed Description

In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention. In addition, the technical features involved in the embodiments of the present invention described below may be combined with each other as long as they do not conflict with each other.

Fig. 1 is a schematic flowchart of a method for predicting remaining life of a lithium battery pack based on optimized variational modal decomposition according to an embodiment of the present invention, where the method shown in fig. 1 includes the following steps:

and S1, measuring the discharge capacity data sequence of the lithium battery pack along with the charge-discharge period, and generating the charge-discharge period data sequence.

The discharge capacity data sequence of the lithium battery pack along with the charge-discharge period is measured as [ C ]₁,...,C_i,...,C_n]Wherein, C_iThe discharge capacity of the lithium battery pack in the ith (i ═ 1, 2.., n) charge-discharge period is shown, and n is the number of charge-discharge periods;

in the embodiment of the invention, in step S1, the data sequence of the charging and discharging period of the lithium battery pack is [ T ]₁,...,T_i,...,T_n]Wherein, T_iIs the number of charge-discharge cycles corresponding to the ith charge-discharge cycle.

And S2, processing the discharge capacity data sequence of the lithium battery pack by adopting a posterior feedback confidence coefficient method, preferably selecting the number of the variable-fractional modal decomposition modal layers, and generating an intrinsic modal component data sequence.

In the embodiment of the present invention, in step S2, a posterior feedback confidence method is used to process a lithium battery pack discharge capacity data sequence, and preferably, the number of layers of the variation modal decomposition modal is selected, which specifically includes:

(a) initializing a variation modal decomposition algorithm, wherein the variation modal decomposition algorithm comprises an intrinsic modal decomposition modal layer number, an iteration number lambda, an algorithm termination condition and the like, and the initial intrinsic modal decomposition modal layer number is 2;

(b) decomposing the initial discharge capacity data sequence mu into mu by using a variational modal decomposition algorithm_λ,1、μ_λ，2Verification of mu and mu_λ,1、μ_λ，2And checking whether an end condition 1 is reached, wherein μ_λ,1、μ_λ，2Two eigenmode components of μ at the λ -th iteration, respectively, and the end condition 1 is defined as: mu and mu_λ,1、μ_λ,2Maximum correlation significance level p of_λ≥0.05(ρ_λ＝max{ρ_λ,1,ρ_λ,2}), i.e., mu and mu_λ,1、μ_λ,2Less than 95% of the maximum correlation confidence;

(c) the data sequence u 'is calculated, the Pearson correlation degree of u with u' is verified, and it is checked whether the end condition 2 is reached, wherein,ε_λ，1is μ and μ at the λ -th iteration_λ,1Of the correlation coefficient of_λ,2Is μ and μ at the λ -th iteration_λ，2The end condition 2 is defined as: significant level of correlation of μ with u'. rho_λ'> 0.05, i.e., the confidence of the correlation of mu with u' is less than 95%;

(d) and if the two tests do not reach the end condition, updating the discharge capacity data sequence mu (mu is mu ') and the iteration number lambda is lambda +1, repeating the steps (b) to (c) until the algorithm end condition is met, otherwise, ending the algorithm, and outputting the number K of the mode layers after optimization when the variation mode decomposition is applied to the initial discharge capacity data sequence mu, wherein the lambda' is the iteration number at the end of the algorithm.

Processing the discharge capacity data sequence of the lithium battery pack by adopting a posterior feedback confidence coefficient method, preferably selecting the number of variable modal decomposition modal layers, and generating an intrinsic modal component data sequence as follows: k in total, wherein j is 1., K,is an intrinsic mode component IMF_jAmplitude during the ith charge-discharge cycle.

And S3, generating a discharge capacity degradation trend component data sequence and a noise component data sequence based on the eigenmode component data sequence.

In the embodiment of the invention, in step S3, the discharge capacity degradation tendency component data sequence is IMF₁The noise component data sequence is IMF₂,...,IMF_KAnd K-1 in total.

And S4, predicting the discharge capacity degradation trend data sequence of the lithium battery pack in the future charge-discharge period by using the particle filter optimized by the nonlinear least square method based on the discharge capacity degradation trend component data sequence and the charge-discharge period data sequence of the lithium battery pack.

In the embodiment of the present invention, in step S4,

the nonlinear least square method optimized particle filtering refers to: IMF (intrinsic mode function) based on discharge capacity degradation trend component of lithium battery pack₁The method comprises the following steps of optimizing an initial value of a particle filter observation equation by using a nonlinear least square method:

(1) defining an observation equation:wherein a, b, c and d are undetermined coefficients, and f (T)_i) Is the T th_iObserved values of discharge capacity in the secondary charge-discharge cycle;

(2) calculate n data points (T)_i,C_i) To discharge capacity observed value f (T)_i) Is a distance ofThe sum of squares J (a, b, c, d), note

(3) And outputting the parameters a, b, c and d corresponding to the minimum J (a, b, c and d), namely the initial values of the particle filter observation equation.

The data of the discharge capacity degradation trend of the lithium battery pack in the future charge-discharge period predicted by the particle filter optimized by the nonlinear least square method isThe discharge capacity degradation trend data sequence of the lithium battery pack predicted by the particle filter optimized by the corresponding nonlinear least square method in the future charge-discharge period isWherein l is the starting point of the charge-discharge cycle number of the discharge capacity degradation trend of the lithium battery pack in the future charge-discharge cycle predicted by the particle filter optimized by the nonlinear least square method,shows a discharge capacity degradation tendency component IMF₁And predicting the discharge capacity degradation trend of the l +1 th charge-discharge period.

And S5, establishing a noise prediction model by applying Gaussian process regression based on the noise component data sequence training set and the charge-discharge period data sequence, and predicting the noise data sequence of the lithium battery pack in the future charge-discharge period.

In the embodiment of the present invention, in step S5, the training set of noise component data sequence isBased on a noise component data sequence training set, noise data sequences of K-1 lithium battery packs in future charge-discharge cycles predicted by applying Gaussian process regression are respectively as follows:whereinRepresenting noise components IMF_KNoise prediction value in the l +1 th charge-discharge period.

And S6, calculating the discharge capacity data sequence of the lithium battery pack in the future charge-discharge period by utilizing the discharge capacity degradation trend data sequence in the future charge-discharge period and the noise data sequence in the future charge-discharge period.

In the embodiment of the present invention, in step S6,

the method for calculating the discharge capacity data sequence of the lithium battery pack in the future charge-discharge cycle by utilizing the discharge capacity degradation trend data sequence of the future charge-discharge cycle and the noise data sequence of the future charge-discharge cycle comprises the following steps:

wherein [ C_{l+1,predicted},...,C_n,predicted]And (4) obtaining a predicted discharge capacity data sequence of the lithium battery pack in the future charge-discharge period.

And S7, calculating the residual service life of the lithium battery pack based on the failure threshold value of the lithium battery pack.

In the embodiment of the present invention, in step S7, the capacity threshold for the failure of the lithium battery pack is that the discharge capacity of the lithium battery pack drops to 70% of the nominal capacity. The residual life is the residual charge-discharge period number before the lithium battery pack is out of work, and RUL is ═ T_true-T_predictedL, where RUL is the residual life of the lithium battery pack, T_trueFor measuring the corresponding charge-discharge period number T when the discharge capacity of the lithium battery pack reaches the failure threshold value_predictedAnd predicting the corresponding charge-discharge cycle number when the discharge capacity of the lithium battery pack in the future charge-discharge cycle reaches the failure threshold value.

The process and estimation performance of the method for predicting the remaining life of the lithium battery pack based on the optimized variation modal decomposition according to the present invention are illustrated as an example.

In a laboratory, six lithium iron phosphate batteries of a certain brand with rated capacity of 2.4Ah and discharge capacity of 2.2Ah are connected in series to form a group, a charge-discharge experiment is carried out on a lithium battery pack, the lithium battery pack is charged at a constant current of 1.2A in a charging stage, when the terminal voltage of the lithium battery pack reaches 24.9V, the terminal voltage is kept unchanged, the charging is continued, and when the charging current is reduced to 48mA, the charging is finished. Discharging with a constant current of 2A after standing for 10s, and finishing the discharging when the voltage of the lithium battery pack is reduced to 19.3V. And (4) repeatedly charging and discharging the lithium battery pack, and ending the experiment when the discharge capacity of the lithium battery pack is lower than 65% of the rated capacity. The experiment totals 729 charge-discharge cycles, and the degradation process of the discharge capacity of the lithium battery pack along with the charge-discharge period is shown in fig. 2. Predicting the residual life of the lithium battery pack based on the discharge capacity degradation data of the lithium battery pack along with the charge-discharge period measured in a laboratory, wherein the specific operation steps are as follows:

(1) based on the lithium battery data measured in the laboratory, a total of 729 sets of data sequences were counted.

(2) Processing the discharge capacity data sequence of the lithium battery pack by adopting a posterior feedback confidence coefficient method, preferably selecting the number of variable modal decomposition modal layers to obtain the optimal number of variable modal decomposition modal layers of the lithium battery pack data sequence as 4, and generating an intrinsic modal component data sequence, wherein fig. 3 is a diagram of the intrinsic modal component data sequence generated after variable modal decomposition is applied to lithium battery pack data.

(3) Generating discharge capacity degradation trend component data sequence IMF based on eigenmode component data sequence₁With noise component data sequences IMF₂,IMF₃,IMF₄。

(4) And predicting the discharge capacity degradation trend data sequence of the lithium battery pack in the future charge-discharge period by using the particle filter optimized by the nonlinear least square method based on the discharge capacity degradation trend component data sequence IMF1 and the charge-discharge period data sequence, wherein the starting point is the 365 th charge-discharge period.

(5) Based on the noise component data sequences IMF2, IMF3 and IMF4 training sets and the charge-discharge period data sequences, taking the 365 th charge-discharge period as a starting point, applying Gaussian process regression to establish a noise prediction model, and predicting the noise data sequences of the future charge-discharge period of the lithium battery pack.

(6) And calculating the discharge capacity data sequence of the lithium battery pack in the future charge-discharge period by utilizing the discharge capacity degradation trend data sequence of the future charge-discharge period and the noise data sequence of the future charge-discharge period.

(7) And calculating the failure threshold value of the lithium battery pack to be 1.68Ah, and calculating the residual service life of the lithium battery pack.

Meanwhile, in order to verify the superiority of the method provided by the invention, a comparison experiment is carried out by adopting particle filtering and Gaussian process regression and the method provided by the invention, fig. 4 is a comparison graph of prediction results of three methods, and table 2 is an average absolute percentage error and a root mean square error of the prediction results of the three methods on the residual life of the lithium battery pack.

TABLE 1

TABLE 2

It can be seen from the comparison graph and the error table of the result of predicting the remaining life of the lithium battery pack, that the discharge capacity degradation data of the future charge-discharge cycle calculated by the method for predicting the remaining life of the lithium battery pack based on the optimized variational modal decomposition provided by the invention is more consistent with the real discharge capacity degradation data, and the same conclusion can be obtained from table 1. The average absolute percentage error and the root mean square error of the lithium battery pack residual life prediction method provided by the invention are lower than those of single particle filter prediction and Gaussian process regression prediction. Meanwhile, aiming at the prediction of the lithium battery pack failure threshold, the error period of the hybrid prediction method provided by the invention is 9, while the error periods of particle filtering and Gaussian process regression are respectively 50 and 31, so that the service life prediction method provided by the invention is obviously higher in precision. In conclusion, the method for predicting the residual life of the lithium battery pack based on the optimized variational modal decomposition has the advantages of small error, high efficiency and the like.

It should be noted that, according to the implementation requirement, each step/component described in the present application can be divided into more steps/components, and two or more steps/components or partial operations of the steps/components can be combined into new steps/components to achieve the purpose of the present invention.

It will be understood by those skilled in the art that the foregoing is only a preferred embodiment of the present invention, and is not intended to limit the invention, and that any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the scope of the present invention.