CN115684972A - A Lithium-ion Battery SOH Estimation Method Based on SSA-SVR Model - Google Patents

Abstract

The invention discloses a lithium ion battery SOH estimation method based on an SSA-SVR model. The technical scheme adopted by the invention is as follows: collecting data; extracting health characteristics; constructing a characteristic processing model based on Savitzky-Golay filtering; taking a matrix F = [ F1, F2, F3 and F4] formed by the extracted and processed health characteristics as input, taking the SOH of the battery as output, and constructing a training set and a prediction set; and 5, constructing an SSA-SVR model: and constructing a reference SVR model, optimizing the hyperparameters of the SVR model by adopting a sparrow search algorithm based on the training set, constructing an SOH estimation model, and outputting the estimated SOH. The invention adopts a SVR model which considers 4 indexes of voltage and current segment data in the charging process, processes characteristics and adjusts parameters by combining a sparrow search algorithm, thereby realizing the SOH prediction of the lithium ion battery.

Description

A Lithium-ion Battery SOH Estimation Method Based on SSA-SVR Model

technical field

The invention relates to the technical field of lithium-ion battery state estimation, in particular to a lithium-ion battery SOH estimation method based on an SSA-SVR model.

Background technique

Lithium-ion batteries play a vital role in the field of grid energy storage systems and new energy vehicles because of their advantages in achieving a good balance between various performance parameters such as cost, life, safety and environmental impact. However, the performance of lithium-ion batteries will gradually decrease with the use cycle, which will pose a safety hazard to the battery system. In order to improve the system's ability to resist risky failures and reduce maintenance costs, accurate estimation of the state of health (SOH) of lithium-ion batteries is required. However, the parameters directly related to battery SOH are internal variables, which are difficult to measure directly with sensors and need to be estimated by measuring characteristic parameters. Lithium-ion battery SOH is generally defined as the ratio of the measured capacity to the nominal capacity. When it drops to a certain level, the battery cell should be replaced and maintained. Existing SOH estimation methods can be divided into three categories: model-based methods, data-driven methods and fusion-based methods. The model-based method needs to build a high-complexity model in order to obtain high-precision SOH estimation results, which is difficult to apply in practice. Fusion-based methods aim to combine the advantages of different methods, but have not yet made great progress, suffer from model compatibility issues, and have high computational complexity. Based on the data-driven method, there is no need to analyze the complex reaction mechanism, and the hidden mapping relationship can be extracted from the data. Therefore, in contrast, the data-driven method with more flexible methods and stronger applicability is a current research hotspot. Data-driven methods are mainly influenced by input features and machine learning algorithms.

High-quality input features have the characteristics of stable required data, easy extraction, and high correlation. Some studies use the data in the discharge process for feature extraction, but because the discharge process is greatly affected by the working conditions and loads, it is not as stable as the data in the charging process, and the use of the characteristics of the discharge data is not conducive to the model's adaptation to complex working conditions. Highly correlated features play an important role in improving the accuracy of modeling, and some processing can be performed after feature extraction to improve the correlation between features and SOH. There are various types of existing machine learning models, different principles, and each has its own advantages; and machine learning methods are more dependent on the setting of model hyperparameters, and improper parameters will lead to insufficient generalization ability of the model. Therefore, feature extraction based on segment charging data, combined with feature processing to further improve the correlation between the extracted features and the target quantity, and at the same time use a high-performance search algorithm to adjust the parameters of the machine learning model to improve the accuracy of SOH estimation is an urgent problem to be solved .

Contents of the invention

In order to overcome the deficiencies in the above-mentioned prior art, the present invention proposes a method for estimating the SOH of lithium-ion batteries based on the SSA-SVR model, in order to dig out features highly correlated with SOH from charging data segments, and adopt Savitzky -The Golay filter algorithm processes the features with poor correlation to further improve the correlation of the existing features. At the same time, the sparrow search algorithm is used to optimize the hyperparameters of SVR, so as to improve the prediction accuracy of the SOH estimation model.

In order to achieve the above object, the technical solution adopted in the present invention is: a method for estimating the SOH of lithium-ion batteries based on the SSA-SVR model, which includes:

Step 1. Data collection: charge and discharge the lithium-ion battery multiple times, and record the current, voltage, time data and the full capacity of each discharge of the lithium-ion battery during the charging and discharging process;

Step 2. Extract health features: Extract four health features from the voltage and current data fragments during charging, namely, the voltage value after charging, the current value after charging, the charging duration of the voltage data fragment, and the current data fragment charging duration;

Step 3. Build a feature processing model based on Savitzky-Golay filtering: select the Savitzky-Golay filtering algorithm to process the extracted health features, eliminate the cluttered fluctuations of the health features due to the impact of charging data on sensor drift noise, and construct feature processing Model;

Step 4. Taking the matrix F=[F1, F2, F3, F4] formed by the extracted and processed health features as input, and the SOH of the battery as output, construct a training set and a prediction set;

Step 5. Construct the SSA-SVR model: Construct the benchmark SVR model, based on the above training set, use the sparrow search algorithm to optimize the hyperparameters of the SVR model, construct the SOH estimation model, and output the estimated SOH.

Further, the step 2 includes:

Step 2.1. Extract the charging duration of the voltage data segment [V _a , V _b ] during the charging process as the first health feature F1, and use the formula (1) to calculate the feature F1:

F1＝t _b -t _a (1)

In the formula, t _a is the charging time corresponding to the voltage rising to V _a , and t _b is the charging time corresponding to the voltage rising to V _b .

Step 2.2, extract the charging duration of the current data segment [I _c , I _d ] in the charging process as the second health feature F2, and use the formula (2) to calculate the feature F2:

F2＝t _d -t _c (2)

In the formula, _tc is the charging time corresponding to the current drop to _Ic , and _td is the charging time corresponding to the current drop to _Id .

Step 2.3, extracting the constant current charging process, starting from the voltage V _e as the timing starting point, and the voltage value after charging t _e time as the third health feature F3, using formula (3) to calculate the feature F3:

F3＝V _e +ΔV _e (3)

Among them, ΔV _e is the voltage value changed within the charging time of t _e .

Step 2.4, extracting the constant voltage charging process, starting from the current I _f as the timing starting point, and the current value after charging t _f time as the fourth health feature F4, using the formula (4) to calculate the feature F4:

F4＝I _f +ΔI _f (4)

Among them, ΔI _f is the current value changed within tf charging time.

Further, the step 3 includes:

Step 3.1, use formula (5) to calculate the correlation between the four extracted health characteristics and the target quantity SOH:

In the formula, X and Y are healthy features and SOH samples, respectively.

Step 3.2, adopt Savitzky-Golay filter to build feature processing model, it is determined by polynomial order N and window length L=2M+1, and M represents the absolute value of maximum and minimum measurement point coordinates in the measurement window; For each measurement point is The data of (-M,-M+1,...,0,1,...,M-1,M) is processed using formulas (6), (7) and (8):

In the formula, p(n) is the limit polynomial, a _k represents the coefficient of the polynomial term, ε _N is the residual error of the least squares fitting, x[n] is the real value of the data set in the nth sequence, and the solution is obtained by ( 8) The obtained equation system finally obtains the optimal polynomial coefficients.

Step 3.3. Set the threshold to 0.97. When the Pearson correlation coefficient between the health feature and SOH is greater than the set threshold, the feature processing model will not be activated, and the original aging feature is directly used to estimate SOH; when the health feature and SOH When the Pearson correlation coefficient between is less than the set threshold, it indicates that the health feature is greatly affected by sensor drift noise, and the health feature is processed to improve the correlation between the health feature and the target quantity.

Further, the step 5 includes:

Step 5.1, Build a benchmark SVR model

S={x _i ,y _i |xi _i ∈R ^m ,y _i ∈R}; i=1,2,...,T (9)

In the formula, x _i is the feature vector of the i-th sample, y _i is the corresponding regression value, T is the number of samples, and m is the dimension of the feature vector;

The SVR function is defined as:

f(x)=ωΦ(x)+b (10)

In the formula, f(x) is the output, Φ(x) is the nonlinear mapping function, ω, b are the parameters to be determined, and the following objective function is minimized to solve ω and b:

In the formula, C is the penalty coefficient, f( _xi ) is the predicted value of the i-th sample, and ε represents the maximum error allowed by regression, which is defined as:

后，化为以下目标函数：|yf(x)| _ε ＝max{0,|yf(x)|-ε} (12) introduce the slack variable ξ _i and

After that, it is transformed into the following objective function:

constraint:

Transform equation (10) into solving the dual problem:

为拉格朗日算子，K(x_i,x_j)为核函数，选择线性逼近能力强的RBF核函数，定义为：In the formula, β _i and

is a Lagrangian operator, K( _xi , x _j ) is a kernel function, and the RBF kernel function with strong linear approximation ability is selected, defined as:

In the formula, σ is the width of the kernel function;

Step 5.2, using the sparrow search algorithm to optimize the parameters of the SVR model, constructing the SOH estimation model, and outputting the estimated SOH.

Further, the step 5.2 includes:

Step 5.2.1, using formula (17) to normalize the sample data, the normalized data is divided into training set and test set,

In the formula, Y and Y' are the sample feature values before and after normalization of the training data, respectively, and Y _min and Y _max represent the minimum and maximum values of the sample feature data before normalization, respectively;

Step 5.2.2, setting the parameters of the sparrow search algorithm;

Step 5.2.3, set the penalty coefficient C and the value range of the kernel function parameter σ, and initialize the population;

Step 5.2.4, use the above training set to train the SVR model, and use formula (18) to calculate the fitness value of each sparrow:

为第n个训练样本的预测值，Y_n为第n个训练样本的真实值，N_tr为训练样本数；In the formula,

is the predicted value of the nth training sample, Y _n is the true value of the nth training sample, and N _tr is the number of training samples;

Step 5.2.5, calculate and update the positions of the discoverer, the joiner, and the vigilant respectively according to formulas (19)-(21);

The finder location update formula is:

表示麻雀单体位置，i是1到P的正整数，j是1到d的正整数，P为种群总数，d为特征数据维度，α为0到1之间的随机数，it_max为最大迭代次数，R₂为预警值，ST为安全值，Q为随机数，E为1×d的全1矩阵；当R₂≥ST时，表示被捕食的风险较大，该位置的麻雀将快速转移；否则可扩大搜索范围；In the formula,

Indicates the position of a single sparrow, i is a positive integer from 1 to P, j is a positive integer from 1 to d, P is the total number of populations, d is the feature data dimension, α is a random number between 0 and 1, and it _max is the maximum The number of iterations, R ₂ is the warning value, ST is the safety value, Q is the random number, E is a 1×d matrix of all 1s; when R ₂ ≥ ST, it means that the risk of being predated is high, and the sparrow at this position will quickly transfer; otherwise the search can be expanded;

The update formula for the joiner position is:

和

分别代表全局最不利和最有利的位置；A⁺是A的广义逆矩阵，A表示1×d的矩阵，其中元素随机预设为1或-1；当i＞n/2时，表示跟随者抢夺食物失败，需进行转移；In the formula,

and

Represent the most unfavorable and most advantageous positions in the world; A ⁺ is the generalized inverse matrix of A, and A represents a 1×d matrix, in which elements are randomly preset to 1 or -1; when i>n/2, it means a follower Failed to snatch food and needs to be transferred;

The update formula for the vigilante's position is:

表示当局最有利位置，γ为标准正态随机数，代表步长控制系数；k是-1到1的随机数，代表麻雀的方向；f_g表示当局适应度最大值，f_w和f_g相反，ε是一个较小的常数，以确保分母不为0；f_i表示第i个麻雀的适应度值；In the formula,

Indicates the most favorable position of the authority, γ is a standard normal random number, representing the step control coefficient; k is a random number from -1 to 1, representing the direction of the sparrow; f _g represents the maximum fitness of the authority, and f _w is opposite to f _g , ε is a small constant to ensure that the denominator is not 0; f _i represents the fitness value of the i-th sparrow;

Step 5.2.6, obtain the current updated position, calculate and obtain the optimal individual and the optimal fitness value;

Step 5.2.7. When the training result reaches the set parameter, stop the calculation and execute the output parameter (C, σ), otherwise, return to step 5.2.4 to recalculate the fitness value;

Step 5.2.8. Establish an estimation model according to the parameters (C, σ), and output the estimated SOH.

Compared with prior art, the beneficial effect that the present invention has is reflected in:

The present invention performs feature extraction based on voltage and current data fragments in the charging process, and extracts four highly correlated health features. These health features have strong correlation, high extraction efficiency, and appropriate quantity, which can be applied to online high-precision estimation of lithium-ion battery SOH.

Aiming at the characteristic that the sensor is easily affected by noise during the data collection process, the present invention adopts the Savitzky-Golay filter to process the features with relatively low correlation, eliminate the influence of noise, and further improve the correlation between the proposed features and the target quantity, The prediction accuracy and stability of the model for SOH estimation are improved.

In view of the fact that the machine learning method is more dependent on the setting of model hyperparameters, and improper parameters will lead to insufficient generalization ability of the model, the present invention adopts the support vector regression with less requirements on the amount of data and strong nonlinear regression ability as the benchmark model , and optimize its hyperparameters through the sparrow search algorithm to ensure the validity of the SOH estimation model.

Description of drawings

Fig. 1 is the variation trend figure of four kinds of initial extraction features after normalization in the specific embodiment of the present invention;

Fig. 2 (a) is before and after Savitzky-Golay filtering in the specific embodiment of the present invention F2 feature change trend contrast figure;

Fig. 2 (b) is before and after Savitzky-Golay filtering F4 feature variation trend contrast figure in the specific embodiment of the present invention;

Fig. 3 is the modeling flowchart of SSA-SVR model in the specific embodiment of the present invention;

Fig. 4 is a comparison chart of prediction results of different methods of B0005 battery in the specific embodiment of the present invention;

Fig. 5 is a comparison chart of the prediction results of different methods for the B0007 battery in the specific embodiment of the present invention.

Detailed ways

In order to make the object, technical solution and advantages of the present invention clearer, the present invention will be further described in detail below in combination with specific embodiments and with reference to the accompanying drawings. It should be understood that these descriptions are exemplary only, and are not intended to limit the scope of the present invention. Also, in the following description, descriptions of well-known structures and techniques are omitted to avoid unnecessarily obscuring the concept of the present invention.

This embodiment is a lithium-ion battery SOH estimation method based on the SSA-SVR model, which can achieve high feature extraction efficiency and high estimation accuracy SOH online estimation as the goal, combined with the feature extraction of the charging process segment data, mining different data of voltage and current The information related to battery degradation in the type is combined with feature processing to optimize the relevant features, enhance the correlation between the relevant features and the target quantity, and optimize the SVR model through the sparrow search algorithm after obtaining the input features, which improves the prediction accuracy of the overall framework. Specifically, the estimation method proceeds as follows:

Step one, data collection. Charge and discharge the lithium-ion battery many times, and record the current, voltage, time data and the full capacity of each discharge of the lithium battery during the charge and discharge process.

The lithium-ion battery degradation data used in this example comes from the NASA PCOE Research Center, which includes charge and discharge test data of several commercial lithium-ion 18650 battery packs. The battery dataset records detailed data on voltage, current, and temperature during battery cycle aging experiments. Batteries B05 and B07 were selected as the experimental objects, and their detailed data were recorded in three working modes (charging, discharging and impedance test) at a temperature of 24°C. The rated capacity of batteries B05 and B07 is 2Ah. The charging process is carried out at 1.5A in constant current (CC) mode until the battery voltage reaches 4.2V, and then charged in constant voltage (CV) mode until the charging current drops to 20mA. The discharge process is carried out in 2A constant current (CC) mode until the battery voltages B05 and B07 drop to 2.7V and 2.2V respectively.

Step 2. Feature extraction

As the number of cycles increases, the voltage and current curves will change regularly. As the battery ages, the duration of the same voltage range gradually becomes shorter; the duration of the same current range gradually becomes longer; the voltage and current value are higher after charging for the same time from a certain voltage and current value. To this end, the present embodiment selects the following four health features: the voltage value after charging, the current value after charging, the charging duration of the voltage data segment and the charging duration of the current data segment for SOH estimation.

Step 2.1. Extract the charging duration of the voltage data segment [V _a , V _b ] in the charging process as the first health feature F1. Use formula (1) to calculate feature F1:

F1＝t _b -t _a (1)

In the formula, t _a is the charging time corresponding to the voltage rising to V _a , and t _b is the charging time corresponding to the voltage rising to V _b .

The present invention selects the optimal segment of F1 through the traversal method, with 3.6V as the lower voltage limit, 4.2V as the upper voltage limit, and 0.01V as the change interval for traversal. From the traversal, V _a =3.7V, V _b =4.2V.

Step 2.2, extracting the charging duration of the current data segment [I _c , I _d ] in the charging process as the second health feature F2. Use formula (2) to calculate feature F2:

F2＝t _d -t _c (2)

In the formula, _tc is the charging time corresponding to the current drop to _Ic , and _td is the charging time corresponding to the current drop to _Id .

The present invention selects the optimal segment of F2 through the traversal method, with 0.3A as the lower limit of the current, 1.5A as the upper limit of the current, and 0.01A as the change interval for traversal. From the traversal, I _c =1.49A, I _d =0.42A.

Step 2.3, extracting the constant current (CC) charging process, starting from the voltage _Ve as the timing start point, and taking the voltage value after charging t _e as the third health feature F3. Use formula (3) to calculate feature F3:

F3＝V _e +ΔV _e (3)

Among them, ΔV _e is the voltage value changed within the charging time of t _e .

The present invention selects the optimal segment of F3 through the traversal method, takes 3.6V as the voltage lower limit, 4.2V as the voltage upper limit, 0.01V as the voltage change interval, and 50s as the time change interval for traversal, and can be obtained by traversal, _Ve = 3.72V , t _e =400s.

The change trend graph of the four initial extracted features after normalization is shown in Figure 1.

Step 2.4, extracting the current value after charging t _f from the current I _f as the timing starting point during constant voltage (CV) charging as the fourth health feature F4. Use formula (4) to calculate feature F4:

F4＝I _f +ΔI _f (4)

Among them, ΔI _f is the current value changed within the charging time of t _f .

The present invention selects the optimal segment of F4 through the traversal method, takes 0.3A as the lower limit of the current, 1.5A as the upper limit of the current, 0.01A as the current change interval, and 50s as the time change interval for traversal, which can be obtained by traversal, I _f =1.41V , t _f =1800s.

Step 3, feature processing

Step 3.1, using formula (5) to calculate the correlation between the extracted four feature quantities and the target quantity SOH:

In the formula, X and Y are healthy features and SOH samples, respectively. The larger the absolute value of the correlation coefficient, the stronger the correlation between the feature and SOH, and the higher the accuracy of estimating SOH with this feature.

Step 3.2, adopt Savitzky-Golay filter to build feature processing model, it is determined by polynomial order N and window length L=2M+1, and M represents the absolute value of maximum and minimum measurement point coordinates in the measurement window; For each measurement point is The data of (-M,-M+1,...,0,1,...,M-1,M) is processed using formulas (6), (7) and (8):

In the formula, p(n) is the limit polynomial, a _k represents the coefficient of the polynomial term, ε _N is the residual error of the least squares fitting, x[n] is the real value of the data set in the nth sequence, and the solution is obtained by ( 8) The obtained equation system finally obtains the optimal polynomial coefficients.

Step 3.3, set the threshold to 0.97, when the Pearson correlation coefficient between the feature quantity and SOH is greater than the set threshold, the feature processing model will not be activated, and directly use the original aging features to estimate SOH; when the feature quantity and SOH When the Pearson correlation coefficient is less than the set threshold, it indicates that the feature is greatly affected by sensor drift noise, and the feature is processed to improve the correlation between the feature and the target quantity.

Compared with the features extracted from the voltage data, the features F2 and F4 extracted from the charging current data have lower correlation, and the change trend of the features fluctuates greatly. This is because the charging current data is easily affected by the drift noise of the current sensor, and this fluctuation is unavoidable. Therefore, the present invention uses Savitzky-Golay (SG) filter algorithm to remove the noise of F2 and F4 features. In order to visualize the effect of feature processing, the comparison charts of feature change trends before and after Savitzky-Golay filtering are shown in Figure 2(a) and Figure 2(b).

Step 4. Use the four health features obtained in the above steps as input and the SOH of the battery as output to establish a training set and a prediction set.

Step five, constructing the SSA-SVR model.

Step 5.1, Build a benchmark SVR model

S＝{x _i , y _i |xi _i ∈ R ⁿ , y _i ∈ R}; (i=1,2,...T) (9)

In the formula, x _i is the feature vector of the i-th sample, y _i is the corresponding regression value, T is the number of samples, and n is the dimension of the feature vector.

The SVR function is defined as:

f(x)=ωΦ(x)+b (10)

In the formula, f(x) is the output, Φ(x) is the nonlinear mapping function, and ω,b are the parameters to be determined. Minimize the following objective function to solve for ω and b:

In the formula, C is the penalty coefficient, f( _xi ) is the predicted value of the i-th sample, and ε represents the maximum error allowed by regression, which is defined as:

|yf(x)| _ε ＝max{0,|yf(x)|-ε} (12)

后，可化为以下目标函数：Introducing slack variables ξ _i and

After that, it can be transformed into the following objective function:

constraint:

Transform equation (10) into solving the dual problem:

为拉格朗日算子，K(x_i,x_j)为核函数，本发明选择线性逼近能力强的RBF核函数，定义为：In the formula, β _i and

is a Lagrangian operator, K( _xi , _xj ) is a kernel function, and the present invention selects an RBF kernel function with strong linear approximation ability, defined as:

In the formula, σ is the width of the kernel function.

Step 5.2, using the sparrow search algorithm to optimize the parameters of the SVR model.

The penalty factor C and the kernel function parameter σ are the key parameters of the SVR model, which determine the estimation accuracy and fitting ability of the estimation model. SSA is a new nature-heuristic algorithm proposed based on the behavior of sparrows foraging and avoiding predators. The sparrow in SSA can converge to the current optimal solution by jumping to the vicinity of the current optimal solution, and has excellent performance in terms of accuracy and convergence speed. Using the sparrow search algorithm to optimize the hyperparameters can effectively improve the estimation performance of the model. The steps of SSA-SVR are briefly described as follows:

Step 5.2.1, using formula (17) to normalize the sample data. The normalized data can be divided into training set and test set.

In the formula, Y and Y' are the sample feature values before and after normalization of the training data, respectively, and Y _min and Y _max represent the minimum and maximum values of the sample feature data before normalization, respectively.

Step 5.2.2, set the parameters of the sparrow search algorithm, the maximum number of iterations N = 150, the population size n = 50, the number of discoverers PD = 0.6, the number of vigilantes SD = 0.3, the security value ST = 0.5, the upper and lower limits of independent variables DL=[-7,7];

Step 5.2.3, set the penalty coefficient C and the value range of the kernel function parameter σ, and initialize the population;

Step 5.2.4, use the above training set to train the SVR model, and use formula (18) to calculate the fitness value of each sparrow:

为第n个训练样本的预测值，Y_n为第n个训练样本的真实值，N_tr为训练样本数。In the formula,

is the predicted value of the nth training sample, Y _n is the real value of the nth training sample, and N _tr is the number of training samples.

Step 5.2.5. Calculate and update the positions of the discoverer, the joiner, and the vigilant respectively according to formulas (19)-(21).

The finder location update formula is:

表示麻雀单位位置，i是1到P的正整数，j是1到d的正整数，P为种群总数，d为特征数据维度，α为0到1之间的随机数，it_max为最大迭代次数，R₂为预警值，ST为安全值，Q为随机数，E为1×d的全1矩阵；当R₂≥ST时，表示被捕食的风险较大，该位置的麻雀将快速转移；否则可扩大搜索范围。In the formula,

Indicates the position of the sparrow unit, i is a positive integer from 1 to P, j is a positive integer from 1 to d, P is the total population, d is the feature data dimension, α is a random number between 0 and 1, and it _max is the maximum iteration The number of times, R ₂ is the warning value, ST is the safety value, Q is the random number, and E is a 1×d matrix of all 1s; when R ₂ ≥ ST, it means that the risk of being predated is high, and the sparrow at this position will move quickly ; Otherwise, the search range can be expanded.

When the finder finds better food, there will be joiners to snatch it. If successful, the position of the discoverer changes, otherwise the position of the joiner changes.

The update formula for the joiner position is:

和

分别代表全局最不利和最有利的位，A⁺是A的广义逆矩阵，A表示1×d的矩阵，其中元素随机预设为1或-1；当i＞n/2时，表示跟随者抢夺食物失败，需进行转移。In the formula,

and

Represent the most unfavorable and most advantageous bits in the world respectively, A ⁺ is the generalized inverse matrix of A, and A represents a 1×d matrix, in which elements are randomly preset to 1 or -1; when i>n/2, it means a follower Failed to snatch food, need to transfer.

The update formula for the vigilante's position is:

表示当局最有利位置；γ为标准正态随机数，代表步长控制系数；k是-1到1的随机数，代表麻雀的方向；f_g表示当局适应度最大值，f_w和f_g相反，ε是一个常数，以确保分母不为0；f_i表示第i个麻雀的适应度值。In the formula,

Indicates the most favorable position of the authority; γ is a standard normal random number, representing the step control coefficient; k is a random number from -1 to 1, representing the direction of the sparrow; f _g represents the maximum fitness of the authority, and f _w is opposite to f _g , ε is a constant to ensure that the denominator is not 0; f _i represents the fitness value of the i-th sparrow.

Step 5.2.6, obtain the current updated position, calculate and obtain the optimal individual and the optimal fitness value.

Step 5.2.7. When the training result reaches the set parameter, stop the calculation and execute the output parameter (C, σ), otherwise, return to step 5.2.4 to recalculate the fitness value.

Step 5.2.8. Establish an estimation model according to the parameters (C, σ), and output the estimated SOH.

The modeling flowchart of the SSA-SVR model is shown in Figure 3.

In order to verify the superiority of the method proposed in the present invention, the following three methods are selected for comparison with the actual SOH. Method 1: The method proposed in the present invention includes SG filter optimization features, and SSA is used to optimize the hyperparameters of SVR. Method 2: Include SG filtering, but do not use SSA to optimize the hyperparameters of SVR. Method 3: Use SSA to optimize hyperparameters, but does not include SG filtering. The existence of method two is to verify the superiority of using SSA to optimize hyperparameters, and the existence of method three is to verify the effectiveness of using SG filter optimization feature method. The present invention utilizes two batteries B0005 and B0007 in the NASA data set to carry out experiments. 60% of the battery life cycle is used as the training set, and the rest is used for testing. The prediction starting point of the B0005 and B0007 batteries is 101. The comparison charts of the prediction results of different methods for B0005 and B0007 batteries are shown in Figure 4 and Figure 5.

In summary, the present invention estimates the state of health of lithium-ion batteries based on the SSA-SVR model, and the four features extracted from charging voltage and current curve segments have a high correlation with SOH, laying the foundation for high-precision prediction of the model. For the two features extracted from the charging current segment with noise influence, SG filtering is used for optimization. Experimental results show that this method can reduce the model error by more than 50%. Using SSA to optimize the parameters of the SVR model globally can also reduce the error by more than 50%, effectively improving the accuracy and generalization ability of the model.

It should be understood that the above specific embodiments of the present invention are only used to illustrate or explain the principle of the present invention, and not to limit the present invention. Therefore, any modification, equivalent replacement, improvement, etc. made without departing from the spirit and scope of the present invention shall fall within the protection scope of the present invention. Furthermore, it is intended that the appended claims of the present invention embrace all changes and modifications that come within the scope and metesques of the appended claims, or equivalents of such scope and metes and bounds.

Claims (10)

1. A lithium ion battery SOH estimation method based on an SSA-SVR model is characterized by comprising the following steps:

step 1, data acquisition: carrying out charging and discharging on the lithium ion battery for multiple times, and recording current, voltage and time data of the lithium ion battery in the charging and discharging process and the capacity of complete discharging in each time;

step 2, health feature extraction: extracting four health characteristics from the voltage and current data segments in the charging process, namely a charged voltage value, a charged current value, the charging duration of the voltage data segment and the charging duration of the current data segment;

step 3, constructing a feature processing model based on Savitzky-Golay filtering: selecting a Savitzky-Golay filtering algorithm to process the extracted health characteristics, eliminating messy fluctuation of the health characteristics caused by the influence of the charging data on the drift noise of the sensor, and constructing a characteristic processing model;

step 4, constructing a training set and a prediction set by taking a matrix F = [ F1, F2, F3, F4] formed by the extracted and processed health characteristics as input and the SOH of the battery as output;

and 5, constructing an SSA-SVR model: and constructing a reference SVR model, optimizing the hyper-parameters of the SVR model by adopting a sparrow search algorithm based on the training set, constructing an SOH estimation model, and outputting the estimated SOH.

2. The method for estimating the SOH of the lithium ion battery based on the SSA-SVR model according to claim 1, wherein the step 2 comprises:

step 2.1, extracting a voltage data segment [ V ] in the charging process _a ,V _b ]Is the first healthy feature F1, the feature F1 is calculated using equation (1):

F1＝t _b -t _a (1)

in the formula, t _a For voltage rising to V _a Corresponding charging time, t _b For voltage rising to V _b Corresponding charging time.

3. The method for estimating the SOH of the lithium ion battery based on the SSA-SVR model according to claim 2, wherein the step 2 further comprises:

step 2.2, extracting a current data segment [ I ] in the charging process _c ,I _d ]Is the second healthy feature F2, the feature F2 is calculated using equation (2):

F2＝t _d -t _c (2)

in the formula, t _c For the current to drop to I _c Corresponding charging time, t _d For the current to drop to I _d Corresponding charging time.

4. The method for estimating the SOH of the lithium ion battery based on the SSA-SVR model according to claim 3, wherein said step 2 further comprises:

step 2.3, extracting the secondary voltage V in the constant current charging process _e As a starting point of the time counting, charging t _e The voltage value after the time is taken as a third healthy feature F3, and the feature F3 is calculated by using equation (3):

F3＝V _e +ΔV _e (3)

wherein, is Δ V _e Is t _e The voltage value that changes during the charging time.

5. The method for estimating SOH of a lithium ion battery based on SSA-SVR model according to claim 4, wherein the step 2 further comprises:

step 2.4, extracting the secondary current I in the constant voltage charging process _f As a starting point of time counting, charging t _f The current value after the time is taken as a fourth healthy feature F4, and the feature F4 is calculated by equation (4):

F4＝I _f +ΔI _f (4)

wherein, delta I _f Is t _f The current value that varies during the charging time.

6. The method for estimating the SOH of the lithium ion battery based on the SSA-SVR model according to claim 1, wherein said step 3 comprises:

step 3.1, calculating the correlation between the extracted 4 health characteristic quantities and the target quantity SOH by using the formula (5):

in the formula, X and Y are respectively a health characteristic and an SOH sample.

7. The method for estimating SOH of a lithium ion battery based on SSA-SVR model according to claim 6, wherein the step 3 further comprises:

step 3.2, constructing a feature processing model by adopting a Savitzky-Golay filter, wherein the feature processing model is determined by a polynomial order N and a window length L =2M +1, and M represents the absolute value of the maximum and minimum measurement point coordinates in a measurement window; for data for each measurement point of (-M, -M + 1.., 0,1.., M-1,M), processing is performed using equations (6), (7), and (8):

in which p (n) is a limiting polynomial, a _k Coefficient, epsilon, representing a polynomial term _N For the residual of the least squares fit, x [ n ]]And (4) solving the equation set obtained in the step (8) to obtain the optimal polynomial coefficient.

8. The method for estimating SOH of a lithium ion battery based on SSA-SVR model according to claim 7, wherein the step 3 further comprises:

step 3.3, setting a threshold value to be 0.97, and when the Pearson correlation coefficient between the health characteristic quantity and the SOH is larger than the set threshold value, not activating the characteristic processing model, and directly estimating the SOH by using the original aging characteristic; and when the Pearson correlation coefficient between the health characteristic quantity and the SOH is smaller than a set threshold value, the health characteristic is greatly influenced by the drift noise of the sensor, and the health characteristic is processed to improve the correlation between the health characteristic and the target quantity.

9. The method for estimating SOH of a lithium ion battery based on SSA-SVR model according to claim 1, wherein the step 5 comprises:

step 5.1, constructing a reference SVR model

S＝{x _i ,y _i |x _i ∈R ^m ,y _i ∈R}；i＝1,2,…,T (9)

In the formula, x _i Is the feature vector of the ith sample, y _i Is the corresponding regression value, T is the number of samples, and m is the dimension of the feature vector;

the SVR function is defined as:

f(x)＝ωΦ(x)+b (10)

in the formula, f (x) is output, Φ (x) is a nonlinear mapping function, ω, b are parameters to be determined, and the following objective function is minimized to solve ω and b:

wherein C is a penalty coefficient, f (x) _i ) Is the predicted value of the ith sample, and epsilon represents the maximum error allowed by regression and is defined as:

|y-f(x)| _ε ＝max{0,|y-f(x)|-ε} (12)

introducing a relaxation variable xi _i And

then, it is transformed into the following objective function:

and (3) constraint:

equation (10) is converted to solve the dual problem:

in the formula, beta _i And

for Lagrangian, K (x) _i ,x _j ) Selecting an RBF kernel function with strong linear approximation capability as a kernel function, and defining the RBF kernel function as follows:

where σ is the width of the kernel function;

and 5.2, optimizing the parameters of the SVR model by adopting a sparrow search algorithm, constructing an SOH estimation model, and outputting the estimated SOH.

10. The method for estimating SOH of a lithium ion battery based on SSA-SVR model according to claim 9, wherein the step 5.2 comprises:

step 5.2.1, normalizing the sample data by using the formula (17), wherein the normalized data is divided into a training set and a testing set,

in the formula, Y and Y' are sample characteristic values before and after the normalization of the training data respectively, Y _min And Y _max Respectively representing the minimum value and the maximum value before normalization in the sample characteristic data;

step 5.2.2, setting parameters of a sparrow searching algorithm;

step 5.2.3, setting the value ranges of the penalty coefficient C and the kernel function parameter sigma, and initializing a population;

and 5.2.4, training an SVR model by using the training set, and calculating the fitness value of each sparrow by using a formula (18):

in the formula (I), the compound is shown in the specification,

is the predicted value of the nth training sample, Y _n Is the true value of the nth training sample, N _tr The number of training samples;

step 5.2.5, calculating and respectively updating the positions of the finder, the joiner and the alertness according to the formulas (19) - (21);

the finder location update formula is:

in the formula (I), the compound is shown in the specification,

representing the position of a sparrow monomer, i is a positive integer from 1 to P, j is a positive integer from 1 to d, P is the total number of the population, d is the dimension of characteristic data, alpha is a random number between 0 and 1, it _max For maximum number of iterations, R ₂ Is an early warning value, ST is a safety value, Q is a random number, and E is a 1 × d all-1 matrix; when R is ₂ ST ≧ represents that the risk of being prey is greater, and the sparrow at that position will be rapidly transferred; otherwise, the searching range can be expanded;

the update formula of the subscriber location is:

in the formula (I), the compound is shown in the specification,

and

representing the global least favorable and most favorable bits, A, respectively ⁺ Is a generalized inverse of A, A representing a 1 × d matrix, where the elements are randomly preset to 1 or-1; when i is more than n/2, the follow-up person fails to snatch food and needs to transfer;

the updated formula for the position of the alert is:

in the formula (I), the compound is shown in the specification,

indicating the most favorable position of the authorities; gamma is a standard normal random number and represents a step length control coefficient; k is a random number from-1 to 1, representing the orientation of a sparrow; f. of _g Represents the maximum value of the fitness of the authority, f _w And f _g Conversely, ε is a constant to ensure that the denominator is not 0; f. of _i Representing the fitness value of the ith sparrow;

step 5.2.6, obtaining the current updated position, and calculating to obtain the optimal individual and the optimal fitness value;

5.2.7, when the result after training reaches the set parameter, stopping calculation, and executing output parameter (C, σ), otherwise, returning to 5.2.4 to recalculate the fitness value;

and 5.2.8, establishing an estimation model according to the parameters (C, sigma) and outputting the estimated SOH.

Images