Battery module optimization design method, battery module and storage medium

Abstract

The invention relates to the technical field of batteries, and provides a battery module optimization design method, a battery module and a storage medium, and the method comprises the steps: obtaining design parameters of the battery module; obtaining a heat dissipation system optimization target and a collision system optimization target of the battery module according to the design parameters; constructing a collision system optimization model according to a hybrid multi-fidelity approximation model construction method; constructing a heat dissipation system optimization model according to a response surface method; the collision system optimization model and the heat dissipation system optimization model are evaluated according to a multi-objective optimization algorithm to obtain optimization design parameters of the battery module, the performance of the collision system and the performance of the heat dissipation system are evaluated through the multi-objective optimization algorithm, it can be ensured that the performance of the battery module in the collision and high-temperature environment meets the safety standard, and the safety of the battery module is improved. The failure rate of the battery in practical application is reduced, and the long-term reliability of the battery module is improved.

Description

Technical Field

[0001] The present application relates to the technical field related to battery module design, and in particular to a battery module optimization design method, a battery module and a storage medium. Background Art

[0002] With the rapid development of new energy vehicles, the performance of power batteries, as the core components of electric vehicles, directly affects the safety, endurance and service life of the entire vehicle. In the current electric vehicle power battery system, lithium-ion batteries have become the most commonly used battery type due to their high energy density, long cycle life and low self-discharge rate. However, despite the many advantages of lithium-ion batteries, they still face many challenges during operation, especially in terms of heat dissipation performance and collision safety.

[0003] At present, the design of battery modules focuses on the optimization of a single goal, and the heat dissipation system and the collision safety system are usually designed as two independent modules. However, the two factors of heat dissipation and collision safety affect each other. Most of the related methods use a simple linear superposition method, or only optimize one aspect separately, lacking a systematic, multi-objective optimization framework to improve the overall safety and reliability of electric vehicle battery systems. Summary of the invention

[0004] In view of this, it is necessary to provide a battery module optimization design method, a battery module and a storage medium that can overcome at least one of the above defects.

[0005] In a first aspect, an embodiment of the present application provides a battery module optimization design method for optimizing a battery module, the battery module optimization design method comprising:

[0006] Obtaining design parameters of the battery module;

[0007] Obtaining a heat dissipation system optimization target and a collision system optimization target of the battery module according to the design parameters;

[0008] The collision system optimization model is constructed according to the hybrid multi-fidelity approximate model construction method. The expression of the collision system optimization model is:

[0009]

[0010] Among them, x is the input sample, is the high-fidelity approximate response when the input is x, is a low-fidelity response, g(y LF ) is the mapping from low-fidelity output to high-fidelity output, w1, w2, γ are unknown coefficients;

[0011] Construct a cooling system optimization model based on the response surface methodology;

[0012] The collision system optimization model and the heat dissipation system optimization model are evaluated according to a multi-objective optimization algorithm to obtain the optimized design parameters of the battery module.

[0013] In an embodiment of the present application, evaluating the collision system optimization model and the heat dissipation system optimization model according to a multi-objective optimization algorithm to obtain the optimal design parameters of the battery module includes:

[0014] Solving the collision system optimization model and the heat dissipation system optimization model respectively according to a particle swarm optimization algorithm to obtain an optimal solution of the collision system optimization model and an optimal solution of the heat dissipation system optimization model;

[0015] Using the optimal solution of the collision system optimization model and the optimal solution of the heat dissipation system optimization model as the initial population of the non-dominated sorting genetic algorithm, and using the initial population as the parent population;

[0016] Generate a child population according to the parent population;

[0017] Generate a new population according to the initial population and the offspring population;

[0018] Performing non-dominated sorting on the individuals in the new population, and performing elite retention operation on the sorted individuals to obtain better individuals in the new population;

[0019] The better individuals are used as the new parent population after iterative updating.

[0020] In an embodiment of the present application, the battery module optimization design method further includes:

[0021] Determine whether the new population meets the stopping condition, wherein the stopping condition includes: reaching a preset maximum number of generations, a convergence condition, or a calculation accuracy requirement;

[0022] If the stopping condition is met, the current population is output as the Pareto frontier;

[0023] If the stopping condition is not met, the process returns to generating a child population according to the parent population, and continues to iterate until the stopping condition is met.

[0024] In an embodiment of the present application, the battery module optimization design method further includes:

[0025] Each solution in the Pareto front is standardized according to the standard boundary crossing method to obtain a standardized solution. The calculation formula of the standardized solution is:

[0026]

[0027] Among them, u a is the lower bound of the standardized solution set, u b is the upper bound of the standardized solution set;

[0028] Taking the origin as a reference, the search radius of each standardized solution is calculated. The calculation formula of the search radius is:

[0029]

[0030] Among them, j is the dimension of the target, q is a constant, R q is the search radius;

[0031] The solution with the smallest search radius among the standardized solutions is selected as the optimal solution.

[0032] In an embodiment of the present application, the method for constructing the hybrid multi-fidelity approximation model further includes:

[0033] The optimal solution is subjected to reverse normalization processing, and the normalization result is restored to the design parameters, and the restored parameters are the actual optimized design parameters of the battery module.

[0034] In an embodiment of the present application, the method for constructing the hybrid multi-fidelity approximation model includes:

[0035] Generate a high-fidelity input sample set based on the optimal Latin hypercube sampling algorithm;

[0036] Generate a low-fidelity input sample set according to the inherited Latin hypercube sampling algorithm;

[0037] constructing a low-fidelity model based on the low-fidelity input sample set and the polynomial response surface method;

[0038] A mapping relationship between low-fidelity responses and high-fidelity responses is constructed based on support vector regression technology;

[0039] The undetermined coefficients are optimized to improve the model accuracy of the hybrid multi-fidelity approximation model.

[0040] In an embodiment of the present application, the optimizing the undetermined coefficients includes:

[0041] Applying a genetic algorithm to optimize the undetermined coefficients to form candidate coefficients;

[0042] Performing crossover and mutation operations on the candidate coefficients to generate new candidate coefficients;

[0043] Evaluating the performance of the candidate coefficients according to a fitness function;

[0044] A parent population is generated according to the performance of the candidate coefficients to iterate the candidate coefficients until the candidate coefficients converge to an optimal value, thereby obtaining the hybrid multi-fidelity approximation model.

[0045] In this embodiment of the present application, the fitness function includes:

[0046] Calculate the candidate coefficients according to the model error to obtain the objective function value, wherein the model error includes a mean square error and an absolute error;

[0047] The objective function values ​​are sorted according to the size of the errors, and the individual with the smallest error in the objective function value is selected as the new candidate coefficient.

[0048] In a second aspect, an embodiment of the present application provides a battery module, including:

[0049] Battery cells;

[0050] A module outer frame, used to wrap the battery core;

[0051] Wherein, the battery module is optimized and designed using the battery module optimization design method as described in the first aspect.

[0052] In a third aspect, an embodiment of the present application provides a computer-readable storage medium, characterized in that it includes instructions, wherein the instructions instruct a device to execute the battery module optimization design method as described in the first aspect.

[0053] The battery module optimization design method, battery module and storage medium provided in the embodiment of the present application can effectively obtain the optimal design parameters of the battery module by comprehensively considering the optimization objectives of the heat dissipation system and the collision protection system, combining the hybrid multi-fidelity approximation model and the response surface method. By evaluating the performance of the collision system and the heat dissipation system through a multi-objective optimization algorithm, it can ensure that the performance of the battery module in a collision and high temperature environment meets safety standards, reduce the failure rate of the battery in actual applications, and improve the long-term reliability of the battery module. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Figure 1 A schematic diagram of a flow chart of a battery module optimization design method provided in one embodiment of the present application.

[0055] Figure 2 A schematic flow chart of a battery module optimization design method provided in another embodiment of the present application.

[0056] Figure 3 A schematic diagram of obtaining a search radius provided in an embodiment of the present application.

[0057] Figure 4 A schematic diagram of an electronic device provided by an embodiment of the present application.

[0058] Main component symbols

[0059] Electronic devices 20

[0060] Processor 21

[0061] Memory 22

[0062] Steps S100-S500; S510-S580 DETAILED DESCRIPTION

[0063] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention. In addition, the technical features involved in the various embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0064] It should be noted that, in the embodiments of the present application, "at least one" refers to one or more, and "more" refers to two or more. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art in the present application. The terms used in the specification of the present application are only for the purpose of describing specific embodiments and are not intended to limit the present application.

[0065] It should be noted that, in the embodiments of the present application, words such as "first" and "second" are only used for the purpose of distinguishing descriptions, and cannot be understood as indicating or implying relative importance, nor can they be understood as indicating or implying order. Features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of the embodiments of the present application, words such as "exemplary" or "for example" are used to indicate examples, illustrations or explanations. Any embodiment or design described as "exemplary" or "for example" in the embodiments of the present application should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of words such as "exemplary" or "for example" is intended to present related concepts in a concrete way.

[0066] Based on the implementations in this application, all other implementations obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0067] With the rapid development of new energy vehicles, the performance of power batteries, as the core components of electric vehicles, directly affects the safety, endurance and service life of the entire vehicle. In the current electric vehicle power battery system, lithium-ion batteries have become the most commonly used battery type due to their high energy density, long cycle life and low self-discharge rate. However, despite the many advantages of lithium-ion batteries, they still face many challenges during operation, especially in terms of heat dissipation performance and collision safety.

[0068] At present, the design of battery modules focuses on the optimization of a single goal, and the heat dissipation system and the collision safety system are usually designed as two independent modules. However, the two factors of heat dissipation and collision safety affect each other. Most of the related methods use a simple linear superposition method, or only optimize one aspect separately, lacking a systematic, multi-objective optimization framework to improve the overall safety and reliability of electric vehicle battery systems.

[0069] In view of this, the embodiments of the present application provide a battery module optimization design method, a battery module and a storage medium, which can effectively obtain the optimal design parameters of the battery module by comprehensively considering the optimization objectives of the heat dissipation system and the collision protection system, combining the hybrid multi-fidelity approximation model and the response surface method. By evaluating the performance of the collision system and the heat dissipation system through a multi-objective optimization algorithm, it can ensure that the performance of the battery module in a collision and high temperature environment meets safety standards, reduce the failure rate of the battery in actual applications, and improve the long-term reliability of the battery module.

[0070] Figure 1 1 is a flow chart of a battery module optimization design method provided by an embodiment of the present application. Figure 1 The battery module optimization design method shown includes at least the following steps: S100: obtaining the design parameters of the battery module; S200: obtaining the heat dissipation system optimization objectives and the collision system optimization objectives of the battery module according to the design parameters; S300: constructing a collision system optimization model according to a hybrid multi-fidelity approximate model construction method; S400: constructing a heat dissipation system optimization model according to a response surface method; S500: evaluating the collision system optimization model and the heat dissipation system optimization model according to a multi-objective optimization algorithm to obtain the optimized design parameters of the battery module.

[0071] S100: Obtain design parameters of the battery module.

[0072] In the embodiment of the present application, the battery module optimization design method includes obtaining design parameters of the battery module in step S100.

[0073] Specifically, the design parameters of the battery module may include but are not limited to: the size, structure, material, energy density, operating voltage, heat dissipation performance, collision protection performance, etc. of the battery cell. In addition, the working environment of the battery module may also need to be considered, such as temperature range, operating pressure, vibration and other external conditions.

[0074] It is understandable that design parameters are the basis for battery module optimization. Only by accurately obtaining these parameters can subsequent optimization analysis be carried out.

[0075] S200: Obtaining a heat dissipation system optimization target and a collision system optimization target of the battery module according to the design parameters.

[0076] In the embodiment of the present application, the battery module optimization design method includes, in step S200, obtaining the heat dissipation system optimization target and the collision system optimization target of the battery module according to the design parameters.

[0077] Specifically, the optimization goals of the heat dissipation system usually involve indicators such as temperature distribution uniformity, thermal resistance and heat dissipation efficiency to ensure that the battery does not overheat under high load. The optimization goals of the collision system may include collision resistance, structural safety and collision energy absorption capacity, aiming to minimize the damage to the battery system caused by the collision.

[0078] It is understandable that the optimization objectives of the heat dissipation system mainly include the maximum temperature Tmax and the voltage drop ΔP of the battery. The maximum temperature Tmax of the battery is a key indicator reflecting the performance of the heat dissipation system. Too high Tmax will affect the capacity and cycle life of the battery and increase the risk of thermal runaway. Therefore, the optimization of heat dissipation performance should be aimed at reducing Tmax as much as possible. At the same time, the voltage drop ΔP is also an important performance indicator. A larger voltage drop will increase the energy consumption of the heat dissipation system. Therefore, the voltage drop ΔP should be minimized.

[0079] It is understandable that the optimization goal of the collision system mainly focuses on the safety performance of the battery module after a collision, including the maximum battery intrusion Dmax and the peak collision force PCF. The maximum battery intrusion Dmax is directly related to the safety status of the battery. Excessive intrusion may cause dangers such as internal short circuit of the battery and diaphragm rupture. Therefore, Dmax should be as small as possible. The peak collision force PCF reflects the maximum impact load that the battery module bears during a collision. Excessive PCF may cause damage to the battery module structure. Therefore, the optimization goal is to minimize Dmax and PCF.

[0080] In the embodiment of the present application, the battery module optimization design method provided in the present application is applied to a certain type of battery module as an example. The inside of the battery module is filled with open-cell foam aluminum. Therefore, the three parameters of aluminum side plate thickness a, foam aluminum thickness b and foam aluminum gradient index n are finally selected as design variables, and the intrusion of the battery module during collision is used as a constraint, and the optimization work is carried out with the goal of minimizing the intrusion amount Dmax, peak collision force PCF, battery maximum temperature Tmax and pressure drop ΔP.

[0081] It is understandable that by clarifying the optimization goals of the heat dissipation system and the collision system, it is possible to provide quantitative standards for the subsequent optimization process and ensure the safety and efficiency of the battery module in actual applications.

[0082] S300: Construct a collision system optimization model based on a hybrid multi-fidelity approximate model construction method.

[0083] In the embodiment of the present application, the battery module optimization design method includes, in step S300, constructing a collision system optimization model according to a hybrid multi-fidelity approximate model construction method.

[0084] Specifically, the expression of the collision system optimization model is:

[0085]

[0086] Among them, x is the input sample, is the high-fidelity approximate response when the input is x, is a low-fidelity response, g(y LF ) is the mapping from low-fidelity output to high-fidelity output, and w1, w2, and γ are unknown coefficients.

[0087] It is understandable that the embodiment of the present application constructs an optimization model of the collision system by mixing low-fidelity and high-fidelity models, thereby reducing the computational complexity and speeding up the computational speed while ensuring high accuracy. This optimization method can better balance computational efficiency and optimization accuracy.

[0088] S400: Construct a cooling system optimization model based on the response surface methodology.

[0089] In the embodiment of the present application, the battery module optimization design method includes, in step S400, constructing a heat dissipation system optimization model according to the response surface method.

[0090] Specifically, response surface methodology is a modeling tool used in optimization processes. It approximates the behavior of the system through design experiments and regression analysis to establish a mathematical model between input parameters and output responses. In the optimization of cooling systems, response surface methodology is used to establish the relationship between cooling performance and design parameters, helping to find the best design parameters to maximize thermal efficiency and minimize temperature fluctuations.

[0091] It can be understood that the response surface method can quickly search for the best solution in multiple design parameter spaces through efficient mathematical modeling, reducing the time cost of experiments and calculations.

[0092] S500: Evaluate the collision system optimization model and the heat dissipation system optimization model according to the multi-objective optimization algorithm to obtain the optimized design parameters of the battery module.

[0093] In an embodiment of the present application, the battery module optimization design method includes, in step S500, evaluating a collision system optimization model and a heat dissipation system optimization model according to a multi-objective optimization algorithm to obtain optimization design parameters of the battery module.

[0094] Specifically, multi-objective optimization algorithms, such as particle swarm optimization algorithm PSO, genetic algorithm, etc., simultaneously consider multiple optimization objectives, namely the heat dissipation system and the collision system, balance between multiple objectives, and obtain multiple Pareto optimal solutions, thereby ensuring that the battery module can achieve the optimal design under different performance requirements.

[0095] It is understandable that the multi-objective optimization algorithm can improve the performance of both systems at the same time by evaluating the optimization models of the collision system and the heat dissipation system, ensuring that the battery module has both excellent heat dissipation performance and good anti-collision ability in actual applications.

[0096] It is understandable that the battery module optimization design method provided in the embodiment of the present application can comprehensively optimize the heat dissipation and collision protection system of the battery module, and combines the multi-fidelity approximation model and the multi-objective optimization algorithm to significantly reduce the amount of calculation and design time while improving the design accuracy. In addition, the response surface method and multi-objective optimization algorithm introduced in the method can ensure the reliability, safety and performance of the battery module in various use environments.

[0097] Figure 2 FIG. 1 is a flow chart of a battery module optimization design method provided by another embodiment of the present application. Figure 2The battery module optimization design method shown includes at least the following steps: S510: solving the collision system optimization model and the heat dissipation system optimization model respectively according to the particle swarm optimization algorithm to obtain the optimal solution of the collision system optimization model and the optimal solution of the heat dissipation system optimization model; S520: using the optimal solution of the collision system optimization model and the optimal solution of the heat dissipation system optimization model as the initial population of the non-dominated sorting genetic algorithm, and using the initial population as the parent population; S530: generating a child population according to the parent population; S540: generating a new population according to the initial population and the child population; S550: performing non-dominated sorting on the individuals in the new population, and performing elite retention operation on the sorted individuals to obtain better individuals in the new population; S560: using the better individuals as the new parent population after iterative update; S570: judging whether the new population meets the stopping condition; S580: outputting the current population as the Pareto frontier.

[0098] S510: Solving the collision system optimization model and the heat dissipation system optimization model respectively according to the particle swarm optimization algorithm to obtain an optimal solution of the collision system optimization model and an optimal solution of the heat dissipation system optimization model.

[0099] In the embodiment of the present application, steps S510 to S570 are Figure 1 Sub-steps of step S500 in the battery module optimization design method shown.

[0100] In an embodiment of the present application, the battery module optimization design method includes, in step S510, solving the collision system optimization model and the heat dissipation system optimization model respectively according to the particle swarm optimization algorithm to obtain the optimal solution of the collision system optimization model and the optimal solution of the heat dissipation system optimization model.

[0101] Specifically, the particle swarm optimization algorithm (PSO) simulates the position and velocity adjustment process of particles in the swarm, iteratively updates the position of each particle based on the objective function value, and finally obtains a set of approximate optimal solutions. In the collision system optimization, the goal is to minimize the maximum intrusion Dmax and peak collision force PCF of the battery module. In the heat dissipation system optimization, the goal is to minimize the maximum battery temperature Tmax and the voltage drop ΔP. The particle swarm optimization algorithm optimizes these two goals through global and local search mechanisms to obtain the optimal design parameters.

[0102] It is understandable that the particle swarm optimization algorithm simulates the behavior of particle groups, conducts global and local searches, and continuously approaches the optimal solution, which enables it to efficiently find ideal design parameters in complex multi-objective optimization problems, and is particularly suitable for collision and heat dissipation optimization problems of battery modules.

[0103] S520: taking the optimal solution of the collision system optimization model and the optimal solution of the heat dissipation system optimization model as the initial population of the non-dominated sorting genetic algorithm, and taking the initial population as the parent population.

[0104] In an embodiment of the present application, in an embodiment of the present application, the battery module optimization design method includes in step S520 taking the optimal solution of the collision system optimization model and the optimal solution of the heat dissipation system optimization model as the initial population of the non-dominated sorting genetic algorithm, and taking the initial population as the parent population.

[0105] Specifically, the non-dominated sorting genetic algorithm is a commonly used method in multi-objective optimization problems. It selects non-dominated solutions as the parents of the next generation population by evaluating the dominance relationship of the solutions in the solution set in the target space. Taking the optimal solutions of the collision system and the heat dissipation system as the initial population can ensure that the parent population contains some potential better solutions, thereby improving the optimization efficiency.

[0106] It is understandable that using the optimal solution obtained by particle swarm optimization as the initial population input into the genetic algorithm can make full use of the existing optimization results and further improve the optimization effect through the genetic algorithm. This method ensures that the initial population has a high quality, thereby improving the convergence speed and solution quality of the entire optimization process.

[0107] S530: Generate a child population according to the parent population.

[0108] In the embodiment of the present application, in the embodiment of the present application, the battery module optimization design method includes generating a child population according to a parent population in step S530.

[0109] Specifically, the generation of offspring populations is usually carried out through crossover and mutation operations. The crossover operation is used to exchange the genetic information of individuals in the parent population, and the mutation operation is used to explore new possible solutions in the search space. The generated offspring population will continue to be used for the next generation of optimization iterations.

[0110] It is understandable that through crossover and mutation operations, genetic algorithms can explore new design solutions, which are generated by genetic operations of individuals in the parent population, and can better expand the search space and improve the diversity of solutions, thereby more comprehensively optimizing the objective function.

[0111] S540: Generate a new population based on the initial population and the offspring population.

[0112] In the embodiment of the present application, in the embodiment of the present application, the battery module optimization design method includes generating a new population according to the initial population and the descendant population in step S540.

[0113] Specifically, after crossover and mutation operations, individuals in the initial population and the offspring population are merged to form a new population. The new population contains more solutions and can provide a wider selection space for further optimization.

[0114] It is understandable that merging the initial population and the offspring population into a new population not only increases the number of new solutions, but also enriches the solution space, which provides more possible options for subsequent optimization, thereby obtaining a better optimized solution.

[0115] S550: Perform non-dominated sorting on the individuals in the new population, and perform elite retention operation on the sorted individuals to obtain better individuals in the new population.

[0116] In an embodiment of the present application, in an embodiment of the present application, the battery module optimization design method includes in step S550 performing non-dominated sorting on the individuals in the new population, and performing an elite retention operation on the sorted individuals to obtain better individuals in the new population.

[0117] Specifically, non-dominated sorting ranks each individual according to its performance in multi-objective optimization and selects individuals with higher priority to enter the parent population of the next round. The elite retention operation ensures that high-quality individuals will not be lost in the genetic operation, effectively ensuring the quality of the solution.

[0118] It is understandable that the combination of non-dominated sorting and elite retention operations can effectively screen out the best design solution, ensuring the retention of high-quality individuals during the optimization process, thereby avoiding the loss of valuable solutions. This mechanism ensures that the optimization process continues to move towards the optimal solution.

[0119] S560: Use the better individuals as the new parent population after iterative update.

[0120] In the embodiment of the present application, in the embodiment of the present application, the battery module optimization design method includes in step S560 taking the better individuals as the new parent population after iterative update.

[0121] Specifically, the better individuals are selected from the new population through non-dominated sorting and elite retention operations, which represent the best design solutions in the current population. By using these individuals as the new parent population, the optimization process can be further driven towards a better solution.

[0122] It is understandable that continuing to iterate with high-quality individuals as the new parent population can not only continue to maintain the advantages of high-quality solutions, but also further improve the quality of solutions through continuous iterative optimization. This process ensures that the algorithm always approaches the optimal solution.

[0123] S570: Determine whether the new population meets the stopping condition.

[0124] In the embodiment of the present application, in the embodiment of the present application, the battery module optimization design method includes, in step S570, determining whether the new population satisfies the stop condition.

[0125] Specifically, the stopping condition may include reaching a preset maximum number of generations, a convergence condition, or a calculation accuracy requirement. If the stopping condition is met, the optimization process ends and enters the result output stage. Otherwise, the next round of iterations is continued, and step S530 is returned to optimize the solution set.

[0126] It is understandable that the stopping condition includes: reaching a preset maximum number of generations, convergence conditions or calculation accuracy requirements. The determination of the stopping condition is to ensure that the optimization process ends after reaching a certain convergence, avoiding unnecessary iterations, thereby saving computing resources. If the stopping condition is not reached, the optimization process continues to ensure that the algorithm fully explores the solution space until the optimal solution is obtained.

[0127] S580: Output the current population as the Pareto front.

[0128] In an embodiment of the present application, in an embodiment of the present application, the battery module optimization design method includes outputting the current population as the Pareto front in step S580.

[0129] Specifically, the Pareto front is the set of optimal solutions in a multi-objective optimization problem, which consists of all non-dominated solutions. These solutions show the best compromise among all objectives and represent different choices for design solutions. Outputting the current population as the Pareto front facilitates further analysis and selection of design solutions.

[0130] It can be understood that the output Pareto front can help decision makers make trade-offs and choices in multi-objective optimization. These solutions show the trade-offs between different objectives and provide a variety of possible solutions to choose from to meet different design requirements.

[0131] This application proposes an efficient battery module design optimization method by combining the particle swarm optimization algorithm and the non-dominated sorting genetic algorithm. First, the particle swarm optimization algorithm is used to solve the collision system optimization model and the heat dissipation system optimization model respectively to obtain their respective optimal solutions. Then, these optimal solutions are used as the initial population of the non-dominated sorting genetic algorithm, and a new population is generated through the crossover and mutation operations of the genetic algorithm. The non-dominated sorting and elite retention operations are used to screen out the better individuals and continuously optimize the design scheme. Finally, after multiple generations of iterations, the output solution set is used as the Pareto frontier, which provides the best compromise solution between different objectives for the design of the battery module.

[0132] It is understandable that by combining the particle swarm optimization algorithm with the non-dominated sorting genetic algorithm, this application achieves efficient and accurate multi-objective optimization, which can not only optimize the heat dissipation performance and crash resistance of the battery module, but also meet the design constraints while ensuring performance. The innovation of this method lies in that by combining the two optimization algorithms, the optimal solution can be found in the complex multi-objective design space, and the multi-faceted requirements of the two major systems of heat dissipation and collision are fully considered.

[0133] In an embodiment of the present application, the battery module optimization design method further includes: normalizing each solution in the Pareto front according to a standard boundary crossing method to obtain a standardized solution.

[0134] The normalized solution is calculated as:

[0135]

[0136] Among them, u a is the lower bound of the standardized solution set, u b is the upper bound of the standardized solution set.

[0137] Next, the search radius of each standardized solution is calculated with the origin as the reference. The calculation formula for the search radius is:

[0138]

[0139] Among them, j is the dimension of the target, q is a constant, R q is the search radius.

[0140] Finally, the solution with the smallest search radius among the standardized solutions is selected as the optimal solution. Figure 3 , Figure 3 This is a schematic diagram of obtaining the search radius provided in an embodiment of the present application.

[0141] Specifically, the battery module optimization design method further standardizes each solution in the Pareto frontier through the standard boundary crossing method to obtain a standardized solution. The purpose of this processing step is to map all solutions to a unified scale range for subsequent comparison and selection. In the standardized solution calculation formula, Nf i is the standardized value of the ith solution, i is the objective function value of the original solution, imax , Of imin are the minimum and maximum objective function values ​​in the original solution set, respectively. The normalization process ensures that the solution set is of the same magnitude within the normalized range, thus avoiding the problem of inconsistent dimensions of targets of different dimensions.

[0142] It can be understood that after calculating the search radius of each standardized solution, the solution with the smallest search radius is selected as the optimal solution. This step ensures that the final selected solution has a smaller error and can show the best compromise result in the overall optimization space, thereby improving the effect and performance of the battery module design.

[0143] In the embodiment of the present application, the mathematical model for optimizing heat dissipation and crashworthiness of the battery system is:

[0144]

[0145] It is understandable that the mathematical model for optimizing heat dissipation and crashworthiness of battery systems aims to optimize multiple objectives simultaneously, including the maximum intrusion Dmax, the peak collision force PCF, and the maximum temperature Tmax of the battery. This optimization problem is subject to a series of constraints, including the range of design variables, namely, the thickness of the aluminum side panel a is between 0.6mm and 1.2mm, the thickness of the foam aluminum b is between 4mm and 7mm, the gradient index of the foam aluminum n is between -1 and 1, and the maximum intrusion Dmax must be less than or equal to 4mm. These constraints ensure the feasibility and safety of the design scheme and avoid battery damage or safety hazards that may be caused by unreasonable design.

[0146] Through the optimization of the above mathematical model, a set of optimal design variables can be found, so that when the battery module is subjected to external impact, it can not only effectively control the temperature and reduce energy consumption, but also maintain the stability of the structure during the collision and reduce the safety risks caused by excessive deformation. The implementation of this optimization model can help designers obtain an optimal battery module design solution that balances heat dissipation, collision protection and other performance requirements under multiple objectives and constraints.

[0147]

[0148] Table 1 Comparison between initial design and optimized design

[0149] Please refer to Table 1, which is a comparison between the initial design and the optimized design. As shown in Table 1, through the optimized design, both the design variables and the target variables have been significantly improved, thereby improving the overall performance. Specifically, the thickness a of the aluminum side panel in the design variable is optimized from 1mm in the initial design to 1.05mm, the thickness b of the foam aluminum is optimized from 6mm in the initial design to 6.73mm, and the gradient index nnn of the foam aluminum is optimized from -1 to -0.53, showing that the optimization process can adjust the design parameters more accurately, thereby achieving the expected optimization goals.

[0150] It is understandable that the optimized design also shows positive changes in the target variables compared to the initial design. For example, the maximum temperature Tmax of the battery is reduced from 34.62℃ to 34.07℃, indicating that the heat dissipation performance has been improved; the pressure drop ΔP of the heat dissipation system is reduced from 129.71Pa to 121.44Pa, indicating that the energy consumption has been reduced; the maximum intrusion Dmax of the collision system is reduced from 3.41mm to 2.55mm, improving the crash resistance of the battery module; and the peak collision force PCF is slightly increased from 5577.89N to 5586.78N. Despite the slight improvement, the overall collision performance is still optimized.

[0151] It can be understood that, as can be seen from Table 1, the optimized design not only achieves better heat dissipation performance (by reducing Tmax and ΔP), but also achieves better performance in collision performance (by reducing Dmax). These results show that the optimization method adopted can effectively improve the overall performance of the battery module, making it reach a more ideal level in terms of heat dissipation and crash resistance, and ultimately improve the reliability and safety of the battery module.

[0152] In an embodiment of the present application, the method for constructing a hybrid multi-fidelity approximation model also includes performing reverse normalization processing on the optimal solution, restoring the normalization result to design parameters, and the restored parameters are the actual optimized design parameters of the battery module.

[0153] It is understandable that in the process of constructing the hybrid multi-fidelity approximation model, the present application also includes reverse normalization of the optimal solution, and restores the normalization result to the design parameters to obtain the actual optimized design parameters of the battery module. The reverse normalization step ensures that the objective function solution in the optimization process can be correctly mapped back to the actual design parameter space, thereby realizing a practical and feasible battery module design.

[0154] In an embodiment of the present application, a method for constructing a hybrid multi-fidelity approximation model includes generating a high-fidelity input sample set according to an optimal Latin hypercube sampling algorithm. Generating a low-fidelity input sample set according to an inherited Latin hypercube sampling algorithm. Constructing a low-fidelity model according to the low-fidelity input sample set and a polynomial response surface method. Constructing a mapping relationship between a low-fidelity response and a high-fidelity response according to a support vector regression technique. Optimizing undetermined coefficients to improve the model accuracy of the hybrid multi-fidelity approximation model.

[0155] It is understandable that in order to further improve the accuracy of the model, the hybrid multi-fidelity approximate model construction method in the embodiment of the present application also includes generating high-fidelity and low-fidelity input sample sets, and combining polynomial response surface method, support vector regression technology and other methods to establish a mapping relationship between the low-fidelity model and the high-fidelity response. Through multi-fidelity modeling, the computational cost can be effectively reduced on the basis of the high-fidelity model, and the optimization efficiency can be improved.

[0156] In an embodiment of the present application, optimizing the undetermined coefficients includes: applying a genetic algorithm to optimize the undetermined coefficients to form candidate coefficients. Performing crossover and mutation operations on the candidate coefficients to generate new candidate coefficients. Evaluating the performance of the candidate coefficients according to a fitness function. Generating a parent population according to the performance of the candidate coefficients to iterate the candidate coefficients until the candidate coefficients converge to an optimal value, thereby obtaining a hybrid multi-fidelity approximation model.

[0157] It is understandable that in order to optimize the undetermined coefficients, the embodiment of the present application uses a genetic algorithm for optimization, generates new candidate coefficients through operations such as crossover and mutation, and evaluates their performance through a fitness function. The fitness function evaluation process includes the calculation of mean square error and absolute error, ensuring that the accuracy of the model can be accurately evaluated during the optimization process. By iteratively optimizing the candidate coefficients, the optimal undetermined coefficients can eventually be found, thereby improving the accuracy of the hybrid multi-fidelity approximation model.

[0158] In the embodiment of the present application, the fitness function includes: calculating the candidate coefficients according to the model error to obtain the objective function value, the model error includes the mean square error and the absolute error. The objective function values ​​are sorted according to the error size, and the individual with the smallest error in the objective function value is selected as the new candidate coefficient.

[0159] It is understandable that the design of the fitness function is a key step in the embodiment of this application. By sorting the model errors and selecting the individuals with the smallest errors as new candidate coefficients, the accuracy and reliability of the final optimization results are guaranteed. This process effectively improves the accuracy of the hybrid multi-fidelity approximation model by continuously optimizing the coefficients, thereby promoting the optimization process of the battery module design.

[0160] This application proposes an efficient battery module optimization design method by combining the particle swarm optimization algorithm with the non-dominated sorting genetic algorithm, which significantly improves the heat dissipation performance and collision resistance of the battery module. By optimizing design variables, such as the thickness of the aluminum side panel, the thickness of the foamed aluminum, and the gradient index of the foamed aluminum, the maximum temperature Tmax and the voltage drop ΔP of the battery can be effectively controlled, thereby optimizing the heat dissipation system, reducing the temperature rise of the battery during use, reducing energy consumption and increasing the cycle life of the battery. At the same time, the optimization of the collision system enables the battery module to better maintain structural integrity when subjected to external impact, and the maximum intrusion Dmax is significantly reduced, reducing the risk of battery short circuit or thermal runaway due to excessive deformation.

[0161] In addition, this solution not only improves the optimization accuracy but also ensures the reliability of the design results by innovatively combining the multi-fidelity approximation model with the optimization algorithm. The stability of the optimization process is further improved through standardization and minimization of the search radius, ensuring that the optimized design can find a balance point among multi-dimensional objectives. Through this series of optimization steps, the final battery module design not only meets the performance requirements of heat dissipation and crash resistance, but also shows superiority in improving battery safety and extending service life. Therefore, this solution provides a comprehensive and widely applicable battery module design optimization method with high engineering application value.

[0162] The present application provides a battery module, including: a battery cell; and a module outer frame for wrapping the battery cell. Figures 1 to 3 The battery module optimization design method shown is used for optimization design.

[0163] Figure 4 This is an electronic device 20 provided in an embodiment of the present application. Figure 4 As shown, the electronic device 20 includes at least the following parts: a processor 21 and a memory 22 .

[0164] In the embodiment of the present application, the memory 22 is used to store instructions executable by the processor 21, and the processor 21 is configured to implement the following when executing the instructions: Figure 1 The battery module optimization design method shown.

[0165] In an embodiment of the present application, a computer-readable storage medium includes instructions, and the instructions instruct a device to execute a battery module optimization design method as described in one aspect. For example, the instructions instruct the device to execute Figure 1 The battery module optimization design method is shown.

[0166] The program operating in the electronic device 20 involved in one embodiment of the present application may be a program (a program that enables a computer to function) that controls a central processing unit (CPU) and the like to realize the functions of the above-mentioned embodiment involved in one scheme of the present invention. Then, the information processed by these devices is temporarily stored in a random access memory (RAM) during its processing, and then stored in various ROMs such as a read-only memory (Flash ROM) and a hard disk drive (HDD), and is read, modified, and written by the CPU as needed.

[0167] It should be noted that part of the electronic device 20 of the above embodiment may be implemented by a computer. In this case, a program for implementing the control function may be recorded in a computer-readable recording medium and implemented by reading the program recorded in the recording medium into a computer system and executing it.

[0168] It should be noted that the "computer system" mentioned here refers to a computer system built into the electronic device 20, and adopts a computer system including hardware such as an OS and peripheral devices. In addition, "computer-readable recording medium" refers to removable media such as a floppy disk, a magneto-optical disk, a ROM, a CD-ROM, and a storage device such as a hard disk built into the computer system.

[0169] Furthermore, "computer-readable recording media" may include: media that dynamically store programs for a short period of time, such as communication lines when sending programs via networks such as the Internet or communication lines such as telephone lines; and media that store programs for a fixed period of time, such as volatile memories inside computer systems that serve as servers or clients in this case. In addition, the above-mentioned program may be a program for realizing a part of the above-mentioned functions, or a program that can realize the above-mentioned functions by combining with a program already recorded in a computer system.

[0170] In addition, the electronic device 20 in the above-mentioned embodiment can also be implemented as a collection (device group) composed of multiple devices. Each device constituting the device group can have a part or all of the functions or functional blocks of the electronic device 20 in the above-mentioned embodiment. As a device group, it is sufficient to have all the functions or functional blocks of the electronic device 20.

[0171] The battery module optimization design method, battery module and storage medium provided in the embodiment of the present application can effectively obtain the optimal design parameters of the battery module by comprehensively considering the optimization objectives of the heat dissipation system and the collision protection system, combining the hybrid multi-fidelity approximation model and the response surface method. By evaluating the performance of the collision system and the heat dissipation system through a multi-objective optimization algorithm, it can ensure that the performance of the battery module in a collision and high temperature environment meets safety standards, reduce the failure rate of the battery in actual applications, and improve the long-term reliability of the battery module.

[0172] It will be easily understood by those skilled in the art that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A battery module optimization design method for optimizing a battery module, characterized in that: The battery module optimization design method comprises: Obtaining design parameters of the battery module; Obtaining a heat dissipation system optimization target and a collision system optimization target of the battery module according to the design parameters; The collision system optimization model is constructed according to the hybrid multi-fidelity approximate model construction method. The expression of the collision system optimization model is: Among them, x is the input sample, is the high-fidelity approximate response when the input is x, is a low-fidelity response, g(y LF ) is the mapping from low-fidelity output to high-fidelity output, w1, w2, γ are unknown coefficients; Construct a cooling system optimization model based on the response surface methodology; The collision system optimization model and the heat dissipation system optimization model are evaluated according to a multi-objective optimization algorithm to obtain the optimized design parameters of the battery module.

2. The battery module optimization design method according to claim 1, characterized in that: The step of evaluating the collision system optimization model and the heat dissipation system optimization model according to a multi-objective optimization algorithm to obtain the optimal design parameters of the battery module includes: Solving the collision system optimization model and the heat dissipation system optimization model respectively according to a particle swarm optimization algorithm to obtain an optimal solution of the collision system optimization model and an optimal solution of the heat dissipation system optimization model; Using the optimal solution of the collision system optimization model and the optimal solution of the heat dissipation system optimization model as the initial population of the non-dominated sorting genetic algorithm, and using the initial population as the parent population; Generate a child population according to the parent population; Generate a new population according to the initial population and the offspring population; Performing non-dominated sorting on the individuals in the new population, and performing elite retention operation on the sorted individuals to obtain better individuals in the new population; The better individuals are used as the new parent population after iterative updating.

3. The battery module optimization design method according to claim 2, characterized in that: The battery module optimization design method also includes: Determine whether the new population meets the stopping condition, wherein the stopping condition includes: reaching a preset maximum number of generations, a convergence condition, or a calculation accuracy requirement; If the stopping condition is met, the current population is output as the Pareto frontier; If the stopping condition is not met, the process returns to generating a child population according to the parent population, and continues to iterate until the stopping condition is met.

4. The battery module optimization design method according to claim 3, characterized in that: The battery module optimization design method also includes: Each solution in the Pareto front is standardized according to the standard boundary crossing method to obtain a standardized solution. The calculation formula of the standardized solution is: Among them, u a is the lower bound of the standardized solution set, u b is the upper bound of the standardized solution set; Taking the origin as a reference, the search radius of each standardized solution is calculated. The calculation formula of the search radius is: Among them, j is the dimension of the target, q is a constant, R q is the search radius; The solution with the smallest search radius among the standardized solutions is selected as the optimal solution.

5. The battery module optimization design method according to claim 4, characterized in that: The method for constructing the hybrid multi-fidelity approximation model also includes: The optimal solution is subjected to reverse normalization processing, and the normalization result is restored to the design parameters, and the restored parameters are the actual optimized design parameters of the battery module.

6. The battery module optimization design method according to claim 1, characterized in that: The method for constructing the hybrid multi-fidelity approximation model includes: Generate a high-fidelity input sample set based on the optimal Latin hypercube sampling algorithm; Generate a low-fidelity input sample set according to the inherited Latin hypercube sampling algorithm; constructing a low-fidelity model based on the low-fidelity input sample set and the polynomial response surface method; A mapping relationship between low-fidelity responses and high-fidelity responses is constructed based on support vector regression technology; The undetermined coefficients are optimized to improve the model accuracy of the hybrid multi-fidelity approximation model.

7. The battery module optimization design method according to claim 6, characterized in that: The optimizing the undetermined coefficients comprises: Applying a genetic algorithm to optimize the undetermined coefficients to form candidate coefficients; Performing crossover and mutation operations on the candidate coefficients to generate new candidate coefficients; Evaluating the performance of the candidate coefficients according to a fitness function; A parent population is generated according to the performance of the candidate coefficients to iterate the candidate coefficients until the candidate coefficients converge to an optimal value, thereby obtaining the hybrid multi-fidelity approximation model.

8. The battery module optimization design method according to claim 7, characterized in that: The fitness function includes: Calculate the candidate coefficients according to the model error to obtain the objective function value, wherein the model error includes a mean square error and an absolute error; The objective function values ​​are sorted according to the size of the errors, and the individual with the smallest error in the objective function value is selected as the new candidate coefficient.

9. A battery module, characterized in that: include: Battery cells; A module outer frame, used to wrap the battery core; Wherein, the battery module is optimized and designed using the battery module optimization design method according to any one of claims 1 to 8.

10. A computer-readable storage medium, characterized in that: The method comprises instructions for instructing a device to execute the battery module optimization design method as described in any one of claims 1 to 8.

Images