Battery pack structure optimization method

Abstract

Provided is a battery pack structure optimization method, which relates to the technical field of power batteries. The method specifically comprises the following steps: defining dimension variables of a battery pack housing; defining shape variables of a structural member of a battery pack on the basis of node coordinates representative of a shape change of the structural member; defining topological variables of the battery pack on the basis of a transfer form of the force of the structural member; defining a constraint condition for the battery pack; constructing a battery pack optimization model on the basis of the dimension variables, the shape variables, the topological variables and the constraint condition; and performing iterative solving on the battery pack optimization model on the basis of an enhanced sine-cosine algorithm, so as to obtain an optimal solution of the optimization model. The present invention aims to realize integrated optimization of the morphology and topology of a battery pack.

Description

A method for optimizing battery pack structure

[0001] This application claims priority to Chinese Patent Application No. 2024110050388, filed on July 25, 2024, entitled "A Method for Optimizing Battery Pack Structure," the contents of which are incorporated herein by reference in their entirety. Technical Field

[0002] This disclosure relates to the field of power battery technology, and in particular to a method for optimizing battery pack structure. Background Technology

[0003] With the transformation of the global energy structure and the increasing awareness of environmental protection, the development of new energy vehicles and energy storage systems has become a hot topic of social concern and a key research focus. Lithium-ion batteries, as the most mainstream energy storage medium, are widely used in new energy vehicles and various energy storage devices due to their excellent energy density, cycle life, and safety performance. As a key system that combines multiple individual batteries, the design and optimization of lithium-ion battery packs directly affect the performance, safety, and economy of the entire system.

[0004] Battery pack design is a complex multi-objective optimization problem, involving factors such as the selection of individual battery cells, module design, thermal management, structural strength, and integration with the BMS (Battery Management System). To ensure the safety of new energy vehicles and energy storage systems during use, battery pack design must meet several stringent safety indicators, including overcharge, over-discharge, overcurrent, and short-circuit protection, as well as stability under various extreme operating conditions. However, while meeting these safety requirements, battery pack design must also strive for minimal size and weight to increase the energy density of the vehicle or energy storage device, thereby extending the driving range or increasing the energy storage capacity.

[0005] Currently, battery pack design and optimization primarily rely on the optimization capabilities of simulation software. These programs typically possess powerful structural and thermal management simulation capabilities, enabling optimization of the battery pack's morphology and topology. However, because these optimization functions are usually one-dimensional, they can only improve specific optimization objectives and are unlikely to yield the optimal solution when considering multiple factors comprehensively. For example, optimizing the structural strength of the battery pack may neglect the requirements of thermal management; conversely, improving thermal management performance may increase the system's weight and volume, thereby affecting energy density.

[0006] Therefore, how to achieve integrated optimization of battery pack morphology and topology has become a technical challenge that urgently needs to be solved. Summary of the Invention

[0007] The main objective of this disclosure is to provide a battery pack structure optimization method that aims to achieve integrated optimization of battery pack morphology and topology.

[0008] To achieve the above objectives, this disclosure proposes a battery pack structure optimization method, comprising the following steps:

[0009] Define the size variables of the battery pack casing;

[0010] The shape variables of the battery pack structural components are defined based on the node coordinates that characterize the shape changes of the structural components;

[0011] Define the topological variables of the battery pack according to the force transmission form of the structural components;

[0012] Define the constraints for the battery pack;

[0013] Based on the size variables, shape variables, topology variables, and constraints, a battery pack optimization model is constructed.

[0014] The battery pack optimization model is iteratively solved using an enhanced sine and cosine algorithm to obtain the optimal solution of the optimization model.

[0015] In one embodiment of this application, the size variable includes at least one of: nodes of the shell element that can characterize the cross-sectional size, node coordinates of the shell thickness that can characterize the cross-sectional size, and node coordinates of the shell solid element that can characterize the cross-sectional size.

[0016] In one embodiment of this application, the constraints include at least one of the following: a preset maximum permissible stress of each structural component of the battery pack, a preset maximum permissible strain of each structural component of the battery pack, and a maximum permissible displacement of each structural component of the battery pack.

[0017] In one embodiment of this application, when the constraint condition is the preset maximum permissible stress of each structural component of the battery pack, the constraint condition is expressed as: g1=ε max1 -ε criteria1 ≤0;

[0018] Where g1 represents the preset maximum permissible stress of each structural component of the battery pack, ε max1 ε represents the maximum strain force of the inner bottom plate. criteria1 This indicates the maximum strain force of the inner bottom plate base material;

[0019] When the constraint condition is the preset maximum permissible strain of each structural component of the battery pack, the constraint condition is expressed as: g2=ε max2 -ε criteria2 ≤0;

[0020] Where g2 represents the preset maximum permissible strain of each structural component of the battery pack, ε max2ε represents the maximum strain of the inner bottom plate weld. criteria2 This indicates the maximum allowable strain of the inner bottom plate weld.

[0021] When the constraint condition is the maximum permissible displacement of each structural component of the battery pack, the constraint condition is expressed as: g3=ε max3 -ε criteria3 ≤0;

[0022] Where g3 represents the maximum permissible displacement of each structural component of the battery pack, ε max3 ε represents the maximum intrusion displacement of the inner bottom plate. criteria3 This indicates the maximum allowable displacement of the inner bottom plate.

[0023] In one embodiment of this application, the steps for iteratively solving the battery pack optimization model using the enhanced sine and cosine algorithm are as follows:

[0024] S1: Initialize the parameters of the battery pack optimization model and randomly generate the initial solution of the battery pack optimization model;

[0025] S2: Initialize fitness values;

[0026] S3: Determine if the current solution is less than or equal to the solution after the previous iteration. If not, retain the solution after the previous iteration, use the solution after the previous iteration as the current optimal solution, and update the verification parameters, the random parameter of the distance to the target, the weight parameter, and the control parameters.

[0027] S4: Generate a random number, determine whether the current random number is less than or equal to the control parameter; if not, generate a random value, determine whether the random value is less than or equal to the preset threshold; if not, use the first defined function to update the solution of the battery pack optimization model.

[0028] S5: Apply penalty functions to the solution of the updated battery pack optimization model to impose conditional constraints;

[0029] S6: Determine if the termination condition is met. If it is, output the optimal solution.

[0030] In one embodiment of this application, a random number is generated, and it is determined whether the current random number is less than or equal to a control parameter. If not, a random value is generated, and it is determined whether the random value is less than or equal to a preset threshold. If so, the solution of the battery pack optimization model is updated using a second defined function.

[0031] In one embodiment of this application, a random number is generated, and it is determined whether the current random number is less than or equal to the control parameter. If not, it is determined whether the current iteration number is even. If so, the solution of the battery pack optimization model is updated using a third defined function.

[0032] In one embodiment of this application, a random number is generated, and it is determined whether the current random number is less than or equal to the control parameter. If not, it is determined whether the current iteration number is even. If not, the solution of the battery pack optimization model is updated using the fourth defined function.

[0033] In one embodiment of this application, the first defined function (a), the second defined function (b), the third defined function (c), and the fourth defined function (d) are respectively represented as follows: r1 = c(1-t / T); r5=a max -a min (t / T);

[0034] Where t represents the current iteration number; T represents the total number of iterations; j represents the j-th individual; and i represents the i-th dimension variable; This represents the position of the j-th individual in the i-th dimension during the t-th iteration; Let r1 represent the position of Q in the i-th dimension at the t-th iteration; r2 represent the verification parameter used to determine the region where the newly generated solution is located; and r3 represent the random parameter for the distance to the target, defined in [0, 2π]. The distance is a random value; r3 is a weight parameter, a random weight value defined in [0,2], used to adjust the distance. The effect on the defined distance; r4 is a random value in [0,1]; c is a constant, taken as 2; |·| represents the absolute value; and Let represent the i-th dimension positions of the top three solutions after t iterations. Γ represents the dot product; levy(β) represents Lévy flight; Γ represents the gamma function, with β set to 1.5; θ i For a random value defined in the range [0, 2π]; It is a parameter that controls the search range. It is the optimal fitness value after t iterations. It is the fitness value of the j-th individual in the t-th iteration, r5 represents the control parameter, and a max a min Defined in [0,1], and a max >a min .

[0035] In one embodiment of this application, the penalty function is expressed as: h(t) = 1 + t / T;

[0036] F(x) represents the fitness value being penalized; f(x) represents the objective function value; m represents the number of constraints; h(t) is the penalty parameter; t represents the current iteration number; T represents the total number of iterations; g i(x) is the constraint function.

[0037] By employing the above technical solution and defining variables at three levels—size, shape, and topology—comprehensive optimization of the battery pack design can be achieved, adapting to different design requirements and constraints. The enhanced sine and cosine algorithms can effectively search for the globally optimal solution, ensuring optimal performance of the battery pack in terms of weight, stiffness, and strength. The optimized battery pack design can minimize material usage, reduce manufacturing costs, and improve economic efficiency. Attached Figure Description

[0038] The present disclosure will now be described in detail with reference to specific embodiments and accompanying drawings, wherein:

[0039] Figure 1 is a flowchart illustrating a first embodiment of this disclosure;

[0040] Figure 2 is a flowchart of the steps for iteratively solving the battery pack optimization model according to the enhanced sine and cosine algorithm. Detailed Implementation

[0041] To make the objectives, technical solutions, and advantages of this disclosure clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the following specific embodiments are merely illustrative of this disclosure and do not constitute a limitation thereof.

[0042] As shown in Figures 1 and 2, in order to achieve the above objectives, this disclosure proposes a battery pack structure optimization method, including the following steps:

[0043] Define the size variables of the battery pack casing;

[0044] The shape variables of the battery pack structural components are defined based on the node coordinates that characterize the shape changes of the structural components;

[0045] Define the topological variables of the battery pack according to the force transmission form of the structural components;

[0046] Define the constraints for the battery pack;

[0047] Based on the size variables, shape variables, topology variables, and constraints, a battery pack optimization model is constructed.

[0048] The battery pack optimization model is iteratively solved using an enhanced sine and cosine algorithm to obtain the optimal solution of the optimization model.

[0049] Specifically, determine the dimensional variables of the battery pack casing. This includes the length, width, and height of the casing, as well as the specific dimensions of each component.

[0050] The shape variables of the internal structural components of the battery pack are defined by node coordinates. The coordinates of each node can represent the specific shape changes of the structural component.

[0051] The topology variables of a battery pack refer to the connection relationships between structural components and the force transmission paths. By defining topology variables, the connection forms between structural components and the force transmission methods can be determined, thereby optimizing mechanical performance.

[0052] Constraints include various limitations imposed on the battery pack during use, such as maximum stress and deformation. These constraints need to be met during the optimization process to ensure the safety and reliability of the battery pack in actual use.

[0053] Based on the defined size, shape, topology, and constraints, an optimization model for the battery pack is constructed. Finally, an enhanced sine / cosine algorithm is used to iteratively solve the optimization model. The enhanced sine / cosine algorithm is an optimization algorithm that searches for the optimal solution by simulating the fluctuation characteristics of sine and cosine functions. Through multiple iterative calculations, the optimization objective is gradually approximated, ultimately yielding the optimal battery pack design.

[0054] By employing the above technical solution and defining variables at three levels—size, shape, and topology—comprehensive optimization of the battery pack design can be achieved, adapting to different design requirements and constraints. The enhanced sine and cosine algorithms can effectively search for the globally optimal solution, ensuring optimal performance of the battery pack in terms of weight, stiffness, and strength. The optimized battery pack design can minimize material usage, reduce manufacturing costs, and improve economic efficiency.

[0055] In one embodiment of this application, the size variable includes at least one of: nodes of the shell element that can characterize the cross-sectional size, node coordinates of the shell thickness that can characterize the cross-sectional size, and node coordinates of the shell solid element that can characterize the cross-sectional size.

[0056] Specifically, shell elements are a commonly used simplified model in finite element analysis to simulate thin-walled structures. Nodes that characterize the cross-sectional dimensions are those used to define the cross-sectional dimensions of the shell element. By changing the coordinates of these nodes, the cross-sectional dimensions of the shell element can be adjusted. Shell elements simplify the modeling process of actual structures, reduce computational complexity, and improve computational efficiency. By adjusting the node coordinates, the cross-sectional dimensions of the shell element can be quickly changed to adapt to different design requirements.

[0057] Shell thickness is a crucial parameter of shell elements, determining the structure's stiffness and strength. The nodal coordinates characterizing the cross-sectional dimensions are those used to define the shell thickness; adjusting these coordinates alters the shell thickness. By adjusting the shell thickness, an optimal balance between weight and strength can be achieved, improving structural performance. Properly adjusting the shell thickness can reduce material usage and lower costs while still meeting performance requirements.

[0058] Solid shell elements are a type of finite element used to simulate thicker structures. The node coordinates, which characterize the cross-sectional dimensions, are the nodes used to define the cross-sectional dimensions of the solid shell element. By changing the coordinates of these nodes, the cross-sectional dimensions of the solid shell element can be adjusted. Solid elements can more accurately simulate the mechanical behavior of thicker structures and are suitable for the analysis of complex structures. By adjusting the node coordinates, various cross-sectional dimensions can be designed to meet different engineering needs.

[0059] In one embodiment of this application, the constraints include at least one of the following: a preset maximum permissible stress of each structural component of the battery pack, a preset maximum permissible strain of each structural component of the battery pack, and a maximum permissible displacement of each structural component of the battery pack.

[0060] Specifically, the preset maximum permissible stress refers to the maximum stress value that each structural component of the battery pack can withstand during the design process. This stress value is usually determined based on material properties, design specifications, and safety requirements.

[0061] By setting a maximum permissible stress, structural components can be prevented from failing due to overload during use, ensuring the safety of the battery pack. Proper stress distribution fully utilizes the strength properties of materials, avoids material waste, and improves structural efficiency.

[0062] The preset maximum permissible strain refers to the maximum amount of deformation (strain) that each structural component of the battery pack can withstand during the design process. The strain value is usually set based on the material's deformation capacity and the operating conditions.

[0063] By setting a maximum permissible strain, excessive deformation of structural components can be prevented during use, ensuring the functional and shape stability of the battery pack. Controlling the strain within a reasonable range helps extend the service life of structural components and reduces fatigue and damage caused by deformation.

[0064] The maximum permissible displacement refers to the maximum amount of displacement allowed for each structural component of the battery pack during the design process. This constraint is usually determined based on the functional requirements and assembly tolerances of the structural components.

[0065] By limiting the maximum displacement, it is possible to ensure that structural components maintain their functional positions during operation, avoiding failure due to excessive displacement. Proper control of displacement ensures accurate assembly of all parts of the battery pack, improving the overall assembly precision.

[0066] In one embodiment of this application, when the constraint condition is the preset maximum permissible stress of each structural component of the battery pack, the constraint condition is expressed as: g1=ε max1 -ε criteria1 ≤0;

[0067] Where g1 represents the preset maximum permissible stress of each structural component of the battery pack, ε max1ε represents the maximum strain force of the inner bottom plate. criteria1 This indicates the maximum strain force of the inner bottom plate base material;

[0068] When the constraint condition is the preset maximum permissible strain of each structural component of the battery pack, the constraint condition is expressed as: g2=ε max2 -ε criteria2 ≤0;

[0069] Where g2 represents the preset maximum permissible strain of each structural component of the battery pack, ε max2 ε represents the maximum strain of the inner bottom plate weld. criteria2 This indicates the maximum allowable strain of the inner bottom plate weld.

[0070] When the constraint condition is the maximum permissible displacement of each structural component of the battery pack, the constraint condition is expressed as: g3=ε max3 -ε criteria3 ≤0;

[0071] Where g3 represents the maximum permissible displacement of each structural component of the battery pack, ε max3 ε represents the maximum intrusion displacement of the inner bottom plate. criteria3 This indicates the maximum allowable displacement of the inner bottom plate.

[0072] Specifically, g1 = ε max1 -ε criteria1 The constraint ≤0 is used to control the maximum stress on the inner bottom plate of the battery pack to not exceed the material's stress limit. As a crucial load-bearing structure of the battery pack, the inner bottom plate must be maintained within a safe stress range during use. During optimization, the maximum strain ε of the inner bottom plate under various operating conditions is calculated. max1 And compare it with the material's maximum allowable strain ε criteria1 By comparison, if g1≤0, it means that the stress constraint conditions are met and the design is safe.

[0073] g2=ε max2 -ε criteria2 The constraint ≤0 is used to control the maximum strain of the inner bottom plate weld to not exceed the material's strain limit. As a structural connection, the strength and deformation capacity of the weld directly affect the overall reliability of the battery pack. During optimization, the maximum strain ε of the inner bottom plate weld is calculated under various operating conditions. max2 and compare it with the maximum allowable strain ε of the weld. criteria2 Compare the results. If g2 ≤ 0, it means the strain constraint conditions are met and the design is safe.

[0074] g3=ε max3 -ε criteria3The constraint ≤0 is used to control the maximum displacement of the inner bottom plate to not exceed the maximum allowable displacement in the design. Excessive displacement may lead to failure or interference of the internal structure of the battery pack, affecting its normal operation. During the optimization process, the maximum intrusion displacement ε of the inner bottom plate under various operating conditions is calculated. max3 and compare it with the maximum allowable displacement ε criteria3 Compare the results. If g3 ≤ 0, it means the displacement constraint conditions are met and the design is safe.

[0075] In one embodiment of this application, the steps for iteratively solving the battery pack optimization model using the enhanced sine and cosine algorithm are as follows:

[0076] S1: Initialize the parameters of the battery pack optimization model and randomly generate the initial solution of the battery pack optimization model;

[0077] S2: Initialize fitness values;

[0078] S3: Determine if the current solution is less than or equal to the solution after the previous iteration. If not, retain the solution after the previous iteration, use the solution after the previous iteration as the current optimal solution, and update the verification parameters, the random parameter of the distance to the target, the weight parameter, and the control parameters.

[0079] S4: Generate a random number, determine whether the current random number is less than or equal to the control parameter; if not, generate a random value, determine whether the random value is less than or equal to the preset threshold; if not, use the first defined function to update the solution of the battery pack optimization model.

[0080] S5: Apply penalty functions to the solution of the updated battery pack optimization model to impose conditional constraints;

[0081] S6: Determine if the termination condition is met. If it is, output the optimal solution.

[0082] Specifically, determine the range of design variables for the battery pack optimization model. Use a random number generator to generate a set of initial solutions within the range of design variables. Record the initial solutions and their corresponding parameter values.

[0083] Calculate the fitness value of the initial solution based on the optimization objective function. Record the calculation result as the initial fitness value.

[0084] Calculate the fitness value of the current solution and compare it with the fitness value of the solution after the last iteration.

[0085] If the fitness value of the current solution is not better than the solution after the previous iteration, retain the solution after the previous iteration.

[0086] Update the validation parameters, the random parameter for the distance to the target, the weight parameters, and the control parameters to prepare for the next iteration.

[0087] Generate a random number and compare it with the control parameter. If the random number is not less than the control parameter, generate another random value and compare it with a preset threshold. If the random value is not less than the preset threshold, update the solution of the optimization model using the first defined function. Check whether the updated solution violates any constraints. For solutions that violate constraints, calculate the corresponding penalty value. Adjust the fitness value, adding the penalty value to the fitness value.

[0088] Check if the current iteration meets the termination conditions (such as maximum number of iterations, rate of change of fitness value, etc.). If the termination conditions are met, output the current optimal solution and end the optimization process. If the termination conditions are not met, return to step S3 and continue iterative optimization.

[0089] By employing the above technical solutions, the random initial solution and random parameter generation methods improve global search capabilities, avoiding local optima and finding the global optimum. Fitness values ​​are updated and compared in each iteration, ensuring continuous optimization and improvement of the optimal solution. The penalty function method flexibly handles various constraints, ensuring the feasibility and practical operability of the optimized solution. Adaptive parameter adjustment improves the algorithm's convergence speed and stability, making the optimization process more efficient and reliable.

[0090] In one embodiment of this application, a random number is generated, and it is determined whether the current random number is less than or equal to a control parameter. If not, a random value is generated, and it is determined whether the random value is less than or equal to a preset threshold. If so, the solution of the battery pack optimization model is updated using a second defined function.

[0091] Using the above technical solution, the first step of random number determination provides global search capability, enabling the algorithm to explore a larger solution space. The second step of preset threshold determination provides local search capability, allowing the algorithm to perform fine-grained searches in local regions. The first random number determination ensures that the algorithm has a certain search capability in the global scope, avoiding local optima. The second random value determination ensures that fine-grained searches in local regions can be performed when necessary, guaranteeing the depth and accuracy of the search.

[0092] In one embodiment of this application, a random number is generated, and it is determined whether the current random number is less than or equal to the control parameter. If not, it is determined whether the current iteration number is even. If so, the solution of the battery pack optimization model is updated using a third defined function.

[0093] By adopting the above technical solution and introducing a third definition function and determining the iteration count, the optimization process can employ diverse update strategies, avoiding the limitations of a single method. Dynamically adjusting the update strategy based on the parity of the iteration count improves the algorithm's flexibility and adaptability. Using different definition functions to update the model solution in different iteration cycles makes the optimization process more diverse, enabling better exploration of the solution space and avoiding getting trapped in local optima that might result from a single method.

[0094] In one embodiment of this application, a random number is generated, and it is determined whether the current random number is less than or equal to the control parameter. If not, it is determined whether the current iteration number is even. If not, the solution of the battery pack optimization model is updated using the fourth defined function.

[0095] By adopting the above technical solution and introducing a fourth definition function and determining the iteration count, the optimization process can employ diverse update strategies, avoiding the limitations of a single method. Dynamically adjusting the update strategy based on the parity of the iteration count improves the algorithm's flexibility and adaptability. Using different definition functions to update the model solution in different iteration cycles makes the optimization process more diverse, enabling better exploration of the solution space and avoiding getting trapped in local optima that might result from a single method.

[0096] In one embodiment of this application, the first defined function (a), the second defined function (b), the third defined function (c), and the fourth defined function (d) are respectively represented as follows: r1 = c(1-t / T); r5=a max -a min (t / T);

[0097] Where t represents the current iteration number; T represents the total number of iterations; j represents the j-th individual; and i represents the i-th dimension variable; This represents the position of the j-th individual in the i-th dimension during the t-th iteration; Let r1 represent the position of Q in the i-th dimension at the t-th iteration; r2 represent the verification parameter used to determine the region where the newly generated solution is located; and r3 represent the random parameter for the distance to the target, defined in [0, 2π]. The distance is a random value; r3 is a weight parameter, a random weight value defined in [0,2], used to adjust the distance. The effect on the defined distance; r4 is a random value in [0,1]; c is a constant, taken as 2; |·| represents the absolute value; and Let represent the i-th dimension positions of the top three solutions after t iterations. Γ represents the dot product; levy(β) represents Lévy flight; Γ represents the gamma function, with β set to 1.5; θ i For a random value defined in the range [0, 2π]; It is a parameter that controls the search range. It is the optimal fitness value after t iterations. It is the fitness value of the j-th individual in the t-th iteration, r5 represents the control parameter, and a max a min Defined in [0,1], and a max >amin .

[0098] By employing the above technical solution, the first defined function (a), second defined function (b), third defined function (c), and fourth defined function (d) are updated under different conditions to improve the diversity of solutions and avoid getting trapped in local optima. The first and second defined functions provide diverse search paths through different trigonometric functions (sin and cos) and random parameters, enhancing the global search capability. The third and fourth defined functions switch search modes based on the parity of the iteration count, providing different local search strategies and improving the algorithm's adaptability. Through the Lévy flight mechanism, a random search strategy of long jumps and short jumps is introduced to further enhance the global search capability, especially when dealing with complex multimodal optimization problems.

[0099] In one embodiment of this application, the penalty function is expressed as: h(t) = 1 + t / T;

[0100] F(x) represents the fitness value being penalized; f(x) represents the objective function value; m represents the number of constraints; h(t) is the penalty parameter; t represents the current iteration number; T represents the total number of iterations; g i (x) is the constraint function.

[0101] Using the above technical solution, the penalty parameter is dynamically adjusted via h(t) = 1 + t / T, with the penalty intensity gradually increasing with the number of iterations. In the initial iterations, the penalty intensity is relatively small, allowing the algorithm to explore the solution space under more lenient conditions; in later iterations, the penalty intensity gradually increases, prompting the solution to converge towards satisfying the constraints. The penalty function directly acts on the fitness value f(x), making the penalized fitness value F(x) better reflect the true quality of the solution. This ensures that during the optimization process, the solution not only pursues the optimal objective function value but also satisfies the constraints, improving the practical feasibility of the optimization results.

[0102] The above description is merely a preferred embodiment of this disclosure and does not limit the patent scope of this disclosure. Any equivalent structural transformations made based on the inventive concept of this disclosure and the contents of the specification and drawings of this disclosure, or direct / indirect applications in other related technical fields, are included within the patent protection scope of this disclosure. Industrial applicability:

[0103] The solution provided in this application can be applied to the field of power battery technology. In this application, by defining variables at three levels—size, shape, and topology—comprehensive optimization of the battery pack design can be achieved, adapting to different design requirements and constraints. Employing an enhanced sine and cosine algorithm, the globally optimal solution can be effectively searched, ensuring optimal performance of the battery pack in terms of weight, stiffness, and strength. The optimized battery pack design can minimize material usage, reduce manufacturing costs, and improve economic efficiency.

Claims

1. A battery pack structure optimization method, comprising the following steps: defining a size variable of a battery pack shell; defining a shape variable of a battery pack structure according to node coordinates capable of representing shape changes of a structure; defining a topology variable of the battery pack according to a transmission form of a force of the structure; defining a constraint condition of the battery pack; constructing a battery pack optimization model according to the size variable, the shape variable, the topology variable, and the constraint condition; iteratively solving the battery pack optimization model according to an enhanced sine cosine algorithm to obtain an optimal solution of the optimization model.

2. The battery pack structure optimization method of claim 1, wherein, The size variable comprises at least one of node coordinates capable of representing a cross-sectional dimension of a shell element, node coordinates capable of representing a thickness of the shell, and node coordinates capable of representing a cross-sectional dimension of a solid element of the shell.

3. The battery pack structure optimization method of claim 1, wherein, The constraint condition comprises at least one of a preset maximum allowable stress of each structure of the battery pack, a preset maximum allowable strain of each structure of the battery pack, and a maximum allowable displacement of each structure of the battery pack.

4. The battery pack structure optimization method of claim 3, wherein, When the constraint condition is a preset maximum allowable stress of each structural member of the battery pack, the constraint condition is represented as: g1=ε max1 -ε criteria1 ≤0; wherein g1 represents a preset maximum allowable stress of each structural member of the battery pack, ε max1 represents the maximum strain force of the inner bottom plate, ε criteria1 represents the maximum strain force of the inner bottom plate base material; When the constraint condition is the preset maximum allowable strain of each structure of the battery pack, the constraint condition is expressed as: g2 = ε max2 - ε criteria2 ≤ 0; wherein g2 represents a preset maximum allowable strain of each structural member of the battery pack, ε max2 represents the maximum strain of the inner bottom plate weld, ε criteria2 represents the maximum allowable strain of the inner bottom plate weld; When the constraint condition is the maximum allowable displacement of each structure of the battery pack, the constraint condition is expressed as: g3 = ε max3 - ε criteria3 ≤ 0; wherein g3 represents the maximum allowable displacement of each structural member of the battery pack, ε max3 represents the maximum intrusion displacement of the inner floor panel, ε criteria3 represents the maximum allowable displacement of the inner floor panel.

5. The battery pack structure optimization method of claim 1, wherein, The step of iteratively solving the battery pack optimization model according to the enhanced sine cosine algorithm is as follows: S1: initializing parameters of the battery pack optimization model, and randomly generating an initial solution of the battery pack optimization model; S2: initializing a fitness value; S3: determining whether a current solution is less than or equal to a solution after a previous iteration, if not, retaining the solution after the previous iteration, taking the solution after the previous iteration as a current optimal solution, and updating a verification parameter, a random parameter of a distance to a target, a weight parameter, and a control parameter; S4: generating a random number, determining whether the current random number is less than or equal to the control parameter, if not, generating a random value, determining whether the random value is less than or equal to a preset threshold, if not, updating a solution of the battery pack optimization model by using a first definition function; S5: performing conditional constraint on the updated solution of the battery pack optimization model by using a penalty function; S6: determining whether a termination condition is met, if yes, outputting the optimal solution.

6. The battery pack structure optimization method of claim 5, wherein, Generating a random number, determining whether the current random number is less than or equal to the control parameter, if not, generating a random value, determining whether the random value is less than or equal to a preset threshold, if yes, updating the solution of the battery pack optimization model by using a second definition function.

7. The battery pack structure optimization method of claim 6, wherein, Generating a random number, determining whether the current random number is less than or equal to the control parameter, if not, determining whether a current iteration number is even, if yes, updating the solution of the battery pack optimization model by using a third definition function.

8. The battery pack structure optimization method of claim 7, wherein, Generating a random number, determining whether the current random number is less than or equal to the control parameter, if not, determining whether the current iteration number is even, if not, updating the solution of the battery pack optimization model by using a fourth definition function.

9. The battery pack structure optimization method of claim 8, wherein, The first definition function (a), the second definition function (b), the third definition function (c), and the fourth definition function (d) are respectively represented as: r1 = c (1 - t / T); r5 = a max -a min (t / T); wherein t denotes the current iteration number; T denotes the total iteration number; j denotes the jth individual; i denotes the ith dimension variable; represents the position of the jth individual in the ith dimension at the tth iteration; Q represents the position of the i-th dimension at the t-th iteration; r1 represents a validation parameter for determining the region in which the newly generated solution lies; r2 represents a random parameter to the target distance, defined in [0, 2π] to the a random value of the distance; r3 is a weight parameter, a random weight value defined in [0, 2], used to adjust an influence on the defined distance; r4is a random value in [0,1]; c is a constant, taken as 2; | · | denotes the absolute value; and respectively denote the i-th dimensional position of the top three solutions at the t-th iteration, denotes the dot product; levy(β) denotes a Lévy flight; Γ denotes the gamma function, β takes the value 1.5; θ i is a random value defined in [0, 2π]; is a parameter that controls the search range, is the best fitness value of the tth iteration, is the fitness value of the jth individual at the tth iteration, r5represents a control parameter, a max , a min is defined in [0, 1] and a max > a min .

10. The battery pack structure optimization method of claim 9, wherein, The penalty function is expressed as: h(t) = 1 + t / T; F(x) represents the fitness value being penalized; f(x) represents the objective function value; m represents the number of constraints; h(t) is a penalty parameter; t represents the current iteration number; T represents the total iteration number; g i (x) is a constraint function.

Images