Lithium battery SOC-SOH joint estimation method based on mechanism and data fusion driving

Abstract

The invention discloses a lithium battery SOC-SOH joint estimation method based on mechanism and data fusion driving, and the method comprises the following steps: firstly, carrying out the training of a prediction model of a least square support vector machine (LSSVM, LSSVM) through a multi-dimensional data set after phase space reconstruction; and then, establishing a state equation based on an ampere-hour integral method and a capacity fading empirical model, establishing an IUPF observation equation through an LSSVM prediction model, and substituting the IUPF observation equation into an improved unscented particle filter algorithm (IUPF, Intermediate UPF) to perform state estimation and parameter updating. And finally, the updated LSSVM model parameters are iterated to a training data set, so that the combined prediction of the SOC (State OfCharge) and the SOH (State OfHealth) of the battery and the online updating of the data driving model are realized. According to the lithium battery SOC-SOH joint estimation method based on mechanism and data fusion driving provided by the invention, joint estimation of the state of charge SOC and the state of health SOH of the lithium battery is realized, and state noise and measurement noise of the system are considered, so that the precision and stability of a battery state estimation result are ensured.

Description

Technical Field

[0001] The invention relates to a lithium battery SOC-SOH joint estimation method based on mechanism and data fusion drive, belonging to the technical field of power energy storage. Background Art

[0002] With the continuous advancement of the dual carbon goals, the installed capacity of intermittent new energy power generation represented by wind power and photovoltaic power generation in the power system has increased significantly, and the demand for flexible regulation such as frequency regulation and peak regulation of the power system has also increased. The conventional pumped storage regulation mode cannot be built and developed on a large scale due to the constraints of the geographical environment, while the electrochemical energy storage system represented by lithium-ion batteries has become an important regulation resource for the new power system with new energy as the main body due to its flexible configuration, easy installation and rapid regulation. At the same time, due to the closed physical structure of lithium batteries as electrochemical reaction systems, it is very difficult to monitor and diagnose their operating status. Traditional lithium battery state of charge estimation (SOC) is mainly based on model-based methods combined with Kalman filter series algorithms. It is limited in terms of environmental dependence and model accuracy, and has problems such as low estimation accuracy, weak model generalization ability, and sensitivity to initial conditions. Most lithium battery health state estimation (SOH) uses neural network algorithms, but neural networks may not be applicable to other types or batches of batteries. As a "black box" model, its internal decision-making process is difficult to explain and limited, resulting in problems in the generalization ability and interpretability of the model. In addition, the separate battery state estimation of lithium battery SOC and SOH ignores the coupling between the two, making the speed and accuracy of SOC and SOH estimation unable to meet the needs of practical applications. Especially in some extreme accidents, thermal inducements, mechanical collisions, and failure to provide timely warnings, which leads to the spread and expansion of battery system failures, causing great harm to people's lives and property safety. Therefore, it is necessary to consider the coupling relationship between lithium battery operation SOC and SOH, combine its mechanism model with the data model, and carry out joint estimation of lithium battery SOC and SOH, so as to further improve the level of lithium battery operation status monitoring, identification, diagnosis and early warning. Summary of the invention

[0003] The purpose of the present invention is to provide a lithium battery SOC-SOH joint estimation method based on mechanism and data fusion driven in view of the above-mentioned defects and shortcomings of the prior art, based on the minimum vector machine-improved unscented Kalman particle filter (LSSVM-IUPF) framework, using a three-layer IUPF algorithm, with the optimal state estimation parameters as the goal, to achieve SOC-SOH joint accurate estimation, so as to achieve the purpose of reducing the estimation error of battery health state SOH and battery capacity state SOC.

[0004] The technical solution adopted by the present invention to solve the technical problem is: a lithium battery SOC-SOH joint estimation method based on mechanism and data fusion drive, comprising the following steps:

[0005] Step 1: Construct a multidimensional dataset of the time series of the lithium battery SOC value and the current maximum available capacity respectively, reconstruct the training data from the initial moment to the k-1 moment using the Takens embedding method, and generate a d-dimensional vector in the phase space.

[0006] Step 2: Phase space reconstruction: reconstruct the multidimensional data set into a phase space with the same topological meaning as the dynamical system.

[0007] Step 3: Perform offline training on the reconstructed least squares support vector machine (LSSVM) prediction model.

[0008] Step 4: Establish the state equations for SOC and SOH estimation respectively according to the ampere-hour integration method and the battery capacity degradation empirical model; establish the observation equations for SOC and SOH estimation respectively through the LSSVM online prediction model.

[0009] Step 5: Train the offline model of LSSVM based on the training data from the initial moment to the k-1 moment, and use the improved unscented particle filter algorithm (IUPF, Intermediate UPF) to estimate the LSSVM parameters of the regression prediction model at the k moment.

[0010] Step 6: When the IUPF algorithm observes new data at time k, the system uses an incremental learning method to learn and update only the information of the new data, update the particle weights, and combine the LSSVM online model to give further prediction values ​​of SOC, SOH, and LSSVM parameters.

[0011] Step 7: The updated LSSVM model parameters are iterated to the training dataset for state estimation at time k+1, thereby achieving joint prediction of battery SOC and SOH and online update of the data-driven model.

[0012] Wherein, step 4 also includes the following steps:

[0013] Step 4.1: Establish the IUPF state equations for SOC and SOH estimation based on the ampere-hour integration method and the battery capacity degradation empirical model respectively;

[0014] Step 4.2: The SOC and SOH obtained by offline LSSVM training are used as virtual observations, and the LSSVM regression model is used as the observation equation.

[0015] Furthermore, the above method of establishing the model state equation, the state equation for SOC estimation established based on the ampere-hour integration method is discretized as follows:

[0016]

[0017] SOC k is the SOC value of the lithium battery at time k; Δt k is the system sampling time; C k is the available capacity of the battery at time k; I k is the system input current at time k.

[0018] Furthermore, the above method of establishing the model observation equation establishes a discretized representation of the state equation of capacity based on the empirical model of capacity decay as follows:

[0019] C k+1 =η c C k +β1exp(-β2 / Δt k )+ξ k

[0020] In the formula, C k is the available capacity of the battery at time k; U oc,k is the open circuit voltage; k is the state noise; β1 and β2 are coefficients to be determined.

[0021] Furthermore, the SOC-SOH joint estimation method based on multi-method fusion, the health status SOH of lithium-ion batteries is defined as follows:

[0022]

[0023] In the formula, SOH k is the SOH value calculated at the kth moment, C new is the factory capacitance of the battery; C current The maximum available capacity of the battery.

[0024] Furthermore, the SOC-SOH joint estimation method based on multi-method fusion, the state equation discretization of SOC and SOH constructed by LSSVM observation equation is as follows:

[0025] y k =f(x k-1 ,s k )+ι k

[0026] In the formula, x i is the state vector, s k is the LSSVM parameter, l k is Gaussian white noise.

[0027] Wherein, step 6 also includes the following steps:

[0028] Step 6.1: Initialize n initial state particles x k ;

[0029] Step 6.2: Apply the improved UKF algorithm to each sampled particle for weight optimization;

[0030] Step 6.3: Calculate the importance weight of the particles and normalize them;

[0031] Step 6.4: Use the resampling algorithm to copy and eliminate particles and recalculate weights;

[0032] Step 6.5: Estimating the state of the corresponding SOC or SOH;

[0033] Step 6.6: Perform online estimation of LSSVM parameter updates.

[0034] A SOC-SOH joint estimation model system, comprising:

[0035] The state equations of the SOC and current capacity of the lithium-ion battery are constructed respectively using the ampere-hour integration method and the battery capacity decay method. Then, the IUPF-LSSVM framework is used to iteratively realize the online training data-driven model and further prediction of the battery state parameters. The state estimation of the SOC value and SOH value is realized using the optimal state estimation module.

[0036] Furthermore, the above-mentioned SOC-SOH joint estimation model system, the IUPF-LSSVM framework module, is specifically used to implement the online training data-driven model and further prediction of battery parameters through iteration.

[0037] Furthermore, in the above-mentioned SOC-SOH joint estimation model system, the optimal state estimation module outputs the estimated SOH value and SOC value.

[0038] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the SOC-SOH joint estimation method based on multi-method fusion as described above.

[0039] A device (controller / computer device / electronic device) includes: a memory for storing data; a processor for executing the algorithm instructions, so that the device performs the operation of a lithium battery SOC-SOH joint estimation method driven by a mechanism and data fusion as described above.

[0040] Beneficial effects:

[0041] 1. The present invention proposes a joint state estimation model of the battery's current capacity state SOC and health state SOH, and independently designs the LSSVM-IPUF algorithm, which overcomes the problems of low accuracy and slow convergence speed of estimation of a single algorithm, achieves higher accuracy in estimating the SOC and SOH values ​​of lithium batteries, and ensures stable operation of the battery.

[0042] 2. The present invention aims to enable lithium-ion batteries to accurately estimate the charge state and health state due to their own possible overcharging or over-discharging problems based on the technical concept of multi-method integration, optimize the limitations of a single algorithm, and use a double-layer algorithm for more accurate estimation, so that the estimation of the current capacity and health state of lithium-ion batteries has higher accuracy and robustness, ensure the safe and flexible operation of the power system, and provide technical support for the efficient use of lithium batteries.

[0043] 3. The system architecture proposed in the present invention updates the SOC value and terminal voltage in real time and substitutes them into the SOH estimation, and updates the ohmic internal resistance in real time and substitutes it into the SOC state estimation, thereby reducing the SOH estimation error caused by the increase of the ohmic internal resistance due to the thickening of the SEI film layer due to multiple charges and discharges, and realizes the two-way flow of information, so that the state estimation can still have the advantages of stability and small estimation error under complex working conditions. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 Shown is a schematic diagram of a SOC-SOH joint estimation block diagram based on the LSSVM-IUPF algorithm of the present invention;

[0045] Figure 2 Shown is a schematic diagram of the LSSVM-IUPF algorithm flow of the present invention. DETAILED DESCRIPTION

[0046] It should be noted that:

[0047] The experimental object is Panasonic 18650 ternary lithium battery with a rated capacity of 1800mAh, a maximum voltage of 4.2V and a cut-off voltage of 2.5V.

[0048] The technical solution of the present invention is described in detail below through the accompanying drawings and specific embodiments. It should be understood that the embodiments of the present invention and the specific features in the embodiments are detailed descriptions of the technical solution of the present invention, rather than limitations on the technical solution of the present invention. The embodiments of the present invention and the technical features in the embodiments may be combined with each other unless there is a conflict.

[0049] Example 1

[0050] This embodiment provides a mechanism and data fusion driven LSSVM-IUPF framework, including the following steps:

[0051] Step 1: Construct a multidimensional dataset of the time series of the lithium battery SOC value and the current maximum available capacity, reconstruct the training data from the initial moment to the k-1 moment using the Takens embedding method, and generate a d-dimensional vector in the phase space;

[0052] Among them, step 1 includes the following contents:

[0053] Step 1.1: Construct a multidimensional dataset of the lithium battery SOC value and the current maximum available capacitance value, and initialize N initial state particles. Among them, i represents the i-th particle, represents the SOC value at the initial moment, represents the current maximum available capacitance value at the initial moment, and x k Obey the proposed distribution function;

[0054] Step 1.2: Takens embedding method is as follows:

[0055] Taking the one-dimensional nonlinear relationship between the SOC value and time of lithium batteries and the maximum available capacitance and time, Takens proposed that the observed seemingly random data, the phase trajectory of the chaotic system still has a certain structure, which can be used to identify and analyze chaotic behavior, assuming that the univariate time series under the nonlinear dynamic system When the embedding dimension d of the phase space is greater than a certain value, there will be a smooth mapping f(.), and the generated d-dimensional vector is as follows:

[0056] z k+1 =f(z k ,z k-1 ,z k-2 …,z k-d+1 )

[0057] Among them, z k Represents the particle training data at time k, and the value range of k is 1 to N;

[0058] Step 2: Phase space reconstruction: reconstruct the multidimensional data set into a phase space with the same topological meaning as the dynamical system;

[0059] Among them, step 2 includes the following contents:

[0060] Step 2.1: Reconstruct the multidimensional data set into a phase space with the same topological meaning as the dynamical system. The phase space reconstruction process is as follows:

[0061]

[0062] Among them, x n is the input vector of the smooth mapping f(.), (z d+1 ,z d+2 ,...,z N ) T is the output vector corresponding to the smooth mapping f(.);

[0063] Step 2.2: Use the false neighbor method to determine the minimum dimension d of phase space reconstruction.

[0064] Step 3: Put the reconstructed data into the LSSVM prediction model for offline training;

[0065] Among them, step 3 includes the following contents:

[0066] Step 3.1: According to statistical theory and Gaussian kernel function, the LSSVM model can be expressed as:

[0067]

[0068] Where α=[α1,α2,...,α n ] is the support vector factor, x k =(x1,x2,...,x N ) is the support vector parameter and b is the bias.

[0069] in addition, σ represents the radial basis kernel width;

[0070] Step 3.2: Bringing in the reconstructed data, we can derive a further prediction model:

[0071]

[0072] Step 4: Establish the state equations for SOC and SOH estimation respectively according to the ampere-hour integration method and the battery capacity degradation empirical model; establish the observation equations for SOC and SOH estimation respectively through the LSSVM online prediction model.

[0073] Wherein, step 4 comprises the following steps:

[0074] Step 4.1: Establish the IUPF state equations for SOC and SOH estimation based on the ampere-hour integration method and the battery capacity degradation empirical model respectively;

[0075] Furthermore, the discretization of the state equation for estimating the current capacity SOC of the lithium-ion battery based on the ampere-hour integration method in step 4.1 is as follows:

[0076]

[0077] SOC kis the SOC value of the lithium battery at time k; Δt k is the system sampling time; C k is the available capacity of the battery at time k; I k is the system input current at time k.

[0078] Furthermore, the state equation for estimating the current capacity of the lithium-ion battery based on the empirical equation of battery capacity degradation in step 4.1 is discretized as follows:

[0079] C k+1 =η c C k +β1exp(-β2 / Δt k )+ξ k

[0080] Among them, C k is the available capacity of the battery at time k; k is the state noise; β1 and β2 are coefficients to be determined.

[0081] Step 4.2: The SOC and SOH obtained by offline LSSVM training are used as virtual observations, and the LSSVM regression model is used as the observation equation.

[0082] Furthermore, the LSSVM observation equation of step 4.4 is as follows:

[0083] y k =h(x k-1 ,s k )+ι k

[0084] Among them, x i is the state vector, s k is the LSSVM parameter, l k is Gaussian white noise.

[0085] Step 5: Train the offline model of LSSVM based on the training data from the initial moment to the k-1 moment, and use the improved unscented particle filter algorithm (IUPF, Intermediate UPF) to estimate the LSSVM parameters of the regression prediction model at the k moment.

[0086] Step 6: When the IUPF algorithm observes new data at time k, the system uses an incremental learning method to learn and update only the information of the new data, update the particle weights, and combine the LSSVM online model to give further prediction values ​​of SOC, SOH, and LSSVM parameters.

[0087] Wherein, step 6 comprises the following steps:

[0088] Step 6.1: Initialize n initial state particles xk :

[0089] When k = 0, N particles are selected from the data sample x0 for initialization, and the particles satisfy the prior distribution function, that is,

[0090]

[0091]

[0092] Among them, the value range of i is 1,2,...,N. It represents the mean calculated based on the data sample. represents the covariance calculated from the data sample.

[0093] Step 6.2: Apply the improved UPF algorithm to each sampled particle for weight optimization;

[0094] When k = 0, the particle mean and covariance are calculated based on the UKF algorithm:

[0095]

[0096] When k≠0, importance sampling is performed and the improved UKF algorithm is applied to optimize the covariance matrix calculation problem.

[0097] 1) Calculate the sigma point:

[0098]

[0099] in, is the mean of the weights, is the covariance of the weights, and β is taken as 2 when solving the Gaussian problem. λ=α 2 (L+κ)-L, κ is the subsampling factor.

[0100] because There are invalid values ​​for the square root of the singular value, so the improved UKF algorithm will perform singular value decomposition (SVD) on the UKF algorithm:

[0101]

[0102] Among them, U is the left singular vector, S is called the singular value matrix, and V is the right singular vector.

[0103] After decomposition, the sigma point is finally calculated using the following formula:

[0104]

[0105] The value range of j is 1, 2, ..., 2L, where L is the dimension of the sigma point.

[0106] 2) The state value and predicted covariance of the particle are further predicted by the sigma sampling point to obtain the state prediction value:

[0107]

[0108] 3) Obtain the observed predicted value from the further state predicted value:

[0109]

[0110] 4) Calculate autocovariance and cross-covariance using weighted values:

[0111]

[0112] 5) Calculate the Kalman gain and update the state and covariance:

[0113]

[0114] 6) Generate important density functions and sample them to generate particles:

[0115]

[0116] Step 6.3: Calculate the importance weights of the particles and normalize them so that their sum is 1;

[0117] 1) Calculate the weight of each particle:

[0118]

[0119] 2) Normalization:

[0120]

[0121] Step 6.4: Using the resampling algorithm, set the threshold N max To determine the effective number of particles N eff , copy and eliminate particles, and recalculate weights;

[0122]

[0123] Step 6.5: Update the state estimation particle of the corresponding SOC or current capacity;

[0124]

[0125] Step 6.6: Perform online estimation of LSSVM parameter updates.

[0126] Furthermore, the state equation and observation equation of the IUPF algorithm framework in step 6.6 are as follows:

[0127]

[0128] Among them, s k =[γ k σ k ], γ k is the regularization coefficient, σ k is the kernel parameter, s k is the state vector in the IUPF state space equation, ω k , k are independent Gaussian white noises.

[0129] Step 7: The updated LSSVM model parameters are iterated to the training dataset for state estimation at time k+1, thereby achieving joint prediction of battery SOC and SOH and online update of the data-driven model.

[0130] Example 2

[0131] This embodiment provides a SOC-SOH joint estimation model, the steps are as follows:

[0132] The state equations of the SOC and the current maximum available capacity of the lithium-ion battery are constructed respectively using the ampere-hour integration method and the battery capacity decay method. Then, the IUPF-LSSVM framework is used to iteratively realize the online training data-driven model and further prediction of the battery state parameters. The state estimation of the SOC value and SOH value is realized using the optimal state estimation module.

[0133] Step 1: Initialize voltage and current states and LSSVM parameter vectors and weights;

[0134] Step 2: Use the ampere-hour integration method and the battery capacity decay method to construct the state equations of the SOC and the current maximum available capacity of the lithium-ion battery respectively, use LSSVM to construct the measurement equations of the SOC and the current maximum available capacity of the lithium-ion battery, and transmit the SOC and the current maximum available capacity to the IUPF algorithm;

[0135] Step 3: Calculate the SOC value and SOH value at time k+1 through the prediction and update module of the IUPF algorithm;

[0136] Step 4: Update the value of the current maximum available capacity and feed it back to the state equation of the spatial mathematical model of the SOC state, and update the LSSVM parameter vector at the next moment;

[0137] Step 5: Realize the joint prediction of the SOC value and SOH value of the lithium-ion battery and iterate repeatedly.

[0138] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0139] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0140] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0141] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process in the computer or other programmable device. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0142] The embodiments of the present invention are described above in conjunction with the accompanying drawings, but the present invention is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the enlightenment of the present invention, ordinary technicians in this field can also make many forms without departing from the scope of protection of the purpose of the present invention and the claims, which all fall within the protection of the present invention.

Claims

1. A lithium battery SOC-SOH joint estimation method based on mechanism and data fusion drive, characterized in that: The steps include: Step 1: Construct a multidimensional dataset of the time series of the lithium battery SOC value and the current maximum available capacity, reconstruct the training data from the initial moment to the k-1 moment using the Takens embedding method, and generate a d-dimensional vector in the phase space; Step 2: Phase space reconstruction: reconstruct the multidimensional data set into a phase space with the same topological meaning as the dynamical system; Step 3: Offline training of the reconstructed least square support vector machine (LSSVM) prediction model; Step 4: Establish the state equations for SOC and SOH estimation respectively according to the ampere-hour integration method and the battery capacity degradation empirical model; establish the observation equations for SOC and SOH estimation respectively through the LSSVM online prediction model; Step 5: Train the offline model of LSSVM based on the training data from the initial moment to the k-1 moment, and use the improved unscented particle filter algorithm (IUPF, Intermediate UPF) to estimate the LSSVM parameters of the regression prediction model at the k moment; Step 6: When the IUPF algorithm observes new data at time k, the system uses an incremental learning method to learn and update only the information of the new data, update the particle weights, and combine the LSSVM online model to give further prediction values ​​of SOC, SOH, and LSSVM parameters; Step 7: The updated LSSVM model parameters are iterated to the training dataset for state estimation at time k+1, thereby achieving joint prediction of battery SOC and SOH and online update of the data-driven model.

2. According to claim 1, a lithium battery SOC-SOH joint estimation method based on mechanism and data fusion drive is characterized in that: Step 4 also includes the following steps: Step 4.1: Establish the IUPF state equations for SOC and SOH estimation based on the ampere-hour integration method and the battery capacity degradation empirical model respectively; Step 4.2: The SOC and SOH obtained by offline LSSVM training are used as virtual observations, and the LSSVM regression model is used as the observation equation.

3. According to claim 2, a lithium battery SOC-SOH joint estimation method based on mechanism and data fusion drive is characterized in that: The state equation for SOC estimation based on the ampere-hour integration method is discretized as follows: SOC k is the SOC value of the lithium battery at time k; Δt k is the system sampling time; C k is the available capacity of the battery at time k; I k is the system input current at time k.

4. According to claim 2, a lithium battery SOC-SOH joint estimation method based on mechanism and data fusion drive is characterized in that: The discretization expression of the state equation of capacity based on the empirical model of capacity decay is as follows: C k+1 =the c C k +β1exp(-β2 / Δt k )+ξ k In the formula, C k is the available capacity of the battery at time k; U oc,k is the open circuit voltage; k is the state noise; β1 and β2 are coefficients to be determined.

5. The SOC-SOH joint estimation method according to claim 2, characterized in that: The definition of lithium-ion battery state of health SOH is as follows: In the formula, SOH k is the SOH value calculated at the kth moment, C new is the factory capacitance of the battery; C current The maximum available capacity of the battery.

6. The lithium battery SOC-SOH joint estimation method based on mechanism and data fusion drive according to claim 2 is characterized in that: The state equations of SOC and SOH constructed by LSSVM observation equation are discretized as follows: y k =f(x k-1 ,s k )+i k Among them, x i is the state vector, s k is the LSSVM parameter, l k is Gaussian white noise.

7. The lithium battery SOC-SOH joint estimation method based on mechanism and data fusion drive according to claim 1 is characterized in that: Step 6 also includes the following steps: Step 6.1: Initialize n initial state particles x k ; Step 6.2: Apply the improved UKF algorithm to each sampled particle for weight optimization; Step 6.3: Calculate the importance weight of the particles and normalize them; Step 6.4: Use the resampling algorithm to copy and eliminate particles and recalculate weights; Step 6.5: Estimating the state of the corresponding SOC or SOH; Step 6.6: Perform online estimation of LSSVM parameter updates.

8. A SOC-SOH joint estimation model system, characterized in that: include: The state equations of the SOC and current capacity of lithium-ion batteries are constructed by using the ampere-hour integration method and the battery capacity decay method. The IUPF-LSSVM framework is then used to iteratively implement the online training data-driven model and further prediction of battery state parameters. The optimal state estimation module is used to achieve state estimation of the SOC value and SOH value. The IUPF-LSSVM framework module is specifically used to implement online training data-driven model and further prediction of battery parameters through iteration; The optimal state estimation module outputs the estimated SOH value and SOC value.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the SOC-SOH joint estimation method based on multi-method fusion as described in any one of claims 1 to 8 is implemented.

10. A device (controller / computer device / electronic device), characterized in that: include: A memory for storing data; A processor is used to execute the algorithm instructions so that the device performs operations to implement a lithium battery SOC-SOH joint estimation method driven by a mechanism and data fusion as described in any one of claims 1-8.

Images