A multi-objective optimization method for battery thermal management system

Abstract

The application belongs to the technical field of new energy automobile battery thermal management, and discloses a battery thermal management system multi-objective optimization method, which comprises the following steps: constructing a double-liquid-cooled-plate battery thermal management system geometric model, comparing the heat dissipation and flow performance of four kinds of fins, i.e., square, circular, fish-shaped and snake-shaped, and selecting the snake-shaped fin with the optimal comprehensive performance; selecting the flow channel width, height, fin bending radius, cooling liquid inlet flow rate and temperature as design variables, and determining the mutual restriction relationship among the variables by the control variable method; obtaining a sample set based on Latin hypercube sampling, and respectively establishing a radial basis function proxy model of the maximum battery temperature and a response surface proxy model of the system pressure drop; combining a multi-objective evolutionary algorithm to solve the Pareto front, and obtaining and verifying the optimal design parameter combination. The multi-objective collaborative optimization assisted by the proxy model can significantly reduce the maximum battery temperature and greatly reduce the system pressure drop, and comprehensively improve the heat dissipation performance and energy consumption efficiency.

Description

Technical Field

[0001] This invention belongs to the technical field of thermal management of new energy vehicle batteries, and more specifically, relates to a multi-objective optimization method for a battery thermal management system. Background Technology

[0002] With the advancement of global energy transition and carbon neutrality goals, the new energy vehicle industry is experiencing unprecedented rapid development. Lithium-ion batteries, due to their high energy density, long cycle life, and low self-discharge rate, have become the mainstream power source for electric vehicles. However, lithium-ion batteries are extremely sensitive to operating temperature; their optimal operating temperature range is typically 20°C to 40°C, and the maximum temperature difference within the battery module should be controlled within 5°C. During charging and discharging, batteries generate significant Joule heating and electrochemical reaction heat. If this heat cannot be dissipated in time or if the temperature distribution is severely uneven, it will directly lead to increased internal resistance, decreased usable capacity, and shortened cycle life. In severe cases, it can even cause thermal runaway accidents, posing a significant threat to vehicle safety. Therefore, designing an efficient and reliable battery thermal management system has become a key technical challenge in the development of new energy vehicles.

[0003] Currently, the main technologies for thermal management of power batteries are divided into four categories: air cooling, liquid cooling, phase change material cooling, and direct cooling. Among them, liquid cooling solutions have been most widely used in passenger vehicles due to their comprehensive advantages such as high heat dissipation efficiency, compact structure, and ease of integration. A typical liquid cooling system arranges liquid cooling plates at the bottom or side of the battery module, using the cooling medium flowing through the channels within the plates to remove the heat generated by the battery, thus achieving active temperature control. To improve the heat exchange capacity of liquid cooling systems, researchers have conducted extensive work from multiple dimensions, including channel topology, enhanced heat exchange structures, and cooling medium parameters. For example, some studies have enhanced convective heat transfer between the coolant and the plate wall by adding flow-disrupting elements such as pin fins, corrugated fins, or baffles within the channels; other studies have adopted unconventional topologies such as biomimetic fractal channels and spiral channels to achieve better flow distribution and temperature uniformity.

[0004] However, while enhanced heat exchange measures improve heat dissipation performance, they inevitably lead to increased flow resistance and pump power consumption. Adding turbulence-inducing elements or employing complex flow channel configurations significantly increases coolant flow resistance, resulting in higher system pressure drop. This necessitates higher-power pumps to drive coolant circulation, directly reducing the vehicle's driving range. The inherent contradiction between heat dissipation performance and system energy consumption constitutes the most critical trade-off in liquid cooling system design. Existing research largely focuses on optimizing single performance indicators, such as simply pursuing the lowest battery temperature or minimum pressure drop, without fully considering the complex interactions and coupling effects between multiple parameters. While some studies involve multi-parameter analysis, they often employ traditional single-factor or orthogonal experimental methods, making it difficult to fully reveal the nonlinear mapping relationship between design variables and multiple performance responses, and even more difficult to efficiently search for the global optimum within the design space. Furthermore, while high-fidelity numerical simulation can accurately predict the performance of battery thermal management systems, its computational cost is high, with a single simulation taking hours or even longer. Directly embedding iterative optimization processes would lead to an unbearable computational burden, limiting optimization efficiency in engineering applications.

[0005] Therefore, there is an urgent need to propose a multi-objective optimization method for battery thermal management systems, which aims to achieve the comprehensive optimization of heat dissipation performance and energy consumption efficiency through a multi-objective collaborative optimization framework assisted by biomimetic fin structure design and surrogate model. Summary of the Invention

[0006] To address the aforementioned deficiencies or improvement needs of existing technologies, this invention provides a multi-objective optimization method for battery thermal management systems. The aim is to enhance heat transfer by introducing biomimetic serpentine fins and constructing an optimization framework based on a surrogate model and a multi-objective evolutionary algorithm to achieve the comprehensive optimization of heat dissipation performance and energy consumption efficiency. This solves the technical problems of prominent contradictions between heat dissipation performance and flow resistance in liquid cooling systems, insufficient multi-parameter collaborative optimization methods, and the difficulty of directly using high-cost simulations for iterative optimization.

[0007] To achieve the above objectives, according to one aspect of the present invention, a multi-objective optimization method for a battery thermal management system is provided, comprising the following steps: S1: Construct a geometric model of the battery thermal management system, set up fins of various shapes in the liquid cooling plate flow channel for simulation comparison, and select the serpentine fin with the best overall performance to be arranged in the flow channel; S2: Select multiple design variables and conduct parameter influence analysis using the control variable method to determine the mutual constraints between the design variables; S3: Based on Latin hypercube sampling, sample datasets are obtained, and surrogate models for the highest battery temperature and system voltage drop are constructed respectively. S4: Using the battery maximum temperature proxy model and the system voltage drop proxy model as optimization objective functions, and combining the multi-objective evolutionary algorithm to solve for the Pareto optimal solution set, and selecting the optimal compromise solution from the Pareto optimal solution set; S5: Reconstruct the geometric model based on the design parameter combination corresponding to the optimal compromise solution and perform high-precision simulation verification to determine the final structural dimensions and operating parameters.

[0008] Preferably, step S1 specifically includes: S11: Construct a geometric model of a battery thermal management system that includes multiple batteries, thermal interface materials, and dual liquid cooling plates; S12: Four types of rib structures, namely square ribs, circular ribs, fish-shaped ribs and snake-shaped ribs, are respectively set in the liquid cooling plate flow channel, and the hydraulic diameter of each rib is consistent. S13: Select the highest battery temperature, battery module temperature difference, system voltage drop and Nusselt number as evaluation indicators for simulation comparison; S14: Comprehensively evaluate the heat dissipation performance and flow resistance of each fin configuration, and select the serpentine fin with the best overall performance to arrange in the flow channel.

[0009] Preferably, step S2 specifically includes: S21: Establish a numerical simulation model based on the geometric model of the battery thermal management system, and set the flow and heat transfer boundary conditions, initial conditions and material property parameters; S22: Select the flow channel width, flow channel height, fin bending radius, coolant inlet velocity, and coolant inlet temperature as design variables; S23: By changing the values ​​of each design variable using the control variable method, the control equations are solved to obtain temperature field and flow field data, and the influence of each design variable on the battery's maximum temperature, system voltage drop, battery module temperature difference, and Nusselt number is quantitatively analyzed. S24: Based on the analysis results, it is determined that there are mutual constraints among the design variables, and the multi-objective optimization method is used to perform collaborative optimization of multiple parameters.

[0010] Preferably, step S3 specifically includes the following sub-steps: S31: Within the range of five design variables—channel width, channel height, fin bending radius, coolant inlet velocity, and coolant inlet temperature—multiple sample points are obtained through Latin hypercube sampling. S32: Based on the sample dataset, candidate surrogate models for the battery's highest temperature and system voltage drop are established using radial basis function model, second-order response surface model, and Kriging model, respectively. S33: The accuracy of each candidate surrogate model is evaluated using the test set. The evaluation metrics include median absolute deviation, mean absolute error, and root mean square error. S34: Select the model with the smallest prediction error based on the evaluation results: select the radial basis function model for the highest battery temperature and the second-order response surface model for the system voltage drop.

[0011] Preferably, the specific range of the design variables is as follows: the flow channel width is 18–22 mm; the flow channel height is 1.8–2.2 mm; the rib bending radius is 1.8–2.2 mm; the coolant inlet velocity is 0.3–0.5 m / s; and the coolant inlet temperature is 293.15–303.15 K.

[0012] Preferably, the expression for the radial basis function surrogate model is:

[0013] In the formula, To predict the response, For basis function weights, For radial basis functions, The Euclidean distance between the input variable and the sample points. The total number of training samples used to construct the surrogate model The sample number; The expression for the second-order response surface surrogate model is:

[0014] In the formula, To predict the response, , , For the undetermined regression coefficients, For random error, As the first design variable, This is the second design variable.

[0015] Preferably, step S4 specifically includes the following sub-steps: S41: Use the battery maximum temperature proxy model and the system voltage drop proxy model as the optimization objective functions; S42: Iterative optimization is performed by combining a multi-objective evolutionary algorithm, wherein the multi-objective evolutionary algorithm includes at least one of the multi-objective particle swarm optimization algorithm, the second-generation non-dominated sorting genetic algorithm, or a multi-objective evolutionary algorithm based on the dominant region. S43: Obtain the Pareto optimal solution set through iterative calculation; S44: Select the optimal compromise solution from the Pareto optimal solution set, which achieves a balance between the battery's maximum temperature and the system voltage drop.

[0016] Preferably, step S5 specifically includes the following sub-steps: S51: Based on the design parameter combination corresponding to the optimal compromise solution, determine the final structural dimensions and operating parameters of the battery thermal management system; S52: Reconstruct the geometric model of the battery thermal management system based on the optimal design parameter combination and perform high-precision simulation verification; S53: By comparing the relative errors between the predicted values ​​of the surrogate model and the high-precision simulation values, the accuracy and feasibility of the optimization scheme are verified.

[0017] Preferably, the serpentine fins extend in a continuous curved shape along the flow direction of the coolant, and their curved profile is used to smoothly guide the fluid flow around, so as to reduce flow separation and shape resistance, while extending the effective flow path of the fluid and increasing the heat exchange area.

[0018] According to another aspect of the present invention, a battery thermal management system is provided, which is designed based on the above-described multi-objective optimization method for battery thermal management systems.

[0019] In summary, compared with the prior art, the multi-objective optimization method for a battery thermal management system provided by this invention has the following advantages: 1. A radial basis function proxy model for the highest battery temperature and a response surface proxy model for the system voltage drop are established for the proposed biomimetic serpentine fin structure. This facilitates rapid optimization of the cold plate structure parameters and operating parameters under different practical application conditions, providing a scalable and efficient thermal management solution for high heat flux density battery modules. 2. This invention replaces high-cost numerical simulation with a surrogate model and combines a multi-objective evolutionary algorithm to optimize geometric and thermofluid parameters in a coordinated manner. It can quickly find the Pareto optimal solution set within the design space, achieving a simultaneous reduction in the battery's maximum temperature and the system's voltage drop, and significantly improving the overall energy efficiency of the system. 3. This invention introduces biomimetic serpentine fins into the flow channel of the liquid cooling plate. Compared with traditional square and circular fins, the continuous curved contour of the serpentine fins can reduce flow separation and shape resistance, while extending the effective flow path of the fluid and increasing the heat exchange area. It can significantly improve the heat exchange intensity and temperature uniformity while effectively controlling the flow resistance and system pressure drop, thus alleviating the contradiction between enhanced heat exchange and increased energy consumption. 4. Compared with the initial unoptimized serpentine fin model, the optimized battery thermal management system of this invention achieves a 6.98K reduction in the highest battery temperature and a 339.84Pa reduction in system voltage drop, while improving the temperature uniformity of the battery module, thus achieving the best overall performance in terms of heat dissipation and energy efficiency. Attached Figure Description

[0020] Figure 1 This is a flowchart of a multi-objective optimization method for a battery thermal management system according to an embodiment of the present invention.

[0021] Figure 2 This is a schematic diagram of the battery thermal management system structure according to an embodiment of the present invention.

[0022] Figure 3 This is a comparative schematic diagram of four rib shapes according to an embodiment of the present invention.

[0023] Figure 4 This is a graph showing the results of the mesh independence verification in an embodiment of the present invention.

[0024] Figure 5 This is a comparison and verification diagram of the temperature simulation and experiment of a single cell in an embodiment of the present invention.

[0025] Figure 6 This is a comparison chart showing the effects of different fin shapes on the battery's maximum temperature, temperature difference, voltage drop, and Nusselt number in embodiments of the present invention.

[0026] Figure 7 This is a distribution diagram of sample points in the sampling space according to an embodiment of the present invention.

[0027] Figure 8 This is a comparison chart of the predicted value and the actual value of the proxy model in an embodiment of the present invention.

[0028] Figure 9 This is a graph showing the evaluation results of the proxy model in an embodiment of the present invention.

[0029] Figure 10 This is a Pareto optimal solution set diagram according to an embodiment of the present invention.

[0030] Figure 11 This is a comparison diagram of battery temperature distribution before and after optimization in an embodiment of the present invention. Detailed Implementation

[0031] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention. Furthermore, the technical features involved in the various embodiments of this invention described below can be combined with each other as long as they do not conflict with each other.

[0032] Please see Figure 1 This embodiment provides a multi-objective optimization method for a battery thermal management system, including the following steps: S1: Construct a geometric model of the battery thermal management system, set up fins of various shapes in the liquid cooling plate flow channel for simulation comparison, and select the serpentine fin with the best overall performance to be arranged in the flow channel; S11: Construct a geometric model of the battery thermal management system based on SolidWorks software, including multiple batteries, thermal interface materials, and dual liquid cooling plates. Refer to section 2. The entire battery thermal management system comprises 12 pouch-type lithium-ion batteries, 11 thermal interface materials, and 2 liquid cooling plates. Specific parameters of the pouch-type lithium-ion batteries are shown in Table 1, including dimensions, mass, specific heat capacity, density, and thermal conductivity. The liquid cooling plates are attached to both sides of the battery module, and the thermal interface material fills the spaces between adjacent batteries with a thickness of 1 mm.

[0033] Table 1. Lithium-ion battery parameter table

[0034] S12: Four biomimetic rib structures—square, circular, fish-shaped, and snake-shaped—are respectively arranged in the liquid cooling plate flow channel. The hydraulic diameter of each rib is consistent. The specific configuration is as follows: Figure 3 As shown, (a) is a square, (b) is a circle, (c) is a fish shape, and (d) is a snake shape. S13: Select the highest battery temperature, battery module temperature difference, system voltage drop and Nusselt number as evaluation indicators for simulation comparison; To ensure the accuracy of the numerical simulation results, mesh independence verification is necessary. Multiple mesh schemes with different densities are used for simulation, and the changes in the battery's maximum temperature and system voltage drop are monitored. Figure 4 As shown, when the number of grid cells increases to 1,358,690, the rate of change of the battery's maximum temperature and the system voltage drop tends to stabilize, and the calculation results reach convergence. At this point, further increasing the number of grid cells has a negligible impact on the simulation results. Considering both computational accuracy and computational resources, the optimal grid partitioning scheme is determined. The accuracy of the model is verified by comparing it with experimental data from the literature. Figure 5 As shown, simulations were conducted under the same initial and boundary conditions. The simulation results agree well with the experimental data, and the relative error between the simulated and experimental values ​​of the single-cell temperature does not exceed 5%, proving that the current simulation model is accurate and reliable. Numerical simulation comparisons were performed using ANSYS Fluent software, with the highest battery temperature, battery module temperature difference, system voltage drop, and Nusselt number as evaluation indicators.

[0035] S14: Comprehensively evaluate the heat dissipation performance and flow resistance of each fin configuration, and select the serpentine fin with the best overall performance to arrange in the flow channel.

[0036] like Figure 6As shown, the fish-shaped fins have the highest Nusselt number of 26.858, followed by the serpentine fins at 21.448. There is no significant difference in their impact on heat dissipation and temperature uniformity. However, the system pressure drop corresponding to the fish-shaped fins reaches 851.74 Pa, significantly higher than the 803.38 Pa of the serpentine fins, leading to higher pump power consumption. Considering heat dissipation performance, flow resistance, and computational efficiency, serpentine fins are selected for placement within the liquid cooling plate channel. The serpentine fins extend in a continuous, curved shape along the coolant flow direction. Their curved profile smoothly guides the fluid flow, reducing flow separation and shape resistance, while simultaneously extending the effective fluid path and increasing the heat exchange area.

[0037] S2: Select multiple design variables and conduct parameter influence analysis using the control variable method to determine the mutual constraints between the design variables; S21: Based on the geometric model of the battery thermal management system, a numerical simulation model is established and meshed using ANSYS Fluent software, and flow and heat transfer boundary conditions, initial conditions and material property parameters are set.

[0038] S22: Select the flow channel width W, flow channel height H, fin bending radius R, coolant inlet velocity v, and coolant inlet temperature Tf as design variables.

[0039] S23: By changing the values ​​of each design variable using the control variable method, the control equations are solved to obtain temperature field and flow field data, and the influence of each design variable on the battery's maximum temperature, system voltage drop, battery module temperature difference, and Nusselt number is quantitatively analyzed. S24: Based on the analysis results, it was determined that there are interdependent relationships among the design variables. The analysis shows that the influence trends of each design variable on heat dissipation performance and flow resistance differ, and some variables exhibit interdependent relationships. Adjusting a single parameter cannot simultaneously achieve optimal performance in both heat dissipation and flow resistance; therefore, a multi-objective optimization method is needed to collaboratively optimize multiple parameters.

[0040] S3: Based on Latin hypercube sampling, sample datasets are obtained, and surrogate models for the highest battery temperature and system voltage drop are constructed respectively. S31: Within the range of five design variables—channel width, channel height, fin bending radius, coolant inlet velocity, and coolant inlet temperature—multiple sample points were obtained using Latin hypercube sampling to represent combinations of these five design variables. Specifically, the channel width ranged from 18 to 22 mm; the channel height from 1.8 to 2.2 mm; the fin bending radius from 1.8 to 2.2 mm; the coolant inlet velocity from 0.3 to 0.5 m / s; and the coolant inlet temperature from 293.15 to 303.15 K. The distribution of the sample points in the sampling space is as follows: Figure 7As shown, the sample dataset is divided into a training set and a test set in an 8:2 ratio, where the training set is used to build the surrogate model and the test set is used to evaluate the accuracy of the surrogate model.

[0041] S32: Based on the training set, candidate surrogate models for the battery's highest temperature and system voltage drop are established using radial basis function model, second-order response surface model and Kriging model, respectively.

[0042] The expression for the radial basis function surrogate model is:

[0043] In the formula, To predict the response, For basis function weights, For radial basis functions, The Euclidean distance between the input variable and the sample points. The total number of training samples used to construct the surrogate model The sample number; The expression for the second-order response surface surrogate model is:

[0044] In the formula, To predict the response, , , For the undetermined regression coefficients, For random error, As the first design variable, This is the second design variable; The expression for the Kriging proxy model is:

[0045] In the formula, To predict the response, For the global polynomial trend term, It is a random process that follows a normal distribution and is used to describe the local fluctuation characteristics of a sample.

[0046] S33: Evaluate the accuracy of each candidate agent model using the test set, with evaluation metrics including median absolute deviation. Mean absolute error and root mean square error The specific formulas for calculating the evaluation indicators are as follows:

[0047]

[0048]

[0049] In the formula, For the first The predicted response value for each sample. For the first The true response value of each sample The total number of training samples used to build the surrogate model.

[0050] S34: Select the model with the smallest prediction error based on the evaluation results: select the radial basis function model for the highest battery temperature and the second-order response surface model for the system voltage drop.

[0051] Comparison of predicted and actual values ​​by the surrogate model Figure 8 As shown, the evaluation results of the proxy model are as follows: Figure 9 As shown. The smaller the value of the evaluation index, the smaller the prediction error of the constructed surrogate model. S4: Using the battery maximum temperature proxy model and the system voltage drop proxy model as optimization objective functions, and combining the multi-objective evolutionary algorithm to solve for the Pareto optimal solution set, and selecting the optimal compromise solution from the Pareto optimal solution set; S41: Use the battery maximum temperature proxy model and the system voltage drop proxy model as the optimization objective function.

[0052] S42: Iterative optimization is performed using a multi-objective evolutionary algorithm, which includes at least one of a multi-objective particle swarm optimization algorithm, a second-generation non-dominated sorting genetic algorithm, or a multi-objective evolutionary algorithm based on dominant regions. In this embodiment, a multi-objective particle swarm optimization algorithm is used for iterative optimization.

[0053] S43: Obtain the Pareto optimal solution set through iterative optimization, such as... Figure 10 As shown.

[0054] S44: Select the optimal compromise solution from the Pareto optimal solution set, which achieves the best balance between the battery's maximum temperature and the system voltage drop.

[0055] S5: Reconstruct the geometric model based on the design parameter combination corresponding to the optimal compromise solution and perform high-precision simulation verification to determine the final structural dimensions and operating parameters.

[0056] S51: Based on the design parameter combination corresponding to the optimal compromise solution, determine the final structural dimensions and operating parameters of the battery thermal management system. The specific parameters are shown in Table 2.

[0057] Table 2

[0058] S52: Reconstruct the geometric model of the battery thermal management system based on the optimal combination of design parameters and perform high-precision simulation verification.

[0059] S53: By comparing the relative errors between the predicted values ​​of the surrogate model and the high-precision simulation values, the accuracy and feasibility of the optimization scheme are verified.

[0060] like Figure 11 As shown, the comparison of battery temperature distribution before and after optimization reveals that the high-temperature region of the battery module is significantly reduced after optimization, and the overall temperature distribution is more uniform. Compared with the initial unoptimized serpentine fin model, the highest battery temperature decreased from 308.92K to 301.94K, a reduction of 6.98K; the system voltage drop decreased from 803.38Pa to 463.54Pa, a reduction of 339.84Pa. By comparing the predicted values ​​of the surrogate model with the high-precision simulation values, the relative errors are only 0.05% and 4.49%, respectively, demonstrating the high accuracy and feasibility of the surrogate model optimization framework.

[0061] This embodiment also provides a battery thermal management system, designed based on the aforementioned multi-objective optimization method for battery thermal management systems. The battery thermal management system includes multiple batteries, a thermal interface material, and dual liquid cooling plates. The liquid cooling plates have serpentine fins arranged within their flow channels, extending in a continuously curved manner along the coolant flow direction. The structural dimensions and operating parameters of the battery thermal management system are determined according to the aforementioned optimization method, enabling it to simultaneously achieve excellent heat dissipation performance and low flow resistance.

[0062] Those skilled in the art will readily understand that the above description is merely 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 within the scope of protection of the present invention.

Claims

1. A multi-objective optimization method for a battery thermal management system, characterized in that, Includes the following steps: S1: Construct a geometric model of the battery thermal management system, set up fins of various shapes in the liquid cooling plate flow channel for simulation comparison, and select the serpentine fin with the best overall performance to be arranged in the flow channel; S2: Select multiple design variables and conduct parameter influence analysis using the control variable method to determine the mutual constraints between the design variables; S3: Based on Latin hypercube sampling, sample datasets are obtained, and surrogate models for the highest battery temperature and system voltage drop are constructed respectively. S4: Using the battery maximum temperature proxy model and the system voltage drop proxy model as optimization objective functions, and combining the multi-objective evolutionary algorithm to solve for the Pareto optimal solution set, and selecting the optimal compromise solution from the Pareto optimal solution set; S5: Reconstruct the geometric model based on the design parameter combination corresponding to the optimal compromise solution and perform high-precision simulation verification to determine the final structural dimensions and operating parameters.

2. The multi-objective optimization method for the battery thermal management system as described in claim 1, characterized in that, Step S1 specifically includes: S11: Construct a geometric model of a battery thermal management system that includes multiple batteries, thermal interface materials, and dual liquid cooling plates; S12: Four types of rib structures, namely square ribs, circular ribs, fish-shaped ribs and snake-shaped ribs, are respectively set in the liquid cooling plate flow channel, and the hydraulic diameter of each rib is consistent. S13: Select the highest battery temperature, battery module temperature difference, system voltage drop and Nusselt number as evaluation indicators for simulation comparison; S14: Comprehensively evaluate the heat dissipation performance and flow resistance of each fin configuration, and select the serpentine fin with the best overall performance to arrange in the flow channel.

3. The multi-objective optimization method for the battery thermal management system as described in claim 1, characterized in that, Step S2 specifically includes: S21: Establish a numerical simulation model based on the geometric model of the battery thermal management system, and set the flow and heat transfer boundary conditions, initial conditions and material property parameters; S22: Select the flow channel width, flow channel height, fin bending radius, coolant inlet velocity, and coolant inlet temperature as design variables; S23: By changing the values ​​of each design variable using the control variable method, the control equations are solved to obtain temperature field and flow field data, and the influence of each design variable on the battery's maximum temperature, system voltage drop, battery module temperature difference, and Nusselt number is quantitatively analyzed. S24: Based on the analysis results, it is determined that there are mutual constraints among the design variables, and the multi-objective optimization method is used to perform collaborative optimization of multiple parameters.

4. The multi-objective optimization method for the battery thermal management system as described in claim 1, characterized in that, Step S3 specifically includes the following sub-steps: S31: Within the range of five design variables—channel width, channel height, fin bending radius, coolant inlet velocity, and coolant inlet temperature—multiple sample points are obtained through Latin hypercube sampling. S32: Based on the sample dataset, candidate surrogate models for the battery's highest temperature and system voltage drop are established using radial basis function model, second-order response surface model, and Kriging model, respectively. S33: The accuracy of each candidate surrogate model is evaluated using the test set. The evaluation metrics include median absolute deviation, mean absolute error, and root mean square error. S34: Select the model with the smallest prediction error based on the evaluation results: select the radial basis function model for the highest battery temperature and the second-order response surface model for the system voltage drop.

5. The multi-objective optimization method for the battery thermal management system as described in claim 4, characterized in that, The specific ranges for the design variables are as follows: the flow channel width is 18–22 mm; the flow channel height is 1.8–2.2 mm; the rib bending radius is 1.8–2.2 mm; the coolant inlet velocity is 0.3–0.5 m / s; and the coolant inlet temperature is 293.15–303.15 K.

6. The multi-objective optimization method for the battery thermal management system as described in claim 4, characterized in that, The expression for the radial basis function surrogate model is: In the formula, To predict the response, For basis function weights, For radial basis functions, The Euclidean distance between the input variable and the sample points. The total number of training samples used to construct the surrogate model The sample number; The expression for the second-order response surface surrogate model is: In the formula, To predict the response, , , For the undetermined regression coefficients, For random error, As the first design variable, This is the second design variable.

7. The multi-objective optimization method for a battery thermal management system as described in claim 1, characterized in that, Step S4 specifically includes the following sub-steps: S41: Use the battery maximum temperature proxy model and the system voltage drop proxy model as the optimization objective functions; S42: Iterative optimization is performed by combining a multi-objective evolutionary algorithm, wherein the multi-objective evolutionary algorithm includes at least one of the multi-objective particle swarm optimization algorithm, the second-generation non-dominated sorting genetic algorithm, or a multi-objective evolutionary algorithm based on the dominant region. S43: Obtain the Pareto optimal solution set through iterative calculation; S44: Select the optimal compromise solution from the Pareto optimal solution set, which achieves a balance between the battery's maximum temperature and the system voltage drop.

8. The multi-objective optimization method for a battery thermal management system as described in claim 1, characterized in that, Step S5 specifically includes the following sub-steps: S51: Based on the design parameter combination corresponding to the optimal compromise solution, determine the final structural dimensions and operating parameters of the battery thermal management system; S52: Reconstruct the geometric model of the battery thermal management system based on the optimal design parameter combination and perform high-precision simulation verification; S53: By comparing the relative errors between the predicted values ​​of the surrogate model and the high-precision simulation values, the accuracy and feasibility of the optimization scheme are verified.

9. The multi-objective optimization method for a battery thermal management system as described in any one of claims 1 to 8, characterized in that: The serpentine fins extend in a continuous curved shape along the flow direction of the coolant. Their curved profile is used to smoothly guide the fluid flow around the fins, thereby reducing flow separation and shape resistance, while extending the effective flow path of the fluid and increasing the heat exchange area.

10. A battery thermal management system, characterized in that, The battery thermal management system is designed based on the multi-objective optimization method of any one of claims 1 to 9.

Images