Intelligent liquid cooling plate optimization method and apparatus based on CFD numerical simulation

Abstract

Disclosed in the present invention are an intelligent liquid cooling plate optimization method and apparatus based on CFD numerical simulation. The method comprises: first, establishing a 3D battery pack model on the basis of liquid cooling plate design parameters and other component design parameters; then, on the basis of the 3D battery pack model, performing thermal simulation modeling and thermal simulation calculation on a battery pack; on the basis of a thermal simulation result, using a particle swarm multi-objective optimization algorithm to optimally calculate locally optimal solutions of the liquid cooling plate design parameters, and determining whether the locally optimal solutions converge; and using a loop iteration mode to calculate the locally optimal solutions of the liquid cooling plate design parameters until the locally optimal solutions converge, and then obtaining globally optimal solutions. Three-dimensional CAD software, CFD simulation software and an optimization interface are interconnected to each other, so that the operation of the whole process of design, simulation and optimization of a liquid cooling plate can be completely and intelligently realized simply by means of inputting key design parameters of the liquid cooling plate in an initial design stage; and an optimal design scheme is obtained by means of implanting the particle swarm multi-objective optimization algorithm. The present invention achieves a high speed and high efficiency.

Description

A liquid cooling plate intelligent optimization method and device based on CFD numerical simulation Technical Field

[0001] The present invention relates to a liquid cooling plate intelligent optimization method and device based on CFD numerical simulation, belonging to battery thermal management technology for electric vehicles. Background Art

[0002] With the rapid development of the global economy, fossil fuel reserves are decreasing year by year. At the same time, the use of fossil fuels also causes serious environmental pollution. Consequently, electric vehicles have emerged as a response to this need. In recent years, the rapid development of electric vehicles has led to increasing attention to the safety and lifespan of electric vehicle power batteries. One of the most significant factors affecting power battery safety and lifespan is the operating temperature. Excessive internal temperature in the battery pack can lead to safety issues such as thermal runaway. Excessive temperature differences between different cells within the battery pack can degrade battery performance. Long-term operation at excessively high temperatures and temperature differences can also shorten battery life. Therefore, proper battery thermal management is a crucial component of power battery design. Battery thermal management involves heating the battery when it is cold, cooling it when it is hot, and thermally balancing all cells within the battery pack to maintain a temperature difference of less than 5°C. Currently, the most common battery thermal management methods are liquid cooling, air cooling, and direct cooling. Liquid cooling is more efficient and reliable than these two methods, making it the most widely used method.

[0003] The liquid cooling plate is an important component of the liquid-cooled battery thermal management system. Currently, the design of the liquid cooling plate mostly relies on the experience of engineers. When optimizing the design of the battery liquid cooling plate, it is necessary to optimize the pressure drop of the liquid cooling plate and the flow distribution of each parallel tributary. Due to the large number of parallel tributaries, multiple optimization designs and CFD (computational fluid dynamics) simulation iterations are required when optimizing the flow distribution of each tributary. Each manual optimization design and CFD simulation takes a lot of time, which is inefficient and time-costly, and it is difficult to find the optimal solution through manual trial and error. There are also automatic optimization methods for liquid cooling plates. For example, the Chinese invention patent with the patent number "CN202110064676.7" and the name "A liquid cooling plate multi-objective optimization method, electronic device and storage medium" realizes the automatic optimization of the liquid cooling plate by simply connecting the liquid cooling plate 3D modeling software, simulation software, and optimization software in series. However, this method only optimizes by setting a preset target. After the optimization result reaches the preset target, the optimization iteration is stopped, and the optimal solution is not optimized. Summary of the Invention

[0004] Purpose of the invention: In order to overcome the deficiencies in the prior art, the present invention provides a method and device for intelligent optimization of liquid cooling plates based on CFD numerical simulation. By simply inputting the key design parameters of the liquid cooling plate in the initial design stage, the entire process of liquid cooling plate design, simulation, and optimization can be fully intelligently implemented to obtain the optimal design solution, greatly reducing the labor and time costs of liquid cooling plate design and simulation.

[0005] Technical solution: To achieve the above purpose, the technical solution adopted by the present invention is:

[0006] An intelligent optimization method for liquid cooling plates based on CFD numerical simulation, wherein a battery pack includes a liquid cooling plate and other components other than the liquid cooling plate; N sets of liquid cooling plate design parameters are used to perform parametric 3D modeling on the liquid cooling plate to generate N liquid cooling plate 3D models; other components are used to perform 3D modeling on other components based on their design parameters to generate a set of other component 3D models; each liquid cooling plate 3D model and other component 3D models are combined to form N battery pack 3D models; thermal simulation modeling and thermal simulation calculation are performed on the battery pack based on the battery pack 3D model, and thermal simulation results including voltage drop, cell temperature difference, and cell maximum temperature are output; based on the thermal simulation results, granular simulation is used to calculate the thermal simulation results. The subgroup multi-objective optimization algorithm optimizes and calculates the local optimal solution of the liquid cooling plate design parameters, and determines whether the local optimal solution converges. If converged, the global optimal solution of the liquid cooling plate design parameters is calculated based on the local optimal solution. Otherwise, the optimization direction of the liquid cooling plate design parameters is formed based on the local optimal solution to guide the optimization of N groups of liquid cooling plate design parameters. The N groups of liquid cooling plate design parameters are updated, and parametric 3D modeling of the liquid cooling plate is performed based on the updated N groups of liquid cooling plate design parameters to generate N liquid cooling plate 3D models. The local optimal solution of the liquid cooling plate design parameters is calculated in a cyclic iterative manner until the local optimal solution converges, and then the global optimal solution is obtained.

[0007] Preferably, a k-epsilon turbulence model is used to perform thermal simulation modeling on the battery pack 3D model.

[0008] Preferably, CATIA software is used to perform 3D modeling on the liquid cooling plate and other components, wherein the 3D modeling of the liquid cooling plate adopts parametric modeling, and the 3D modeling of other components adopts 3D drawing modeling.

[0009] Preferably, 3D modeling is performed through a 3D modeling module, and thermal simulation modeling is performed through a CFD simulation module. A data interface is established between the 3D modeling module and the CFD simulation module. The local optimal solution calculated by the CFD simulation module is sent to the 3D modeling module through the data interface. The 3D modeling module generates N liquid cooling plate 3D models through the parameterized 3D modeling and sends them to the CFD simulation module through the data interface.

[0010] Specifically, the method includes the following steps:

[0011] S1. Perform 3D modeling through the 3D modeling module and thermal simulation modeling through the CFD simulation module. Establish a data optimization interface between the 3D modeling module and the CFD simulation module. Send the local optimal solution calculated by the CFD simulation module to the 3D modeling module through the data optimization interface. The 3D modeling module generates N 3D models of liquid cooling plates based on the parameterized 3D modeling and sends them to the CFD simulation module through the data optimization interface.

[0012] S2. Dividing the value ranges of the liquid cooling plate design parameters and initializing the liquid cooling plate design parameters; the liquid cooling plate design parameters include external contour parameters and internal flow channel domain parameters, and the internal flow channel domain parameters include the number of parallel flow channels, flow channel thickness, branch flow channel width, and main flow channel width;

[0013] S3. Perform 3D modeling based on the design parameters of other components to generate a set of 3D models of other components;

[0014] S4. Perform parametric 3D modeling based on N sets of liquid cooling plate design parameters to generate N liquid cooling plate 3D models, where all N liquid cooling plate 3D models have the same external contour parameters but different internal flow channel domain parameters;

[0015] S5. Perform thermal simulation modeling on the battery pack and record a macro file, including the following steps:

[0016] S51, turn on the macro file recording button;

[0017] S52. Import N 3D models of liquid cooling plates and other component 3D models, and combine them with boundary constraints of the battery pack 3D model to form N 3D battery pack models.

[0018] S53. Based on the external shape, size, and position of the liquid cooling plate 3D model and the 3D models of other components in the battery pack 3D model, stamp each physical component of the battery pack to obtain a contact relationship between the physical components of the battery pack;

[0019] S54, extracting the air domain of the battery pack and the flow channel domain of each liquid cooling plate, and dividing the inlet and outlet boundaries of the flow channel domain of each liquid cooling plate;

[0020] S55. Setting grid parameters and reconstructing the grids of the physical components of the battery pack to generate a volume grid;

[0021] S56. Setting property parameters including fluid and solid material parameters, fluid inlet and outlet boundaries, fluid inlet temperature, battery initial temperature, and battery cell heating value;

[0022] S57. Select the k-epsilon turbulence model to perform thermal simulation on N 3D battery pack models, set the inlet and outlet pressure drop constraints of the liquid cooling plate, set the liquid cooling plate design parameters as solution parameters, select the time step, number of iterations, and solution time, and use the particle swarm multi-objective optimization algorithm to calculate the local optimal solution of the liquid cooling plate design parameters;

[0023] S58. Determine whether the local optimal solution of the liquid cooling plate design parameters converges. If so, calculate the global optimal solution of the liquid cooling plate design parameters based on the local optimal solution. Otherwise, form an optimization direction for the liquid cooling plate design parameters based on the local optimal solution and guide the optimization of N groups of liquid cooling plate design parameters. Update the N groups of liquid cooling plate design parameters, return to step S4, and calculate the local optimal solution of the liquid cooling plate design parameters in a loop iterative manner until the local optimal solution converges, thereby obtaining the global optimal solution.

[0024] A liquid cooling plate intelligent optimization device based on CFD numerical simulation includes a 3D modeling module, a CFD simulation module and a data optimization interface; the 3D modeling module performs parameterized 3D modeling on the liquid cooling plate based on the liquid cooling plate design parameters to generate a liquid cooling plate 3D model, and performs 3D modeling on other components based on the design parameters of other components to generate a set of 3D models of other components; the CFD simulation module combines the liquid cooling plate 3D model and other component 3D models into a battery pack 3D model, performs thermal simulation modeling and thermal simulation calculations on the battery pack 3D model, outputs thermal simulation results including voltage drop, battery cell temperature difference, and battery cell maximum temperature, optimizes and calculates local optimal solutions of the liquid cooling plate design parameters using a particle swarm multi-objective optimization algorithm based on the thermal simulation results, and determines whether the local optimal solutions converge; the data optimization interface is used to establish a data connection between the 3D modeling module and the CFD simulation module, and the local optimal solutions calculated by the CFD simulation module are sent to the 3D modeling module through the data optimization interface. The 3D modeling module generates N liquid cooling plate 3D models through the parameterized 3D modeling and sends them to the CFD simulation module through the data optimization interface.

[0025] Preferably, the liquid cooling plate intelligent optimization device calculates the local optimal solution of the liquid cooling plate design parameters in a cyclic iterative manner until the local optimal solution converges, thereby obtaining the global optimal solution.

[0026] Preferably, the data optimization interface forms an optimization direction of the liquid cooling plate design parameters based on the local optimal solution and guides the optimization of the liquid cooling plate design parameters, updates N groups of liquid cooling plate design parameters, and sends the updated liquid cooling plate design parameters to the 3D modeling module.

[0027] Beneficial effects: The intelligent optimization method and device for liquid cooling plates based on CFD numerical simulation provided by the present invention, through the interconnection between three-dimensional CAD software, CFD simulation software, and optimization interface, only needs to input the key design parameters of the liquid cooling plate in the initial design stage, so as to realize the whole process of liquid cooling plate design, simulation, and optimization in a fully intelligent manner. By implanting the particle swarm multi-objective optimization algorithm, the optimal design scheme is obtained with high speed and efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] FIG1 is a schematic diagram of the implementation process of the method of the present invention. DETAILED DESCRIPTION

[0029] The present invention will be described in detail below with reference to the accompanying drawings and specific embodiments.

[0030] As shown in Figure 1, a liquid cooling plate intelligent optimization method based on CFD numerical simulation is shown. The battery pack includes a liquid cooling plate and other components other than the liquid cooling plate. Based on N sets of liquid cooling plate design parameters, the liquid cooling plate is parametrically 3D modeled to generate N liquid cooling plate 3D models. Based on the design parameters of other components, other components are 3D modeled to generate a set of other component 3D models. Each liquid cooling plate 3D model and other component 3D models are combined to form N battery pack 3D models. Based on the battery pack 3D model, thermal simulation modeling and thermal simulation calculation are performed on the battery pack, and thermal simulation results including voltage drop, cell temperature difference, and cell maximum temperature are output. Based on the thermal simulation results A particle swarm multi-objective optimization algorithm is used to optimize and calculate the local optimal solution of the liquid cooling plate design parameters, and determine whether the local optimal solution converges. If converged, the global optimal solution of the liquid cooling plate design parameters is calculated based on the local optimal solution. Otherwise, the optimization direction of the liquid cooling plate design parameters is formed based on the local optimal solution and the optimization of N groups of liquid cooling plate design parameters is guided. The N groups of liquid cooling plate design parameters are updated, and the liquid cooling plate is parametrically modeled based on the updated N groups of liquid cooling plate design parameters to generate N liquid cooling plate 3D models. The local optimal solution of the liquid cooling plate design parameters is calculated in a cyclic iterative manner until the local optimal solution converges, thereby obtaining the global optimal solution. In this case, the method specifically includes the following steps:

[0031] S1. Perform 3D modeling through the 3D modeling module and thermal simulation modeling through the CFD simulation module. Establish a data optimization interface between the 3D modeling module and the CFD simulation module. Send the local optimal solution calculated by the CFD simulation module to the 3D modeling module through the data optimization interface. The 3D modeling module generates N 3D models of liquid cooling plates based on the parameterized 3D modeling and sends them to the CFD simulation module through the data optimization interface.

[0032] S2. Dividing the value ranges of the liquid cooling plate design parameters and initializing the liquid cooling plate design parameters; the liquid cooling plate design parameters include external contour parameters and internal flow channel domain parameters, and the internal flow channel domain parameters include the number of parallel flow channels, flow channel thickness, branch flow channel width, and main flow channel width;

[0033] S3, based on the design parameters of other components, 3D modeling is performed using a 3D drawing method to generate a set of 3D models of other components;

[0034] S4. Perform parametric 3D modeling based on N sets of liquid cooling plate design parameters to generate N liquid cooling plate 3D models, where all N liquid cooling plate 3D models have the same external contour parameters but different internal flow channel domain parameters;

[0035] S5. Perform thermal simulation modeling on the battery pack and record a macro file, including the following steps:

[0036] S51, turn on the macro file recording button;

[0037] S52. Import N 3D models of liquid cooling plates and other component 3D models, and combine them with boundary constraints of the battery pack 3D model to form N 3D battery pack models.

[0038] S53. Based on the external shape, size, and position of the liquid cooling plate 3D model and the 3D models of other components in the battery pack 3D model, stamp each physical component of the battery pack to obtain a contact relationship between the physical components of the battery pack;

[0039] S54, extracting the air domain of the battery pack and the flow channel domain of each liquid cooling plate, and dividing the inlet and outlet boundaries of the flow channel domain of each liquid cooling plate;

[0040] S55. Setting grid parameters and reconstructing the grids of the physical components of the battery pack to generate a volume grid;

[0041] S56. Setting property parameters including fluid and solid material parameters, fluid inlet and outlet boundaries, fluid inlet temperature, battery initial temperature, and battery cell heating value;

[0042] S57. Select the k-epsilon turbulence model to perform thermal simulation on N 3D battery pack models, set the inlet and outlet pressure drop constraints of the liquid cooling plate, set the liquid cooling plate design parameters as solution parameters, select the time step, number of iterations, and solution time, and use the particle swarm multi-objective optimization algorithm to calculate the local optimal solution of the liquid cooling plate design parameters;

[0043] S58. Determine whether the local optimal solution of the liquid cooling plate design parameters converges. If so, calculate the global optimal solution of the liquid cooling plate design parameters based on the local optimal solution. Otherwise, form an optimization direction for the liquid cooling plate design parameters based on the local optimal solution and guide the optimization of N groups of liquid cooling plate design parameters. Update the N groups of liquid cooling plate design parameters, return to step S4, and calculate the local optimal solution of the liquid cooling plate design parameters in a loop iterative manner until the local optimal solution converges, thereby obtaining the global optimal solution.

[0044] In this case, the liquid cooling plate intelligent optimization device based on CFD numerical simulation for realizing the above-mentioned method includes a 3D modeling module, a CFD simulation module and a data optimization interface; the 3D modeling module performs parameterized 3D modeling of the liquid cooling plate based on the design parameters of the liquid cooling plate to generate a 3D model of the liquid cooling plate, and performs 3D modeling of other components based on the design parameters of other components to generate a set of 3D models of other components; the CFD simulation module combines the 3D model of the liquid cooling plate and the 3D models of other components into a 3D model of the battery pack, performs thermal simulation modeling and thermal simulation calculation on the 3D model of the battery pack, and outputs thermal simulation results including voltage drop, cell temperature difference, and maximum cell temperature. Based on the thermal simulation results, a particle swarm multi-objective optimization algorithm is used to optimize and calculate the local optimal solution of the liquid cooling plate design parameters, and whether the local optimal solution converges is determined; the data optimization interface is used to establish a data connection between the 3D modeling module and the CFD simulation module, and the local optimal solution calculated by the CFD simulation module is sent to the 3D modeling module through the data optimization interface. The 3D modeling module generates N liquid cooling plate 3D models by parameterized 3D modeling and sends them to the CFD simulation module through the data optimization interface; the liquid cooling plate intelligent optimization device uses a cyclic iterative method to calculate the local optimal solution of the liquid cooling plate design parameters until the local optimal solution converges, and then obtains the global optimal solution.

[0045] All data optimization interfaces form optimization directions for the liquid cooling plate design parameters based on local optimal solutions and guide the optimization of the liquid cooling plate design parameters, update N sets of liquid cooling plate design parameters, and send the updated liquid cooling plate design parameters to the 3D modeling module. The 3D modeling module is designed based on CATIA software.

[0046] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the above embodiments do not limit the present invention in any form, and any technical solutions obtained by equivalent replacement or equivalent transformation fall within the scope of protection of the present invention.

Claims

1. An intelligent optimization method for liquid cooling plates based on CFD numerical simulation, where the battery pack includes a liquid cooling plate and other components other than the liquid cooling plate; characterized in that: Based on N sets of liquid cooling plate design parameters, parametric 3D modeling of the liquid cooling plate is carried out to generate N 3D models of the liquid cooling plate. Based on the design parameters of other components, 3D modeling of other components is carried out to generate a set of 3D models of other components. Each 3D model of the liquid cooling plate and the 3D models of other components are combined to form N 3D models of the battery pack. Based on the 3D models of the battery pack, thermal simulation modeling and thermal simulation calculation of the battery pack are carried out, and thermal simulation results including pressure drop, cell temperature difference, and maximum cell temperature are output. Based on the thermal simulation results, the particle swarm multi-objective optimization algorithm is used to optimize and calculate the local optimal solution of the liquid cooling plate design parameters, and it is judged whether the local optimal solution converges: if it converges, the global optimal solution of the liquid cooling plate design parameters is calculated according to the local optimal solution; otherwise, based on the local optimal solution, the optimization direction of the liquid cooling plate design parameters is formed and used to guide the optimization of N sets of liquid cooling plate design parameters. The N sets of liquid cooling plate design parameters are updated, and based on the updated N sets of liquid cooling plate design parameters, parametric 3D modeling of the liquid cooling plate is carried out to generate N 3D models of the liquid cooling plate. The local optimal solution of the liquid cooling plate design parameters is calculated in a cyclic iterative manner until the local optimal solution converges, and then the global optimal solution is obtained.

2. The intelligent optimization method of the liquid cooling plate based on CFD numerical simulation according to claim 1, wherein: The k-epsilon turbulence model is used for thermal simulation modeling of the 3D model of the battery pack.

3. The intelligent optimization method of the liquid cooling plate based on CFD numerical simulation according to claim 1, characterized in that: The CATIA software is used for 3D modeling of the liquid cooling plate and other components. Among them, the 3D modeling method of the liquid cooling plate is parametric modeling, and the 3D modeling of other components is carried out by 3D drawing.

4. The intelligent optimization method of the liquid cooling plate based on CFD numerical simulation according to claim 1, wherein: 3D modeling is carried out through the 3D modeling module, and thermal simulation modeling is carried out through the CFD simulation module. A data interface is established between the 3D modeling module and the CFD simulation module. The local optimal solution calculated by the CFD simulation module is sent to the 3D modeling module through this data interface, and the N 3D models of the liquid cooling plate generated by parametric 3D modeling by the 3D modeling module are sent to the CFD simulation module through this data interface.

5. The intelligent optimization method of the liquid cooling plate based on CFD numerical simulation according to claim 1, wherein: It includes the following steps: S1. 3D modeling is carried out through the 3D modeling module, and thermal simulation modeling is carried out through the CFD simulation module. A data optimization interface is established between the 3D modeling module and the CFD simulation module. The local optimal solution calculated by the CFD simulation module is sent to the 3D modeling module through the data optimization interface, and the N 3D models of the liquid cooling plate generated by parametric 3D modeling by the 3D modeling module are sent to the CFD simulation module through this data optimization interface. S2. The value range of the liquid cooling plate design parameters is divided, and the liquid cooling plate design parameters are initialized; the liquid cooling plate design parameters include external contour parameters and internal flow channel domain parameters, and the internal flow channel domain parameters include the number of parallel flow channels, flow channel thickness, branch flow channel width, and main flow channel width. S3. Based on the design parameters of other components, 3D modeling is carried out to generate a set of 3D models of other components. S4. Based on N sets of liquid cooling plate design parameters, parametric 3D modeling of the liquid cooling plate is carried out to generate N 3D models of the liquid cooling plate. The external contour parameters of all N 3D models of the liquid cooling plate are the same, and the internal flow channel domain parameters are different. S5. Thermal simulation modeling of the battery pack is carried out and a macro file is recorded, including the following steps: S51. Turn on the macro file recording button. S52. Import the 3D models of N liquid cooling plates and the 3D models of other components, and combine the boundary constraint conditions of the battery pack 3D model to form N battery pack 3D models; S53. Based on the external shapes, dimensions and positions of the liquid cooling plate 3D model and other component 3D models in the battery pack 3D model, imprint each physical component of the battery pack to obtain the contact relationships of each physical component of the battery pack; S54. Extract the air domain of the battery pack and the flow channel domain of each liquid cooling plate, and divide the inlet and outlet boundaries of the flow channel domain of each liquid cooling plate; S55. Set the mesh parameters, and reconstruct the mesh of each physical component of the battery pack to generate volume meshes; S56. Set the property parameters including fluid and solid material parameters, fluid inlet and outlet boundaries, fluid inlet temperature, initial battery temperature, and heat generation of the battery cells; S57. Select the k-epsilon turbulence model to perform thermal simulation modeling on the N battery pack 3D models, set the inlet and outlet pressure drop constraint conditions of the liquid cooling plate, set the liquid cooling plate design parameters as the solution parameters, select the time step, the number of iteration steps and the solution time, and use the particle swarm multi-objective optimization algorithm to calculate the local optimal solution of the liquid cooling plate design parameters; S58. Judge whether the local optimal solution of the liquid cooling plate design parameters converges: If it converges, calculate the global optimal solution of the liquid cooling plate design parameters according to the local optimal solution; Otherwise, form the optimization direction of the liquid cooling plate design parameters based on the local optimal solution and guide the optimization of N groups of liquid cooling plate design parameters, update N groups of liquid cooling plate design parameters, return to step S4, and calculate the local optimal solution of the liquid cooling plate design parameters in a cyclic iteration manner until the local optimal solution converges, and then obtain the global optimal solution.

6. An intelligent optimization device for a liquid cooling plate based on CFD numerical simulation, characterized in that: It includes a 3D modeling module, a CFD simulation module and a data optimization interface; the 3D modeling module performs parametric 3D modeling on the liquid cooling plate based on the liquid cooling plate design parameters to generate a liquid cooling plate 3D model, and performs 3D modeling on other components based on other component design parameters to generate a set of other component 3D models; the CFD simulation module combines the liquid cooling plate 3D model and other component 3D models into a battery pack 3D model, performs thermal simulation modeling and thermal simulation calculation on the battery pack 3D model, outputs thermal simulation results including pressure drop, temperature difference between battery cells, and maximum temperature of battery cells, optimizes and calculates the local optimal solution of the liquid cooling plate design parameters using the particle swarm multi-objective optimization algorithm based on the thermal simulation results, and judges whether the local optimal solution converges; the data optimization interface is used to establish a data connection between the 3D modeling module and the CFD simulation module, and the local optimal solution calculated by the CFD simulation module is sent to the 3D modeling module through this data optimization interface, and the 3D modeling module sends the N liquid cooling plate 3D models generated by parametric 3D modeling to the CFD simulation module through this data optimization interface.

7. The intelligent optimization device for the liquid cooling plate based on CFD numerical simulation according to claim 6, characterized in that: This liquid cooling plate intelligent optimization device calculates the local optimal solution of the liquid cooling plate design parameters in a cyclic iteration manner until the local optimal solution converges, and then obtains the global optimal solution.

8. The intelligent optimization device for liquid cooling plates based on CFD numerical simulation according to claim 6, characterized in that: The data optimization interface forms the optimization direction of the liquid cooling plate design parameters based on the local optimal solution and guides the optimization of the liquid cooling plate design parameters, updates N groups of liquid cooling plate design parameters, and sends the updated liquid cooling plate design parameters to the 3D modeling module.

Images