Method and apparatus for predicting the remaining life of lithium-ion batteries

Abstract

This invention proposes a method and apparatus for predicting the remaining lifespan of lithium-ion batteries. It uses a dual-exponential model to fit degradation data corresponding to the lithium-ion battery capacity to determine initial values ​​for the model parameters. Based on the number of charge-discharge cycles and the initial model parameters, an initial particle set is established using a particle filtering algorithm, and a random particle set for the t-th iteration is constructed through importance sampling. The fitness values ​​of each particle in the initial particle set and the random particle set are calculated until t equals the maximum number of iterations T. A target particle set is then constructed based on the particles with the smallest fitness values. The importance weights of each particle in the target particle set are then used for random resampling to eliminate and retain particles. The remaining lifespan of the lithium-ion battery is determined based on the optimal particle set after elimination and retention. Thus, by iteratively optimizing the importance-sampled particles, a target particle set composed of particles with the smallest fitness values ​​is obtained, reducing particle diversity and improving prediction accuracy.

Description

Technical Field

[0001] This invention relates to the field of lithium-ion battery technology, and in particular to a method and apparatus for predicting the remaining life of lithium-ion batteries. Background Technology

[0002] Lithium-ion batteries, with their high energy density, have become the primary choice for modern electric vehicles. However, during continuous charge-discharge cycles, the capacity of lithium-ion batteries undergoes non-linear decay, which can lead to severe performance degradation and potentially serious accidents. In related technologies, empirical models combined with particle filtering algorithms can effectively fit degradation data to predict the remaining lifespan of lithium-ion batteries. However, the randomness of particle sampling in particle filtering algorithms leads to particle degradation issues during subsequent resampling, limiting prediction accuracy. Therefore, a more reliable method for predicting the remaining lifespan of lithium-ion batteries is urgently needed. Summary of the Invention

[0003] The present invention aims to at least partially solve one of the technical problems in the related art.

[0004] Therefore, the first objective of this invention is to propose a method for predicting the remaining life of lithium-ion batteries. By iteratively optimizing the importance sampling particles, a target particle set consisting of particles with the smallest fitness value is obtained, thereby reducing particle diversity and improving prediction accuracy.

[0005] The second objective of this invention is to provide a device for predicting the remaining life of a lithium-ion battery.

[0006] The third objective of this invention is to provide an electronic device.

[0007] The fourth objective of this invention is to provide a non-transitory computer-readable storage medium storing computer instructions.

[0008] To achieve the above objectives, a first aspect of the present invention provides a method for predicting the remaining life of a lithium-ion battery, the method comprising:

[0009] The degradation data of lithium-ion battery capacity nonlinear decay was fitted by a double exponential model to determine the initial values ​​of the four model parameters of the double exponential model. The initial values ​​of the four model parameters include internal impedance a0 and c0, and degradation rate b0 and d0.

[0010] Based on the number of charge-discharge cycles k of the lithium-ion battery and a0, c0, b0, d0, an initial particle set for each charge-discharge cycle is established using a particle filtering algorithm, and a random particle set for the t-th iteration is constructed through importance sampling, wherein the maximum number of iterations for importance sampling is T.

[0011] Calculate the fitness value of each particle in the initial particle set and the random particle set at the t-th iteration, until the iteration number t equals T, and construct the target particle set based on the particle with the smallest fitness value;

[0012] Substitute the target particle set into the double exponential prediction model, calculate the predicted battery capacity value corresponding to each particle in the target particle set, and normalize the importance weight of each particle based on the predicted battery capacity value.

[0013] Random resampling is used to eliminate particles with lower importance weights and retain particles with higher importance weights to obtain an optimal particle set. The filter in the particle filtering algorithm is then updated based on the optimal particle set to predict the remaining lifespan of the lithium-ion battery.

[0014] To achieve the above objectives, a second aspect of the present invention provides a lithium-ion battery remaining life prediction device, the device comprising:

[0015] The fitting module is used to fit the degradation data of lithium-ion battery capacity nonlinear decay using a double exponential model to determine the initial values ​​of the four model parameters of the double exponential model, including internal impedance a0 and c0, and degradation rate b0 and d0.

[0016] The first construction module is used to establish an initial particle set for each charge-discharge cycle based on the number of charge-discharge cycles k of the lithium-ion battery and the values ​​a0, c0, b0, and d0, and to construct a random particle set for the t-th iteration through importance sampling, wherein the maximum number of iterations for importance sampling is T.

[0017] The second construction module is used to calculate the fitness value of each particle in the initial particle set and the random particle set at the t-th iteration, until the iteration number t equals T, and then construct the target particle set based on the particle with the smallest fitness value.

[0018] The calculation module is used to substitute the target particle set into the double exponential prediction model, calculate the predicted battery capacity value corresponding to each particle in the target particle set, and normalize the importance weight of each particle based on the predicted battery capacity value.

[0019] The prediction module is used to eliminate particles with lower importance weights and retain particles with higher importance weights by random resampling to obtain an optimal particle set, and to update the filter in the particle filtering algorithm based on the optimal particle set in order to predict the remaining life of the lithium-ion battery.

[0020] To achieve the above objectives, a third aspect of the present invention provides an electronic device comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to perform the method described in the first aspect.

[0021] To achieve the above objectives, a fourth aspect of the present invention provides a non-transitory computer-readable storage medium storing computer instructions for causing the computer to perform the method described in the first aspect.

[0022] The lithium-ion battery remaining life prediction method, apparatus, electronic device, and storage medium provided in this invention fit the degradation data corresponding to the lithium-ion battery capacity using a dual exponential model to determine the initial values ​​of the model parameters. Based on the number of charge-discharge cycles and the initial values ​​of the model parameters, an initial particle set is established using a particle filtering algorithm, and a random particle set for the t-th iteration is constructed through importance sampling. The fitness values ​​of each particle in the initial particle set and the random particle set are calculated until t equals the maximum number of iterations T. A target particle set is constructed based on the particle with the smallest fitness value. The importance weights of each particle in the target particle set are used to eliminate and retain particles through random resampling. The remaining life of the lithium-ion battery is determined based on the optimal particle set after elimination and retention. Thus, by iteratively optimizing the importance-sampled particles, a target particle set composed of particles with the smallest fitness value is obtained, reducing particle diversity and improving prediction accuracy.

[0023] Additional aspects and advantages of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0024] The above and / or additional aspects and advantages of the present invention will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, wherein:

[0025] Figure 1 This is a flowchart illustrating a method for predicting the remaining life of a lithium-ion battery according to an embodiment of the present invention.

[0026] Figure 2 A flowchart of an algorithm for a target particle set provided in an embodiment of the present invention;

[0027] Figure 3 This is a schematic diagram of a lithium-ion battery remaining life prediction device provided in an embodiment of the present invention. Detailed Implementation

[0028] Embodiments of the present invention are described in detail below, examples of which are illustrated in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain the present invention, and should not be construed as limiting the present invention.

[0029] It should be noted that the acquisition, storage, use, and processing of data in the technical solution of this invention all comply with the relevant provisions of relevant laws and regulations.

[0030] The following describes a method and apparatus for predicting the remaining life of a lithium-ion battery according to embodiments of the present invention with reference to the accompanying drawings.

[0031] Figure 1 This is a flowchart illustrating a method for predicting the remaining life of a lithium-ion battery according to an embodiment of the present invention.

[0032] like Figure 1 As shown, the method includes the following steps:

[0033] Step 101: Fit the degradation data of lithium-ion battery capacity when it undergoes nonlinear decay using a double exponential model to determine the initial values ​​of the four model parameters of the double exponential model. The initial values ​​of the four model parameters include internal impedance a0 and c0, and degradation rates b0 and d0.

[0034] Alternatively, lithium-ion batteries can be used in electric vehicles, but are not limited to this.

[0035] In some embodiments, the specific expression of the double exponential model can be:

[0036] Q(k) = a*exp(b*k) + c*exp(d*k)

[0037] Where a, b, c, and d are the model parameters of the double exponential model, where a and c are the internal impedance, b and d are the degradation rates, k is the number of lithium-ion battery cycles, and Q is the lithium-ion battery capacity. Therefore, when the double exponential model is used to fit the degradation data when the lithium-ion battery capacity undergoes nonlinear decay, a0, c0, b0, and d0 are the initial values ​​of the four model parameters, where a0 and c0 are the initial internal impedances, and b0 and d0 are the initial degradation rates.

[0038] Step 102: Based on the number of charge-discharge cycles k of the lithium-ion battery and a0, c0, b0, d0, an initial particle set for each charge-discharge cycle is established using a particle filtering algorithm, and a random particle set for the t-th iteration is constructed through importance sampling, where the maximum number of iterations for importance sampling is T.

[0039] Optionally, the lithium-ion battery capacity can be defined as Q, the initial battery capacity can be normalized to 1, and k can be set as the number of charge-discharge cycles of the lithium-ion battery.

[0040] In some embodiments, based on the number of charge-discharge cycles k of the lithium-ion battery and a0, c0, b0, d0, an initial particle set for each charge-discharge cycle is established using a particle filtering algorithm. Then, a random particle set for the t-th iteration is constructed through importance sampling. One implementation where the maximum number of iterations for importance sampling is T can be as follows: based on the number of charge-discharge cycles k of the battery capacity and a0, c0, b0, d0, a particle set for each charge-discharge cycle is established using a particle filtering algorithm. Process noise is added to a0, c0, b0, d0 to initialize the state of the particle set, thus obtaining the initial particle set. Based on the number of charge-discharge cycles k of the battery capacity and a0, c0, b0, d0, a random particle set for the t-th iteration is constructed through importance sampling. This achieves the initialization of the particle set and the rapid generation of the random particle set.

[0041] Furthermore, by setting the number of charge-discharge cycles to k = n, and the particle set to... When n=1, extract m initial state particles (initialized particle set). Where i = 1, 2...n.

[0042] Furthermore, by setting the maximum number of iterations for importance sampling to T, at the t-th iteration, a set of random particles is... Random particle set initialized to

[0043] Wherein, at the t-th iteration, the position of the random particle set is at the maximum value d of the particle set in the t-th iteration. t The minimum value c of the random particle set in the t-th iteration t Within the range, the maximum value d t It is determined by the random parameters in the t-th iteration and the maximum value of the particle set in the initial iteration, with a minimum value c. t It is determined by the random parameters in the t-th iteration and the minimum value of the initial iteration of the random particle set.

[0044] Specifically, the formula for the minimum value of the random particle set in the t-th iteration is as follows:

[0045] In the formula, w is a random parameter under different iteration numbers; c1 is the minimum value of the initial iteration of the random particle set.

[0046] The formula for the maximum value of the particle set in the t-th iteration is as follows:

[0047] In the formula, w is a random parameter under different iteration numbers; d1 is the maximum value of the particle set in the initial iteration.

[0048] Among them, random particle set The position, at the maximum value d t With minimum value c t Within the range, the expression for the position of the random particle set after the t-th iteration is as follows:

[0049]

[0050] In the formula, cumsum is the accumulation function, and the expression for r(t) is:

[0051] z is a random number between 0 and 1.

[0052] Step 103: Calculate the fitness values ​​of each particle in the initial particle set and the random particle set at the t-th iteration, until the iteration number t equals T, and construct the target particle set based on the particle with the smallest fitness value.

[0053] In some embodiments, the fitness values ​​of each particle in the initial particle set and the random particle set at the t-th iteration are calculated. This process continues until the iteration number t equals T. One implementation of constructing a target particle set based on the particle with the smallest fitness value can be as follows: A fitness function is constructed based on a Gaussian function, the predicted battery capacity after filtering by a particle filter algorithm, and the observed lithium-ion battery capacity. The fitness values ​​of each particle in the initial particle set and the random particle set at the t-th iteration are calculated based on this fitness function. The fitness values ​​of the initial particle set and the random particle set are compared, and the m particles with the smallest fitness values ​​are determined as the positions of the particle set. This process continues until the iteration number t equals T, yielding the target particle with the smallest fitness value and its particle set position. The target particle set is then constructed based on the target particle and its particle set position. Thus, by comparing the fitness functions of the particle set and the random particle set, the smaller fitness values ​​are used to replace the positions of the particle sets, and the positions of the particle sets are adjusted. After iteration to the maximum value, the optimal N solutions are obtained. All particles are optimized to obtain the solution set with the best fitness value.

[0054] Specifically, the predicted battery capacity value after filtering by the particle filter algorithm is Z. pf Given that the observed capacity of the lithium-ion battery is Z, the fitness function expression is:

[0055]

[0056] In the formula, x is a random number in the range [-4, 4], F is a Gaussian function, and F follows a standard normal distribution. The purpose of introducing the Gaussian function F is to prevent the fitness function from getting trapped in local convergence after t iterations. The particle sets are sorted from smallest to largest, and the set with the smallest fitness value is the target particle set.

[0057] Optionally, the position of the particle set is smaller than The specific location of the particle set can be: after the t-th iteration, take the average of the target particle set location and the location of the particle with the smaller fitness value.

[0058] In summary, this invention proposes an algorithm flowchart for a target particle set, as follows: Figure 2 As shown, at the start of the nth charge-discharge cycle, an initial particle set for each charge-discharge cycle is established using a particle filtering algorithm. A random particle set for the t-th iteration is constructed using importance sampling. When t equals 1, the fitness values ​​of each particle in the particle set and the random particle set are calculated and compared. The m particles with the smallest fitness values ​​are identified as the positions of the particle set, and the records are adjusted. Then, the next iteration begins. In the t-th iteration, random particles are reselected, and the fitness values ​​of the random particles in the t-th iteration are compared with those in the (t-1)-th iteration. The particles with the smallest fitness values ​​in the t-th iteration are identified as the positions of the particle set, and the records are adjusted. This process continues until t = t + 1, until t = T. After the T-th iteration, the target particle set is constructed based on the particle with the smallest fitness value.

[0059] Step 104: Substitute the target particle set into the double exponential prediction model, calculate the predicted battery capacity value corresponding to each particle in the target particle set, and normalize the importance weight of each particle based on the predicted battery capacity value.

[0060] In some embodiments, the importance weight of each particle is determined by comparing the similarity between the particle's observed data (predicted battery capacity) and the actual observed data (observed battery capacity Z).

[0061] Step 105: Particles with lower importance weights are eliminated by random resampling, while particles with higher importance weights are retained to obtain the optimal particle set. The filter in the particle filtering algorithm is then updated based on the optimal particle set to predict the remaining lifespan of the lithium-ion battery.

[0062] Optionally, the predicted remaining lifespan of a lithium-ion battery refers to the number of charge-discharge cycles required from the current moment to the end of the battery's lifespan.

[0063] In some embodiments, random resampling is used to eliminate particles with lower importance weights and retain particles with higher importance weights to obtain an optimal particle set. The filter in the particle filtering algorithm is then updated based on this optimal particle set to predict the remaining lifespan of the lithium-ion battery. Another implementation method involves using random resampling to eliminate particles with lower importance weights and retain particles with higher importance weights to obtain an optimal particle set, and then updating the filter X in the particle filtering algorithm based on this optimal particle set. pf The updated filter determines the values ​​of the four target model parameters in the double exponential model, and the predicted battery capacity Z is calculated based on these four target model parameter values. pf Based on the predicted battery capacity, the remaining lifespan of the lithium-ion battery is predicted.

[0064] Furthermore, we can let k = k + 1, until k = N, to complete the particle filter X. pf The predicted battery capacity Z after filtering pf Updates provide accurate predictions of the remaining lifespan of lithium-ion batteries in real time.

[0065] Furthermore, the cycle start period can be set to R = NF, where N represents the total number of cycle periods and F represents the number of predicted cycle periods, based on the updated particle filter X described above. pf Determine the parameter values ​​of the four target models in the double exponential model, based on the predicted battery capacity Z. pf To determine the remaining lifespan of lithium-ion batteries.

[0066] The lithium-ion battery remaining life prediction method of this invention fits the degradation data corresponding to the lithium-ion battery capacity using a double exponential model to determine the initial values ​​of the model parameters. Based on the number of charge-discharge cycles and the initial values ​​of the model parameters, an initial particle set is established using a particle filtering algorithm, and a random particle set for the t-th iteration is constructed through importance sampling. The fitness values ​​of each particle in the initial particle set and the random particle set are calculated until t equals the maximum number of iterations T. A target particle set is constructed based on the particle with the smallest fitness value. The importance weights of each particle in the target particle set are used to eliminate and retain particles through random resampling. The remaining life of the lithium-ion battery is determined based on the optimal particle set after elimination and retention. Thus, by iteratively optimizing the importance-sampled particles, a target particle set composed of particles with the smallest fitness value is obtained, reducing particle diversity and improving prediction accuracy.

[0067] To achieve the above embodiments, the present invention also proposes a lithium-ion battery remaining life prediction device.

[0068] Figure 3 This is a schematic diagram of a lithium-ion battery remaining life prediction device provided in an embodiment of the present invention.

[0069] like Figure 3 As shown, the lithium-ion battery remaining life prediction device 30 includes: a fitting module 31, a first construction module 32, a second construction module 33, a calculation module 34, and a prediction module 35.

[0070] The fitting module 31 is used to fit the degradation data of lithium-ion battery capacity when it undergoes nonlinear decay through a double exponential model, so as to determine the initial values ​​of the four model parameters of the double exponential model. The initial values ​​of the four model parameters include internal impedance a0 and c0, and degradation rate b0 and d0.

[0071] The first construction module 32 is used to establish an initial particle set for each charge-discharge cycle based on the number of charge-discharge cycles k of the lithium-ion battery and the a0, c0, b0, and d0, and to construct a set of random particles for the t-th iteration through importance sampling, wherein the maximum number of iterations for importance sampling is T.

[0072] The second construction module 33 is used to calculate the fitness value of each particle in the initial particle set and the random particle set at the t-th iteration, until the iteration number t equals T, and then construct the target particle set based on the particle with the smallest fitness value.

[0073] The calculation module 34 is used to substitute the target particle set into the double exponential prediction model, calculate the predicted battery capacity value corresponding to each particle in the target particle set, and normalize the importance weight of each particle based on the predicted battery capacity value.

[0074] The prediction module 35 is used to eliminate particles with lower importance weights and retain particles with higher importance weights by random resampling to obtain an optimal particle set, and to update the filter in the particle filtering algorithm based on the optimal particle set in order to predict the remaining life of the lithium-ion battery according to the updated filter.

[0075] Furthermore, in one possible implementation of this invention, the first construction module 32 is specifically used for:

[0076] Based on the number of charge-discharge cycles k of the battery capacity and the values ​​a0, c0, b0, and d0, a particle set for each charge-discharge cycle is established using a particle filtering algorithm. Process noise is added to a0, c0, b0, and d0 to initialize the state of the particle set, thus obtaining an initialized particle set.

[0077] Based on the battery capacity, the number of charge-discharge cycles k, and a0, c0, b0, d0, a set of random particles for the t-th iteration is constructed through importance sampling.

[0078] Furthermore, in one possible implementation of this invention, the position of the random particle set at the t-th iteration is at the maximum value d of the particle set at the t-th iteration. t The minimum value c of the random particle set in the t-th iteration t Within the range, the maximum value d t It is determined by the random parameters in the t-th iteration and the maximum value of the particle set in the initial iteration, with a minimum value c. t It is determined by the random parameters in the t-th iteration and the minimum value of the initial iteration of the random particle set.

[0079] Furthermore, in one possible implementation of this invention, the second building module 33 is specifically used for:

[0080] A fitness function is constructed based on the predicted battery capacity after filtering by the Gaussian function and the particle filter algorithm, as well as the observed lithium-ion battery capacity.

[0081] The fitness values ​​of each particle in the initial particle set and the random particle set at the t-th iteration are calculated based on the fitness function. The fitness values ​​of each particle in the initial particle set and the random particle set are compared. The m particles with smaller fitness values ​​are determined as the positions of the particle set. This process continues until the iteration number t equals T, at which point the target particle with the smallest fitness value and the position of the target particle in the particle set are obtained.

[0082] Based on the target particle and its particle set position, construct the target particle set.

[0083] Furthermore, in one possible implementation of this invention, the position of the particle set is less than...

[0084] Furthermore, in one possible implementation of this invention, the prediction module 35 is specifically used for:

[0085] Random resampling is used to eliminate particles with lower importance weights and retain particles with higher importance weights to obtain the optimal particle set, and the filter in the particle filtering algorithm is updated based on the optimal particle set.

[0086] The updated filter determines the values ​​of four target model parameters in the double exponential model, and the predicted battery capacity is calculated based on the four target model parameter values.

[0087] Based on the predicted battery capacity, the remaining lifespan of the lithium-ion battery is predicted.

[0088] The lithium-ion battery remaining life prediction device of this invention fits the degradation data corresponding to the lithium-ion battery capacity using a dual exponential model to determine the initial values ​​of the model parameters. Based on the number of charge-discharge cycles and the initial values ​​of the model parameters, an initial particle set is established using a particle filtering algorithm, and a random particle set for the t-th iteration is constructed through importance sampling. The fitness values ​​of each particle in the initial particle set and the random particle set are calculated until t equals the maximum number of iterations T. A target particle set is constructed based on the particle with the smallest fitness value. The importance weights of each particle in the target particle set are used to eliminate and retain particles through random resampling. The remaining life of the lithium-ion battery is determined based on the optimal particle set after elimination and retention. Thus, by iteratively optimizing the importance-sampled particles, a target particle set composed of particles with the smallest fitness value is obtained, reducing particle diversity and improving prediction accuracy.

[0089] To achieve the above embodiments, the present invention also proposes an electronic device, comprising:

[0090] At least one processor; and

[0091] A memory communicatively connected to the at least one processor; wherein,

[0092] The memory stores instructions that can be executed by the at least one processor to enable the at least one processor to perform the aforementioned method.

[0093] To implement the above embodiments, the present invention also proposes a non-transitory computer-readable storage medium storing computer instructions for causing the computer to perform the aforementioned method.

[0094] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0095] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this invention, "a plurality of" means at least two, such as two, three, etc., unless otherwise explicitly specified.

[0096] Any process or method description in the flowchart or otherwise herein can be understood as representing a module, segment, or portion of code comprising one or more executable instructions for implementing custom logic functions or processes, and the scope of preferred embodiments of the invention includes additional implementations in which functions may be performed not in the order shown or discussed, including substantially simultaneously or in reverse order depending on the functions involved, as should be understood by those skilled in the art to which embodiments of the invention pertain.

[0097] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-included system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of computer-readable media include: an electrical connection having one or more wires (electronic device), a portable computer disk drive (magnetic device), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Furthermore, the computer-readable medium may even be paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium and then editing, interpreting or processing it in another suitable manner if necessary, and then storing it in a computer memory.

[0098] It should be understood that various parts of the present invention can be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0099] Those skilled in the art will understand that all or part of the steps of the methods in the above embodiments can be implemented by a program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, the program includes one or a combination of the steps of the method embodiments.

[0100] Furthermore, the functional units in the various embodiments of the present invention can be integrated into a processing module, or each unit can exist physically separately, or two or more units can be integrated into a module. The integrated module can be implemented in hardware or as a software functional module. If the integrated module is implemented as a software functional module and sold or used as an independent product, it can also be stored in a computer-readable storage medium.

[0101] The storage medium mentioned above can be a read-only memory, a disk, or an optical disk, etc. Although embodiments of the present invention have been shown and described above, it is to be understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, substitutions, and variations to the above embodiments within the scope of the present invention.

Claims

1. A method for predicting the remaining life of a lithium-ion battery, characterized in that, The method includes: The degradation data of lithium-ion battery capacity nonlinear decay was fitted by a double exponential model to determine the initial values ​​of the four model parameters of the double exponential model. The initial values ​​of the four model parameters include internal impedance a0 and c0, and degradation rate b0 and d0. Based on the number of charge-discharge cycles k of the lithium-ion battery and a0, c0, b0, d0, an initial particle set for each charge-discharge cycle is established using a particle filtering algorithm, and a random particle set for the t-th iteration is constructed through importance sampling, wherein the maximum number of iterations for importance sampling is T. Calculate the fitness value of each particle in the initial particle set and the random particle set at the t-th iteration, until the iteration number t equals T, and construct the target particle set based on the particle with the smallest fitness value; Substitute the target particle set into the double exponential prediction model, calculate the predicted battery capacity value corresponding to each particle in the target particle set, and normalize the importance weight of each particle based on the predicted battery capacity value. Random resampling is used to eliminate particles with low importance weights and retain particles with high importance weights to obtain an optimal particle set. The filter in the particle filtering algorithm is then updated based on the optimal particle set to predict the remaining lifespan of the lithium-ion battery.

2. The method for predicting the remaining life of a lithium-ion battery according to claim 1, characterized in that, The number of charge-discharge cycles k based on the lithium-ion battery, and a0, c0, b0, and d0, are used to establish an initial particle set for each charge-discharge cycle through a particle filtering algorithm. Furthermore, a random particle set for the t-th iteration is constructed through importance sampling, including: Based on the number of charge-discharge cycles k of the battery capacity and the values ​​a0, c0, b0, and d0, a particle set for each charge-discharge cycle is established using a particle filtering algorithm. Process noise is added to a0, c0, b0, and d0 to initialize the state of the particle set, thus obtaining an initialized particle set. Based on the battery capacity, the number of charge-discharge cycles k, and a0, c0, b0, d0, a set of random particles for the t-th iteration is constructed through importance sampling.

3. The method for predicting the remaining life of a lithium-ion battery according to claim 2, characterized in that, in, At the t-th iteration, the position of the random particle set is at the maximum value d of the particle set in the t-th iteration. t The minimum value c of the random particle set in the t-th iteration t Within the range, the maximum value d t It is determined by the random parameters in the t-th iteration and the maximum value of the particle set in the initial iteration, with a minimum value c. t It is determined by the random parameters in the t-th iteration and the minimum value of the initial iteration of the random particle set.

4. The method for predicting the remaining life of a lithium-ion battery according to claim 1, characterized in that, The process of calculating the fitness values ​​of each particle in the initial particle set and the random particle set at the t-th iteration, until the iteration number t equals T, and constructing the target particle set based on the particle with the smallest fitness value, includes: A fitness function is constructed based on the predicted battery capacity after filtering by the Gaussian function and the particle filter algorithm, as well as the observed lithium-ion battery capacity. The fitness values ​​of each particle in the initial particle set and the random particle set at the t-th iteration are calculated based on the fitness function. The fitness values ​​of each particle in the initial particle set and the random particle set are compared. The m particles with smaller fitness values ​​are determined as the positions of the particle set. This process continues until the iteration number t equals T, at which point the target particle with the smallest fitness value and the position of the target particle in the particle set are obtained. Based on the target particle and its particle set position, construct the target particle set.

5. The method for predicting the remaining life of a lithium-ion battery according to claim 4, characterized in that, The position of the particle set is less than 6. The method for predicting the remaining life of a lithium-ion battery according to claim 1, characterized in that, The process involves randomly resampling to eliminate particles with low importance weights and retaining particles with high importance weights to obtain an optimal particle set. The filter in the particle filtering algorithm is then updated based on this optimal particle set to predict the remaining lifespan of the lithium-ion battery. This includes: Random resampling is used to eliminate particles with low importance weights and retain particles with high importance weights to obtain the optimal particle set, and the filter in the particle filtering algorithm is updated based on the optimal particle set. The updated filter determines the values ​​of four target model parameters in the double exponential model, and the predicted battery capacity is calculated based on the four target model parameter values. Based on the predicted battery capacity, the remaining lifespan of the lithium-ion battery is predicted.

7. A device for predicting the remaining life of a lithium-ion battery, characterized in that, The device comprises: The fitting module is used to fit the degradation data of lithium-ion battery capacity nonlinear decay using a double exponential model to determine the initial values ​​of the four model parameters of the double exponential model, including internal impedance a0 and c0, and degradation rate b0 and d0. The first construction module is used to establish an initial particle set for each charge-discharge cycle based on the number of charge-discharge cycles k of the lithium-ion battery and the values ​​a0, c0, b0, and d0, and to construct a random particle set for the t-th iteration through importance sampling, wherein the maximum number of iterations for importance sampling is T. The second construction module is used to calculate the fitness value of each particle in the initial particle set and the random particle set at the t-th iteration, until the iteration number t equals T, and then construct the target particle set based on the particle with the smallest fitness value. The calculation module is used to substitute the target particle set into the double exponential prediction model, calculate the predicted battery capacity value corresponding to each particle in the target particle set, and normalize the importance weight of each particle based on the predicted battery capacity value. The prediction module is used to eliminate particles with low importance weights and retain particles with high importance weights by random resampling to obtain an optimal particle set, and to update the filter in the particle filtering algorithm based on the optimal particle set, so as to predict the remaining life of the lithium-ion battery according to the updated filter.

8. The lithium-ion battery remaining life prediction device according to claim 7, characterized in that, The first building module is specifically used for: Based on the number of charge-discharge cycles k of the battery capacity and the values ​​a0, c0, b0, and d0, a particle set for each charge-discharge cycle is established using a particle filtering algorithm. Process noise is added to a0, c0, b0, and d0 to initialize the state of the particle set, thus obtaining an initialized particle set. Based on the battery capacity, the number of charge-discharge cycles k, and a0, c0, b0, d0, a set of random particles for the t-th iteration is constructed through importance sampling.

9. The lithium-ion battery remaining life prediction device according to claim 8, characterized in that, in, At the t-th iteration, the position of the random particle set is at the maximum value d of the particle set in the t-th iteration. t The minimum value c of the random particle set in the t-th iteration t Within the range, the maximum value d t It is determined by the random parameters in the t-th iteration and the maximum value of the particle set in the initial iteration, with a minimum value c. t It is determined by the random parameters in the t-th iteration and the minimum value of the initial iteration of the random particle set.

10. The lithium-ion battery remaining life prediction device according to claim 7, characterized in that, The second building module is specifically used for: A fitness function is constructed based on the predicted battery capacity after filtering by the Gaussian function and the particle filter algorithm, as well as the observed lithium-ion battery capacity. The fitness values ​​of each particle in the initial particle set and the random particle set at the t-th iteration are calculated based on the fitness function. The fitness values ​​of each particle in the initial particle set and the random particle set are compared. The m particles with smaller fitness values ​​are determined as the positions of the particle set. This process continues until the iteration number t equals T, at which point the target particle with the smallest fitness value and the position of the target particle in the particle set are obtained. Based on the target particle and its particle set position, construct the target particle set.

11. The lithium-ion battery remaining life prediction device according to claim 10, characterized in that, The position of the particle set is less than 12. The lithium-ion battery remaining life prediction device according to claim 7, characterized in that, The prediction module is specifically used for: Random resampling is used to eliminate particles with low importance weights and retain particles with high importance weights to obtain the optimal particle set, and the filter in the particle filtering algorithm is updated based on the optimal particle set. The updated filter determines the values ​​of four target model parameters in the double exponential model, and the predicted battery capacity is calculated based on the four target model parameter values. Based on the predicted battery capacity, the remaining lifespan of the lithium-ion battery is predicted.

13. An electronic device, characterized in that, include: At least one processor; as well as A memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor to enable the at least one processor to perform the method of any one of claims 1-6.

14. A non-transitory computer-readable storage medium storing computer instructions, characterized in that, The computer instructions are used to cause the computer to perform the method according to any one of claims 1-6.

Images