CN114814592A - Lithium battery state of health estimation and remaining service life prediction method and equipment - Google Patents

Abstract

The invention provides a method and equipment for estimating the state of health of a lithium battery and predicting the remaining service life. The method includes: S1: during a battery charging cycle, extracting health features related to capacity degradation on the voltage, current, and temperature curves; S2: verifying the relationship between the extracted health features and battery capacity degradation by using a grey correlation analysis method Indirect health features are obtained by a multi-health feature fusion method; S3: Use the improved gravitational search algorithm to optimize the support vector regression model to estimate the state of health of lithium batteries; S4: Optimize the support vector regression model based on the polynomial regression model and the improved gravitational search algorithm Prediction of the remaining service life of lithium batteries. The health characteristics obtained by the invention have a higher correlation with battery aging, and the calculation amount is small; the RUL is predicted based on the estimation result of the SOH of the lithium battery, the joint prediction of the two is realized, and the battery health status is more comprehensively evaluated.

Description

Lithium battery state of health estimation and remaining service life prediction method and equipment

technical field

Embodiments of the present invention relate to the technical field of batteries, and in particular, to a method and device for estimating the state of health of a lithium battery and predicting its remaining service life.

Background technique

With the popularization of new energy vehicles, the global energy crisis and environmental pollution problems have been effectively alleviated. As the main power source of new energy vehicles, lithium batteries, battery state of health (SOH) and remaining service life (Remaining Useful Life) , RUL) has become the current research focus. SOH is generally expressed as the ratio between the current maximum available capacity of the battery and the initial capacity, and RUL is defined as the number of cycles required for the battery to decay from the current state to the end of life (EOL). Accurate SOH estimation and RUL prediction can effectively avoid overcharge, overdischarge and even thermal runaway. In the existing studies, the separate estimation of SOH and RUL is more common, and the joint estimation of the two is less. However, the battery is a complex dynamic system, and there is a complex coupling relationship between the battery SOH and RUL. SOH is often used as the basis for predicting RUL. Usually, the moment when the SOH drops to 80% is regarded as the end of battery life (End of Life , EOL). The two influence each other in the whole life cycle of the battery, and if only one of the parameters is considered, it may lead to a large estimation error.

The commonly used prediction methods for lithium battery SOH and RUL include model-based methods and data-driven methods. Model-based methods can reflect battery characteristics well, but since the accuracy of estimation is closely related to the accuracy of model parameters, the model The precision requirements are high and the calculation amount is large. The data-driven method does not rely on an accurate mathematical model to describe the battery aging process, and only needs to extract the information reflecting the battery state of health from the historical data through a specific learning algorithm to perform SOH estimation and RUL prediction.

Support Vector Regression (SVR) has become one of the most effective algorithms for estimating battery SOH and RUL in the field of machine learning due to its fast convergence and strong generalization ability. Due to its own parameter settings, if the parameter settings are improper, it will cause a large estimation error. In addition, the selection of Health Feature (HF) is also the key to SOH and RUL prediction. In the existing research work, the battery health feature is mostly extracted from the voltage, current, and temperature curves of the battery during charging and discharging. However, the physical and chemical reactions in the battery discharge stage are very complex, and it is often difficult to obtain a complete charge-discharge curve in practice. Therefore, developing a method and device for estimating the state of health of a lithium battery and predicting the remaining service life, which can effectively overcome the above-mentioned defects in the related technologies, has become an urgent technical problem to be solved in the industry.

SUMMARY OF THE INVENTION

In view of the above problems existing in the prior art, embodiments of the present invention provide a method and device for estimating the state of health of a lithium battery and predicting the remaining service life.

In a first aspect, an embodiment of the present invention provides a method for estimating the state of health of a lithium battery and predicting a remaining service life, including: S1: During a battery charging cycle, extract the health related to capacity degradation on the voltage, current, and temperature curves Features; S2: Use the grey relational analysis method to verify the degree of correlation between the extracted health features and battery capacity degradation, and propose a multi-health feature fusion method to obtain indirect health features; S3: Use the improved gravitational search algorithm to optimize the support vector regression model for lithium Estimation of battery state of health; S4: Remaining service life prediction of lithium battery based on polynomial regression model and improved gravity search algorithm to optimize support vector regression model.

Based on the content of the above method embodiments, in the method for estimating the state of health of a lithium battery and predicting the remaining service life provided in the embodiment of the present invention, in the step S1, the extracted health features include:

Step S101 : in the battery aging process, the capacity directly indicates the aging degree of the battery, and health characteristic parameters related to the capacity degradation are selected to be extracted from the voltage, current, and temperature curves of the battery charging process;

Step S102: extracting the constant current charging time (CCCT) as HF1; extracting the constant voltage charging time (CDCT) in the range of 3.8V to 4.2V in the constant current charging phase as HF2; extracting the temperature change rate in the constant current charging phase ( ROTC) is HF3.

On the basis of the content of the above method embodiments, in the method for estimating the state of health of a lithium battery and predicting the remaining service life provided in the embodiment of the present invention, in the step S2, the gray correlation analysis method and the multi-health feature fusion method specifically include:

Step S201: The battery capacity degradation is regarded as a reference sequence, the health feature is regarded as a comparison sequence, and the gray correlation coefficient of the i-th health feature is:

Among them, z ₀ (k) is the reference sequence, z _i (k) is the comparison sequence, k=1, 2, ..., n, ρ is the resolution coefficient, ρ∈[0,1], the value of ρ is 0.5, the grey relational degree γi takes the average value of the grey relational coefficient ξi( _k ), namely:

Among them, the closer the value of γ _i is to 1, the higher the degree of association between the comparison sequence and the reference sequence;

Step S202: Propose a multi-health feature fusion method to obtain an indirect health feature (Indirect HealthFeature, IHF). The mathematical expression is:

IHF=e*CCCT _N +g*CDCT _N +h*(1-ROTC _N )

where e, g, h are the coefficients of the IHF, and the footnote N indicates that the variable is standardized.

Based on the content of the above method embodiments, in the method for estimating the state of health of a lithium battery and predicting the remaining service life provided in the embodiment of the present invention, in the step S3, the improved gravity search algorithm to optimize the support vector regression model specifically includes:

Step S301: establishing a support vector machine and selecting the radial basis function as the kernel, which is defined as follows:

Among them, x _i is any point in the space, x _j is the center of the kernel function, ||x _i -x _j || ² is the Euclidean distance, and σ represents the width of the kernel function.

Step S302: Use the improved gravitational search algorithm to optimize the penalty factor and kernel function parameters of the support vector regression. In order to avoid the traditional gravitational search algorithm easily falling into the problem of local optimal solution and insufficient memory, the improved algorithm is proposed as follows:

1) The chaotic sequence generates the initial population;

2) Introduce global memory;

Step S303: Divide the battery data into a training set and a test set, and use an improved gravity search algorithm to optimize the support vector regression model to estimate the state of health of the lithium battery.

Based on the content of the above method embodiments, in the method for estimating the state of health of a lithium battery and predicting the remaining service life provided in the embodiment of the present invention, in the step S4, the polynomial regression model specifically includes:

Step S401: When predicting the remaining service life of the lithium battery, the IHF needs to be used as an input, but the IHF of the number of future cycles cannot be measured at present, and a polynomial regression model is used to describe the relationship between the IHF and the number of cycles, so as to predict the IHF at a moment in the future. , the polynomial regression equation is as follows:

y=a ₀ +a ₁ x+a ₂ x ² +…+a _i x ⁱ +μ

Among them, x is the number of cycles, y is the fitted IHF, a _i is the unknown parameter, and μ is the random error.

Step S402: Since there is a certain coupling relationship between the remaining service life of the lithium battery and the state of health, and there is a mapping relationship between the state of health and the IHF, a coupling framework is adopted, and the values of the IHF and the current state of health are used to optimize through an improved gravitational search algorithm. The support vector regression model predicts the remaining service life of lithium batteries, and realizes the joint estimation of battery health status and remaining service life.

In a second aspect, an embodiment of the present invention provides an apparatus for estimating the state of health of a lithium battery and predicting a remaining service life, including: a first main module for extracting the difference between the voltage, the current, and the temperature curve during the battery charging cycle. Health features related to capacity degradation; the second main module is used to verify the degree of correlation between the extracted health features and battery capacity degradation by using the grey relational analysis method, and a multi-health feature fusion method is proposed to obtain indirect health features; the third main module, It is used to use the improved gravity search algorithm to optimize the support vector regression model to estimate the state of health of lithium batteries; the fourth main module is used to optimize the support vector regression model based on the polynomial regression model and the improved gravity search algorithm to predict the remaining service life of lithium batteries.

In a third aspect, an embodiment of the present invention provides an electronic device, including:

at least one processor; and

at least one memory communicatively coupled to the processor, wherein:

The memory stores program instructions executable by the processor, and the processor invokes the program instructions to execute the method for estimating the state of health of the lithium battery and predicting the remaining service life provided by any one of the various implementations of the first aspect.

In a fourth aspect, embodiments of the present invention provide a non-transitory computer-readable storage medium, where the non-transitory computer-readable storage medium stores computer instructions, and the computer instructions cause a computer to execute any one of the various implementations of the first aspect A lithium battery state of health estimation and remaining service life prediction method provided by an implementation manner.

In the method and device for estimating the state of health of a lithium battery and predicting the remaining service life provided by the embodiment of the present invention, a normalized one-dimensional indirect health feature is obtained by a multi-health feature fusion method combined with a gray correlation analysis method. The feature has a higher correlation with battery aging, and the amount of calculation is small; the IGSA-SVR estimation model is proposed, which solves the problems of the traditional support vector regression model that is easy to fall into the local optimal solution and insufficient memory, and improves the estimation accuracy of the model ; Predict the RUL based on the estimation result of the SOH of the lithium battery, realize the joint prediction of the two, and make a more comprehensive assessment of the battery health.

Description of drawings

In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following will briefly introduce the accompanying drawings that need to be used in the description of the embodiments or the prior art. Obviously, the accompanying drawings in the following description These are some embodiments of the present invention. For those of ordinary skill in the art, other drawings can also be obtained according to these drawings without creative efforts.

1 is a flowchart of a method for estimating the state of health of a lithium battery and predicting a remaining service life according to an embodiment of the present invention;

2 is a schematic structural diagram of an apparatus for estimating state of health and predicting remaining service life of a lithium battery according to an embodiment of the present invention;

3 is a schematic diagram of a physical structure of an electronic device provided by an embodiment of the present invention;

4 is a flowchart of another method for estimating the state of health of a lithium battery and predicting the remaining service life according to an embodiment of the present invention;

FIG. 5 is a schematic diagram of the result of estimating SOH on a B0005 lithium battery according to an embodiment of the present invention;

FIG. 6 is a schematic diagram of a result of predicting RUL on a B0005 lithium battery according to an embodiment of the present invention.

Detailed ways

In order to make the purposes, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments These are some embodiments of the present invention, but not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention. In addition, the technical features in each embodiment or a single embodiment provided by the present invention can be arbitrarily combined with each other to form a feasible technical solution. This combination is not restricted by the sequence of steps and/or the structural composition mode, but must be in the order of Those of ordinary skill in the art can realize based on that, when the combination of technical solutions is contradictory or cannot be realized, it should be considered that the combination of such technical solutions does not exist and is not within the protection scope of the present invention.

An embodiment of the present invention provides a method for estimating the state of health and predicting the remaining service life of a lithium battery. Referring to FIG. 1 , the method includes: S1 : during the charging cycle of the battery, extract the voltage, current, and temperature curves related to capacity degradation. Health features; S2: Use the grey relational analysis method to verify the degree of correlation between the extracted health features and battery capacity degradation, and propose a multi-health feature fusion method to obtain indirect health features; S3: Use the improved gravity search algorithm to optimize the support vector regression model. Lithium battery state of health estimation; S4: Based on polynomial regression model and improved gravity search algorithm to optimize the support vector regression model (ie IGSA-SVR) to predict the remaining service life of lithium batteries.

Specifically, the present invention proposes a lithium battery SOH estimation and RUL prediction method based on multi-health feature fusion. Some parameters are shown in Table 1.

The basic information and operating parameters of the selected battery are shown in Table 1 (part of the parameters of the experimental lithium-ion battery).

Table 1

battery number Ambient temperature/℃ Discharge current/A Cut-off voltage/V End of life/Ah B0005 twenty four 2 2.7 1.40 B0006 twenty four 2 2.5 1.40 B0007 twenty four 2 2.2 1.45 B0018 twenty four 2 2.5 1.40

Based on the content of the above method embodiment, as an optional embodiment, in the method for estimating the state of health of a lithium battery and predicting the remaining service life provided in the embodiment of the present invention, in the step S1, the extracted health features include:

Step S101: Collect historical operating data of the battery to be tested, and define the battery SOH from the perspective of capacity:

In the battery aging process, the capacity can directly indicate the aging degree of the battery, but it is not easy to measure directly, so the present invention chooses to extract the health characteristic parameters related to the capacity degradation from the voltage, current and temperature curves of the battery charging process.

Step S102 : extracting the constant current charging time (CCCT) as HF1; extracting the constant current charging stage voltage in the range of 3.8V to 4.2V as the constant voltage charging time (CDCT) as HF2; extracting the rate of temperature change (ROTC) in the CC stage as HF3.

Based on the content of the above method embodiment, as an optional embodiment, in the method for estimating the state of health of a lithium battery and predicting the remaining service life provided in the embodiment of the present invention, in the step S2, the gray correlation analysis method and the The health feature fusion method specifically includes:

Step S201 : Use the grey correlation analysis method to give the correlation degree evaluation between the extracted health features and battery capacity degradation. Taking battery capacity degradation as a reference sequence and health features as a comparison sequence, the grey correlation coefficient of the i-th health feature is:

Among them, z ₀ (k) is the reference sequence, z _i (k) is the comparison sequence k=1, 2,...,n, ρ is the resolution coefficient, ρ∈[0,1], the value of ρ is 0.5, gray The correlation degree γ _i takes the average value of the grey correlation coefficient ξ _i (k), namely:

Among them, the closer the value of γ _i is to 1, the higher the degree of association between the comparison sequence and the reference sequence. The gray correlation values between the three health characteristics and battery capacity are shown in Table 2. (Grey correlation of selected health characteristics to battery capacity).

Table 2

battery number HF1 HF2 HF3 B0005 0.8616 0.9133 0.8296 B0006 0.7745 0.9073 0.7861 B0007 0.8917 0.9044 0.7411 B0018 0.9328 0.8886 0.7616

Step S202: Propose a multi-health feature fusion method to obtain an indirect health feature (Indirect Health Feature, IHF). The mathematical expressions are shown in equations (4) and (5).

IHF=e*CCCT _N +g*CDCT _N +h*(1-ROTC _N ) (4)

where e, g, h are the coefficients of IHF, and footnote N indicates that the variable is standardized.

Based on the content of the above method embodiment, as an optional embodiment, in the method for estimating the state of health of a lithium battery and predicting the remaining service life provided in the embodiment of the present invention, in step S3, the improved gravitational search algorithm optimizes the support vector The regression model specifically includes:

Step S301: Support Vector Regression (SVR) is a machine learning algorithm used to deal with regression problems, and a training sample set is given: D={(x ₁ , y ₁ ), (x ₂ , y ₂ ), ... , (x _i , y _i )}∈R ^N ×R

where x _i , y _i , and n are the input vector, output value, and dataset, respectively.

Map the input vector _xi to a linear function f( _xi ), and make f( _xi ) and _yi as close as possible, f( _xi ) is defined as follows:

y _i =f(x _i )=(ω×x _i )+b (6)

where ω is the weight vector and b is the displacement term.

The error function δ is defined as:

定义为第i个样本上边界与下边界的松弛变量。引入拉格朗日乘子，得到最终的回归函数：Among them, δ is the objective function to minimize, C is the penalty factor, δ _i ,

Defined as the slack variable for the upper and lower bounds of the ith sample. Introduce Lagrange multipliers to get the final regression function:

α为拉格朗日系数，K(x_i，x_j)为核函数，本发明选用径向基函数作为内核，定义如下：in,

α is the Lagrangian coefficient, K(x _i , x _j ) is the kernel function, the present invention selects the radial basis function as the kernel, and is defined as follows:

Among them, x _i is any point in the space, x _j is the center of the kernel function, ||x _i -x _j || ² is the Euclidean distance, and σ is the width of the kernel function.

Step S302: The Gravity Search Algorithm (GSA) regards the solution of the optimization problem as a group of particles moving in space. When the particle moves to the optimal position, it is the optimal solution of the problem and the update equation of particle velocity and position. for:

分别为t时刻粒子i在d维空间的速度、加速度和位置。为避免传统引力搜索算法易陷入局部最优解和内存不足的问题，提出改进算法。where R is a random number in the range of [0, 1],

are the velocity, acceleration and position of particle i in d-dimensional space at time t, respectively. In order to avoid the problem that the traditional gravitational search algorithm is easy to fall into the local optimal solution and insufficient memory, an improved algorithm is proposed.

1) The chaotic sequence generates the initial population

First generate a d-dimensional random vector:

将ρ₀作为初始迭代值，由logistic映射得到方程：in,

Taking ρ ₀ as the initial iteration value, the equation is obtained by logistic mapping:

where μ=4, i=1, 2, . . . , n. The traversal range of the chaotic sequence is mapped to the search interval of the optimization variable to get:

Among them, low is the lower limit of the value, and up is the upper limit of the value.

2) Introduce global memory

As the iteration progresses, GSA may cause the particle with the most fitness to be attracted by other particles and lose because there is not enough memory to save all the optimal solutions so far. To overcome this shortcoming, a global memory g _best is introduced to memorize the optimal solution obtained so far. The velocity equation can be improved to:

和

分别为t时刻粒子i和g_best在d维空间的位置，c₁为加速度系数，t为当前迭代次数，T为最大的迭代次数。in,

and

are the positions of particles i and g _best in the d-dimensional space at time t, respectively, c ₁ is the acceleration coefficient, t is the current number of iterations, and T is the maximum number of iterations.

Step S303: The first 80 groups of data in the cycle experiment of the four batteries in Table 1 are used as the training set, and the rest of the data are used as the test set, that is, the first 80 cycles of B0005, B0006, and B0007 are used as the training set and the last 88 cycles are used as the test set. The first 80 cycles are used as the training set and the next 42 cycles are used as the test set, and the SOH estimation of the lithium battery is performed by the IGSA-SVR model, as shown in Figure 3.

Based on the content of the above method embodiment, as an optional embodiment, in the method for estimating the state of health and predicting the remaining service life of a lithium battery provided in the embodiment of the present invention, in the step S4, the polynomial regression model specifically includes:

Step S401: When predicting the remaining service life of the lithium battery, the IHF needs to be used as an input, but the IHF of the number of future cycles cannot be measured at present, and a polynomial regression model is used to describe the relationship between the IHF and the number of cycles, so as to predict the IHF at a moment in the future. , the polynomial regression equation is as follows:

y=a ₀ +a ₁ x+a ₂ x ² +…+a _i x ⁱ +μ (17)

Among them, x is the number of cycles, y is the fitted IHF, a _i is the unknown parameter, and μ is the random error.

Step S402: Since there is a certain coupling relationship between the remaining service life of the lithium battery and the state of health, and there is a mapping relationship between the state of health of the battery and the IHF, a coupling framework is adopted, using the values of the IHF and the current state of health, through the improved support vector The regression model predicts the remaining service life of the lithium battery, and realizes the joint estimation of the battery state of health and the remaining service life, as shown in Figure 2. The end-of-life threshold of the lithium battery is set to 0.75, that is, when the SOH of the battery is less than 0.75, the battery life is considered to be terminated, as shown in Figure 4.

In the method for estimating the state of health and predicting the remaining service life of a lithium battery provided by the embodiment of the present invention, a normalized one-dimensional indirect health feature is obtained by a multi-health feature fusion method combined with a gray correlation analysis method. The health feature obtained by the method is the same as the The battery aging has a higher degree of correlation, and the amount of calculation is small; the IGSA-SVR estimation model is proposed, which solves the problems of the traditional support vector regression model that it is easy to fall into the local optimal solution and insufficient memory, and improves the estimation accuracy of the model; based on The estimated result of lithium battery SOH predicts RUL, realizes the joint prediction of the two, and enables a more comprehensive assessment of battery health.

The realization basis of each embodiment of the present invention is realized through programmed processing performed by a device having a processor function. Therefore, in practical engineering, the technical solutions and functions of the various embodiments of the present invention can be encapsulated into various modules. Based on this reality, and on the basis of the above embodiments, the embodiments of the present invention provide an apparatus for estimating the state of health of a lithium battery and predicting a remaining service life. State estimation and remaining useful life prediction methods. Referring to FIG. 5 , the device includes: a first main module for extracting health features related to capacity degradation on the voltage, current and temperature curves during a battery charging cycle; a second main module for using grey relational analysis method Verify the degree of correlation between the extracted health features and battery capacity degradation, and propose a multi-health feature fusion method to obtain indirect health features; the third main module is used to optimize the support vector regression model using the improved gravity search algorithm for lithium battery state of health estimation ; The fourth main module is used to predict the remaining service life of lithium batteries based on the polynomial regression model and the improved gravity search algorithm to optimize the support vector regression model.

The method in the embodiment of the present invention is implemented by relying on electronic equipment, so it is necessary to introduce the related electronic equipment. For this purpose, an embodiment of the present invention provides an electronic device, as shown in FIG. 6 , the electronic device includes: at least one processor (processor), a communication interface (Communications Interface), at least one memory (memory) and a communication device A bus, wherein at least one processor, a communication interface, and at least one memory communicate with each other through the communication bus. At least one processor may call logic instructions in at least one memory to execute all or part of the steps of the methods provided by the foregoing method embodiments.

In addition, the above-mentioned logic instructions in the at least one memory can be implemented in the form of software functional units and can be stored in a computer-readable storage medium when sold or used as an independent product. Based on this understanding, the technical solution of the present invention can be embodied in the form of a software product in essence, or the part that contributes to the prior art or the part of the technical solution. The computer software product is stored in a storage medium, including Several instructions are used to cause a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the various method embodiments of the present invention. The aforementioned storage medium includes: U disk, removable hard disk, Read-Only Memory (ROM, Read-Only Memory), Random Access Memory (RAM, Random Access Memory), magnetic disk or optical disk and other media that can store program codes .

From the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary general hardware platform, and certainly can also be implemented by hardware. Based on this understanding, the above-mentioned technical solutions can be embodied in the form of software products in essence or the parts that make contributions to the prior art, and the computer software products can be stored in computer-readable storage media, such as ROM/RAM, magnetic A disc, an optical disc, etc., includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to perform the methods described in various embodiments or some parts of the embodiments.

The flowchart and block diagrams in the Figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present invention. With this recognition, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which contains one or more functions for implementing the specified logical function(s) executable instructions. It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It is also noted that each block of the block diagrams and/or flowchart illustrations, and combinations of blocks in the block diagrams and/or flowchart illustrations, can be implemented in dedicated hardware-based systems that perform the specified functions or actions , or can be implemented in a combination of dedicated hardware and computer instructions.

In this patent, the terms "comprising", "comprising" or any other variation thereof are intended to encompass a non-exclusive inclusion such that a process, method, article or device comprising a series of elements includes not only those elements, but also Include other elements not expressly listed, or which are inherent to such a process, method, article or apparatus. Without further limitation, an element defined by the phrase "comprises" does not preclude the presence of additional identical elements in a process, method, article or apparatus that includes the element.

Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, but not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that it can still be The technical solutions described in the foregoing embodiments are modified, or some technical features thereof are equivalently replaced; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims (8)

1. A lithium battery health state estimation and remaining service life prediction method is characterized by comprising the following steps: s1: extracting health characteristics related to capacity degradation on voltage, current and temperature curves in a battery charging period; s2: verifying the correlation degree between the extracted health characteristics and the battery capacity degradation by using a grey correlation analysis method, and providing a multi-health characteristic fusion method to obtain indirect health characteristics; s3: optimizing a support vector regression model by utilizing an improved gravity search algorithm to estimate the health state of the lithium battery; s4: and optimizing the lithium battery residual service life prediction of the support vector regression model based on the polynomial regression model and the improved gravity search algorithm.

2. The method for estimating the state of health and predicting the remaining service life of a lithium battery as claimed in claim 1, wherein the extracted health features in step S1 include:

step S101: in the battery aging process, the capacity directly indicates the aging degree of the battery, and health characteristic parameters related to capacity degradation are extracted from voltage, current and temperature curves in the battery charging process;

step S102: extracting the constant current charging time as HF 1; extracting equal differential charging time of the voltage in the constant current charging stage within the interval of 3.8V to 4.2V as HF 2; the temperature change rate in the extracted constant current charging phase is HF 3.

3. The method for estimating the state of health and predicting the remaining service life of the lithium battery as claimed in claim 2, wherein in step S2, the gray correlation analysis method and the multi-health-feature fusion method specifically include:

step S201: considering the battery capacity degradation as a reference sequence and the health features as a comparison sequence, the gray correlation coefficient of the ith health feature is:

wherein z is ₀ (k) As a reference sequence, z _i (k) For comparison of sequences, k is 1, 2]Where ρ is 0.5 and the gray level of correlation γ is _i Taking a grey correlation coefficient xi _i (k) I.e.:

wherein, γ _i The closer to 1, the higher the degree of association between the comparison sequence and the reference sequence;

step S202: providing a multi-health characteristic fusion method to obtain an indirect health characteristic IHF, wherein the mathematical expression is as follows:

wherein e, g, h are weight coefficients of IHF, and the subscript N indicates that the variable is normalized.

4. The method for estimating the state of health and predicting the remaining service life of a lithium battery as claimed in claim 3, wherein in the step S3, the optimizing the support vector regression model by the improved gravity search algorithm specifically comprises:

step S301: establishing a support vector regression model, selecting a radial basis function as a kernel, and defining the following parameters:

wherein x is _i Is any point in space, x _j Is the kernel center, | x _i -x _j || ² Is the euclidean distance, σ represents the kernel function width;

step S302: the improved gravity search algorithm is adopted to optimize the penalty factor and the kernel function parameter of the support vector regression, and in order to avoid the problems that the traditional gravity search algorithm is easy to fall into the local optimal solution and the memory is insufficient, the improved algorithm is provided as follows:

1) generating an initial population by the chaotic sequence;

2) introducing global memory;

step S303: and dividing the battery data into a training set and a testing set, and optimizing a support vector regression model by an improved gravity search algorithm to estimate the health state of the lithium battery.

5. The method for estimating the state of health and predicting the remaining service life of a lithium battery as claimed in claim 4, wherein in the step S4, the polynomial regression model specifically includes:

step S401: when the residual service life of the lithium battery is predicted, indirect health characteristics are required to be used as input, but the IHF of future cycle number cannot be measured at present, a polynomial regression model is adopted to describe the relation between the IHF and the cycle number, so that the IHF at a future moment is predicted, and the polynomial regression equation is as follows:

y＝a ₀ +a ₁ x+a ₂ x ² +…+a _i x ⁱ +μ

wherein x is the cycle number, y is the IHF after fitting, a _i μ is a random error for unknown parameters;

step S402: because a certain coupling relation exists between the remaining service life and the health state of the lithium battery, and a mapping relation exists between the health state and the IHF, a coupling frame is adopted, the value of the IHF and the current health state is utilized, and the remaining service life of the lithium battery is predicted by optimizing a support vector regression model through an improved gravity search algorithm, so that the joint estimation of the battery health state and the remaining service life is realized.

6. A lithium battery state of health estimation and remaining service life prediction device is characterized by comprising: the first main module is used for extracting health characteristics related to capacity degradation on voltage, current and temperature curves in a battery charging period; the second main module is used for verifying the correlation degree between the extracted health characteristics and the battery capacity degradation by utilizing a grey correlation analysis method, and providing a multi-health characteristic fusion method to obtain indirect health characteristics; the third main module is used for optimizing a support vector regression model by utilizing an improved gravitation search algorithm to estimate the health state of the lithium battery; and the fourth main module is used for optimizing the lithium battery residual service life prediction of the support vector regression model based on the polynomial regression model and the improved gravitation search algorithm.

7. An electronic device, comprising:

at least one processor, at least one memory, and a communication interface; wherein,

the processor, the memory and the communication interface are communicated with each other;

the memory stores program instructions executable by the processor, the processor invoking the program instructions to perform the method of any of claims 1 to 5.

8. A non-transitory computer-readable storage medium storing computer instructions for causing a computer to perform the method of any one of claims 1 to 5.

Images