Symmetrically arranged topological multi-channel single battery liquid cooling plate and topological channel optimization method

Abstract

The invention discloses a symmetrically-arranged topological multi-runner single battery liquid cooling plate and a topological runner optimization method, the liquid cooling plate comprises a substrate, a topological runner is formed on the substrate, the topological runner is symmetrical about a center line in the width direction, and the topological runner specifically comprises a liquid inlet straight runner, a liquid outlet straight runner, a liquid outlet straight runner and a liquid outlet straight runner; the liquid inlet straight flow channel is located on the center line of the liquid cooling plate in the width direction and used for liquid inlet. The liquid outlet straight flow channel is located on the center line of the liquid cooling plate in the width direction and used for discharging liquid; and the sub flow channel is located between the liquid inlet straight flow channel and the liquid outlet straight flow channel and used for conveying the cooling liquid flowing into the liquid inlet straight flow channel to the liquid outlet straight flow channel. According to the invention, through the ingenious layout of the single-inlet single-outlet straight flow channel and the sub-flow channel, the symmetrically-arranged topological multi-flow-channel single battery liquid cooling plate is designed; according to the liquid cooling plate, parameters are further optimized according to a topological flow channel optimization method to realize topological flow channel optimization, and compared with a straight flow channel liquid cooling plate and a snake-shaped flow channel liquid cooling plate, the liquid cooling plate has relatively good performance in the aspects of cooling and power consumption reduction, and is a liquid cooling plate structure with relatively good performance.

Description

Technical Field

[0001] This invention relates to a symmetrically arranged topological multi-channel single-cell liquid cooling plate and a topological channel optimization method, belonging to the technical field of heat dissipation equipment for lithium batteries. Background Technology

[0002] With the widespread application of electric vehicles and energy storage systems, the energy density and charge / discharge efficiency of lithium batteries are constantly improving, making thermal management an increasingly important issue. Lithium batteries release a significant amount of heat during charging and discharging, especially during high-power, high-rate discharge. Increased battery temperature can severely impact performance and safety. Excessive temperature not only accelerates battery capacity decay but can also trigger thermal runaway, potentially leading to fires or explosions in severe cases. Therefore, efficiently dissipating heat from the battery system and maintaining its operating temperature within a safe range is a key focus of current battery technology research.

[0003] Currently, there are two main common cooling systems for lithium batteries: natural cooling and liquid cooling. Natural cooling relies on air to dissipate heat from the battery and is suitable for low-power applications, but its cooling effect is limited and cannot meet the needs of high-power battery packs. Liquid cooling systems, on the other hand, cool the battery with liquid, providing more efficient heat dissipation and are therefore widely used in electric vehicles, energy storage systems, and other fields. In liquid cooling systems, the coolant flows through the channels of the cooling plate, carrying away the heat generated by the battery. The design of the cooling plate, especially the layout of the channels and the path of the coolant flow, is crucial to the thermal management effect.

[0004] Currently, the flow channel design of liquid cooling plates mainly relies on existing standardized designs, but this design method often has the following shortcomings: (1) Unreasonable flow channel layout: Traditional liquid cooling plate flow channel designs often adopt simple straight or uniformly distributed flow channel layouts, failing to fully consider the flow path of the coolant and the temperature distribution of the battery, resulting in uneven heat conduction in some areas and low cooling efficiency. (2) Uneven heat conduction: Due to the insufficient refinement of the flow channel design, the coolant fails to uniformly carry away the heat of the entire battery surface when flowing through the flow channel, resulting in higher temperatures in some areas, thereby affecting the working stability and service life of the battery. (3) Low fluid dynamic efficiency: Many existing cooling plate flow channel designs do not fully consider the optimization of fluid dynamics, resulting in large flow channel resistance and pressure loss, reducing the efficiency of coolant flow, and thus affecting the performance of the entire cooling system. (4) Complex manufacturing process: Some advanced cooling plate flow channel designs require complex manufacturing processes, increasing production costs and difficulty, making it difficult to achieve large-scale production and application.

[0005] Therefore, designing an optimized cooling plate channel to improve heat transfer efficiency, reduce flow resistance, and simplify manufacturing processes is a pressing issue in current liquid cooling systems. Topology optimization can find an optimal material distribution or structural layout within a given design space to maximize cooling performance. However, existing cooling plate channel designs often have limitations, such as inflexible channel layout, uneven cooling, and excessive flow resistance. Therefore, a new channel design method is urgently needed to optimize cooling plate channels and achieve better thermal management. Summary of the Invention

[0006] This invention provides a symmetrically arranged topological multi-channel liquid cooling plate for a single-cell battery, which is cleverly designed with a single-inlet, single-outlet flow channel and sub-flow channels. Furthermore, it provides a topology flow channel optimization method for the symmetrically arranged topological multi-channel liquid cooling plate for a single-cell battery, which optimizes the parameters to achieve topology flow channel optimization and obtain an optimized topological multi-channel liquid cooling plate for a single-cell battery.

[0007] The technical solution of this invention is:

[0008] According to a first aspect of the present invention, a symmetrically arranged topological multi-channel single-cell battery liquid cooling plate is provided, comprising a substrate on which topological channels are formed, the topological channels being symmetrical about a centerline in the width direction, specifically comprising: an inlet direct channel located on the centerline of the liquid cooling plate in the width direction for liquid inlet; an outlet direct channel located on the centerline of the liquid cooling plate in the width direction for liquid outlet; and a sub-channel located between the inlet direct channel and the outlet direct channel for conveying the coolant flowing into the inlet direct channel to the outlet direct channel.

[0009] Furthermore, the sub-channel includes a first sub-channel and a second sub-channel; the first sub-channel is distributed on the outer periphery, and the second sub-channel is located in the area enclosed by the first sub-channel; the coolant flows from the inlet into the inlet direct channel, and the coolant flowing into the inlet direct channel converges through the first sub-channel and the second sub-channel to the outlet direct channel, and is discharged through the outlet of the outlet direct channel.

[0010] Furthermore, the volume fraction of the topological flow channel region in the liquid cooling plate is 30%-70%.

[0011] According to a second aspect of the present invention, a method for optimizing the topology flow channels of a liquid cooling plate for a symmetrically arranged multi-channel single-cell battery as described in any one of the above-mentioned methods is provided, comprising:

[0012] S1. Construct a single-cell heat generation model based on the Bernardi heat generation model;

[0013] S2. Based on the contact surface between the battery and the liquid cooling plate, the dimensions of the liquid cooling plate substrate are designed using finite element analysis software, and a two-dimensional model is constructed by selecting a design domain on the substrate.

[0014] S3. Given the assumptions;

[0015] S4. Based on the assumptions, add governing equations;

[0016] S5. Select interpolation functions to define the mappings for the reverse osmosis function, thermal conductivity function, density function, and specific heat capacity function of the design variables in the governing equations;

[0017] S6. Use a Helmholtz density filter to filter the design variables, and then apply hyperbolic tangent projection to the filtered design variables to obtain the projected design variables.

[0018] S7. Construct an objective function with the goal of minimizing the overall average temperature and power consumption of the liquid cooling plate;

[0019] S8. Based on the volume fraction of the fluid domain in the liquid cooling plate, construct the constraint conditions;

[0020] S9. Based on the objective function, constraints, and design variables, construct a topology optimization model;

[0021] S10. Mesh the two-dimensional model of the design domain using triangular elements.

[0022] S11. The topology optimization model is solved iteratively using an optimization algorithm. During the solution process, the volume fraction is set as a preset value, the initial design variables are set, and the termination condition for the topology optimization model iteration is set as the objective function error value being lower than a preset error threshold. The design variables are updated using the optimization algorithm, and the solution is executed with the objective function minimization as the objective.

[0023] S12. Investigate the effects of different values ​​of weight coefficients and Reynolds number in the objective function on the topology optimization structure, in order to determine the optimal weight coefficients and Reynolds number;

[0024] S13. When post-processing the topology optimization results obtained in S12, a filter is used to smooth the optimized structure corresponding to the optimal parameters; where the optimal parameters are the optimal weight coefficient and the Reynolds number.

[0025] The determination of the two penalty parameters involved in the reverse osmosis function, thermal conductivity function, density function, and specific heat capacity function of the control equations regarding the design variables is specifically as follows: A second penalty parameter is set to a fixed value; within a preset range, the topological flow channel structure of the liquid-cooled plate obtained by taking different values ​​for the first penalty parameter is determined according to preset screening conditions, thereby determining the optimal value of the first penalty parameter; wherein, the first penalty parameter is a constant parameter in the reverse osmosis function, thermal conductivity function, and density function regarding the design variables; and the second penalty parameter is a constant parameter in the specific heat capacity function regarding the design variables.

[0026] Furthermore, using continuity and jagged regions as filtering conditions, the first penalty parameter value with the fewest discontinuous and jagged regions is selected as the optimal value.

[0027] The beneficial effects of this invention are:

[0028] This invention designs a symmetrically arranged multi-channel liquid cooling plate for a single-cell battery by cleverly combining a single-inlet, single-outlet DC channel with sub-channels. Furthermore, it optimizes the parameters using a topology optimization method to achieve topology optimization. Compared with straight-channel liquid cooling plates and serpentine-channel liquid cooling plates, the liquid cooling plate of this invention has better performance in both cooling and power consumption reduction, making it a high-performance liquid cooling plate structure. Attached Figure Description

[0029] Figure 1 The topological structure diagram of the present invention ( ).

[0030] Figure 2 The effect of different penalty parameters on the topology optimization structure.

[0031] Figure 3 The objective function and the objective function error vary with the grid.

[0032] Figure 4 Topology optimization structures with different weighting coefficients and Reynolds numbers.

[0033] Figure 5 This is a schematic diagram of a three-dimensional fluid heat transfer model and a cold plate structure.

[0034] Figure 6 The velocity distribution and inlet / outlet pressure difference of the cross-sections with different topologies.

[0035] Figure 7 Temperature distribution in different topologies. Detailed Implementation

[0036] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention. It should be noted that, unless otherwise specified, the embodiments and features in the embodiments of this application can be arbitrarily combined with each other.

[0037] Example 1: As Figures 1-7 As shown, according to a first aspect of the present invention, a symmetrically arranged topological multi-channel single-cell liquid cooling plate is provided, comprising a substrate, wherein topological channels are formed on the substrate, the topological channels being symmetrical about a centerline in the width direction, specifically including:

[0038] A liquid inlet direct flow channel is located on the center line of the liquid cooling plate along the width direction and is used for liquid inlet.

[0039] A liquid outlet direct channel is located on the center line of the liquid cooling plate along the width direction and is used for liquid outlet;

[0040] Sub-channel, located between the inlet direct channel and the outlet direct channel, is used to transport the coolant flowing into the inlet direct channel to the outlet direct channel.

[0041] For example, such as Figure 1 As shown, the centerline of the topological flow channel about the width direction is horizontal, and the topological flow channel of the liquid cooling plate is symmetrical about the centerline of the width direction. The sub-flow channel is used to connect the inlet flow channel and the outlet flow channel. The sub-flow channel includes a first sub-flow channel and a second sub-flow channel. The first sub-flow channel is distributed on the outer periphery, and the second sub-flow channel is located in the area enclosed by the first sub-flow channel. The coolant flows into the inlet flow channel from the inlet port. The coolant flowing into the inlet flow channel converges into the outlet flow channel through the first sub-flow channel and the second sub-flow channel, and is discharged through the outlet port of the outlet flow channel.

[0042] Furthermore, the thickness of the liquid cooling plate is 2mm-5mm; the length and width of the liquid cooling plate are determined according to the required battery shape. The inlet and outlet channels have the same dimensions and a rectangular cross-section. In this embodiment, the liquid cooling plate is 120mm long, 80mm wide, and the cross-sectional dimensions of the inlet and outlet channels are 1*5mm, with a length of 5mm. Figure 1 In the illustration, length is the dimension in the horizontal direction, and width is the dimension in the vertical direction.

[0043] Furthermore, the volume fraction of the topological flow channel region in the liquid cooling plate is 30%-70%. For example, the volume of the topological flow channel region can be 30%, 50%, 70%, etc., of the total volume of the substrate.

[0044] Furthermore, the substrate material of the liquid cooling plate is an aluminum plate.

[0045] Furthermore, the coolant can be water or a mixture of water and ethanol (volume ratio 1:1).

[0046] According to a second aspect of the present invention, a method for optimizing the topology flow channels of a symmetrically arranged multi-channel liquid cooling plate for a single-cell battery is provided, comprising the following steps:

[0047] S1. Construct a single-cell heat generation model based on the Bernardi heat generation model.

[0048] S2. Based on the contact surface between the battery and the liquid cooling plate, the dimensions of the liquid cooling plate substrate are designed using the finite element analysis software COMSOL Multiphysics. A two-dimensional model (the area that needs to be optimized) is constructed on the substrate by selecting the design domain. The length of the liquid cooling plate substrate is 120mm and the width of the substrate is 80mm. The design domain is slightly smaller than the substrate dimensions (based on the substrate being reduced inward by 1-2mm).

[0049] S3. Given the assumptions; the assumptions include:

[0050] Based on a Reynolds number below 2300, the flow is assumed to be laminar.

[0051] Assume the fluid is incompressible.

[0052] Based on the battery's heat generation rate characteristics and other researchers' assumptions, it is assumed that the heat dissipation of the liquid cooling plate is uniform.

[0053] Assume the design domain is a porous medium.

[0054] Assume that the thermal properties of the coolant (such as thermal conductivity, specific heat capacity, and dynamic viscosity) are constant.

[0055] S4. Within the liquid-cooled plate, heat exchange occurs between the fluid and the plate wall, constituting a fluid-solid conjugate heat transfer problem, which satisfies the three governing equations of fluid mechanics under steady-state conditions. Based on the assumptions, additional governing equations are added, including the fluid's energy balance equation, the design domain continuity equation, the momentum conservation equation, and the solid's energy equation.

[0056] The energy balance equation for a fluid is:

[0057] .

[0058] In the formula: For flow rate; For temperature; As a heat source; For temperature gradient; For Hamiltonian operators; Regarding design variables The thermal conductivity function; Regarding design variables Specific heat capacity function; Regarding design variables The density function. In the equation, material density, specific heat capacity, and thermal conductivity are all design variables. The function couples the control equations with the design variables in functional form.

[0059] Design domain continuity equation:

[0060] .

[0061] Momentum conservation equation:

[0062] .

[0063] In the formula: For pressure; Dynamic viscosity; This is a volume force term, which is added to the momentum equation to simulate the frictional resistance experienced by fluid flow in porous media. For pressure gradient; This represents the velocity gradient.

[0064] According to the Brinkman penalty model, the resistance to fluid flow in porous media is related to... Related, and also introduce reverse osmosis rate To control the local permeability of porous media, volume forces can be expressed as:

[0065] .

[0066] in, Regarding design variables The reverse osmosis function.

[0067] when When the frictional resistance is zero, fluid flow is unimpeded. However, as the frictional resistance approaches infinity, the frictional resistance becomes infinite. Therefore, changing... The value of can control the distribution of fluid and solid, while Also a design variable The function.

[0068] The energy equation for a solid is shown below:

[0069] .

[0070] In the formula, Density of solid materials; Specific heat capacity of solid materials; is the thermal conductivity of the solid material.

[0071] S5. Perform density-based topology optimization design, designing variables. Representing a localized material, it allows for control over the changes in the fluid-solid two-phase structure; ideally... The value is either 0 or 1 (0 represents solid, 1 represents fluid). However, from the perspective of numerical stability, It varies continuously from 0 to 1; in other words, intermediate values ​​are allowed to exist within the design domain; the effective thermal conductivity of the medium. and local reverse osmosis rate Depends on design variables To avoid local (non-global) optimization, the control equations regarding design variables... The mapping of the reverse osmosis function, thermal conductivity function, density function, and specific heat capacity function is defined by selecting an interpolation function, as follows:

[0072] .

[0073] .

[0074] .

[0075] .

[0076] In the formula: , , These are the thermal conductivity, density, and specific heat capacity of the solid material, respectively. , , These are the thermal conductivity, density, and specific heat capacity of the fluid material, respectively. These are the reverse osmosis rates for solids and fluids, respectively. It is usually a very small value; Theoretically, it is infinite, but for numerical reasons, a finite value is usually chosen. In this work, , ,when When the Reynolds number is fixed, It is a large constant, representing that the fluid does not flow inside the solid; The Darcy number is set to 0.001. When γ=0, the drag approaches infinity, and the design unit is a solid. When γ=1, the drag term is 0, and the design unit is a fluid. , , As the first penalty parameter, This is the second penalty parameter.

[0077] Furthermore, the control equations regarding design variables The determination of two penalty parameters involved in the reverse osmosis function, thermal conductivity function, density function, and specific heat capacity function is specifically as follows: The second penalty parameter is set to a fixed value; within a preset range, the topology of the liquid-cooled plate obtained after taking different values ​​for the first penalty parameter is determined according to preset screening conditions, based on the topological flow channel structure obtained; wherein, the first penalty parameter is related to the design variables. The constant parameters in the reverse osmosis function, thermal conductivity function, and density function; the second penalty parameter is about the design variables. The constant parameter in the specific heat capacity function.

[0078] Furthermore, using continuity and jagged regions as filtering conditions, the first penalty parameter value with the fewest discontinuous and jagged regions is selected as the optimal value.

[0079] , , and The selection process is as follows: When conducting topology optimization studies on structures, the penalty parameter has a significant impact on the final structure. If the penalty parameter is too large, the penalty effect is insufficient, leading to blurred boundaries or computational divergence. Conversely, if the penalty parameter is too small, the penalty effect is sufficient, but numerical divergence and chessboard problems are easily generated. According to existing literature, generally... Set to 100, in this case, the main discussion , , The impact of changes on topological results, in Figure 2 It shows The topological structure was studied when When equal to 100, , , The impact of a change from 0.01 to 0.2 on the final topology (red circles represent discontinuities, green squares represent jagged areas): The outermost channels of the five topologies form relatively clear boundaries, and the internal mesh area decreases. When... , , When it is 0.01 (e.g.) Figure 2 -a) The topology exhibits discontinuous and jagged regions, which are widely distributed. , , As the viscosity increases, the serrated area will gradually disappear, forming a smoother fluid-solid interface. , , After exceeding 0.1 (e.g.) Figure 2 -d、 Figure 2 -e), although the jagged regions disappear, multiple discontinuities appear, and the high penalty effect creates sharp fluid-structure interface configurations in the design domain, ultimately preventing convergence through iteration. , , When the value is equal to 0.1, the discontinuous and jagged regions in the topology optimization structure are relatively few. Therefore, in order to obtain a topology structure that meets the conditions, we choose... This will serve as a penalty parameter for future research.

[0080] The formula for calculating the Reynolds number is as follows:

[0081] .

[0082] In the formula, The characteristic length is set to 5 mm. The inlet velocity of the coolant is given. Generally, when the Reynolds number is less than approximately 2000, the fluid flow is laminar. In laminar flow, the fluid layers are parallel and do not mix, resulting in relatively stable flow.

[0083] S6. Use a Helmholtz density filter to filter the design variables, and then apply hyperbolic tangent projection to the filtered design variables to obtain the projected design variables.

[0084] Since the design variables are artificially assumed, there will be intermediate states of 0 to 1 after topology optimization. This will lead to unclear structures, numerical instability, and checkerboard distribution phenomena. To eliminate grid dependence and the checkerboard problem, a Helmholtz density filter is used to filter the design variables. Its expression is:

[0085] .

[0086] In the formula: These are the filtered design variables; The filter radius; for gradient.

[0087] Filtered design variables may result in a large number of grayscale cells. Therefore, by adopting hyperbolic tangent projection, the intermediate values ​​after filtering are reduced to obtain a clear topological shape. The expression is as follows:

[0088] .

[0089] In the formula: These are the projected design variables; and The projection slope and projection point are 16 and 0.5 respectively in this invention.

[0090] S7. Setting the objective function: The design requirements for battery liquid cooling plates generally include heat dissipation performance indicators and power consumption performance indicators, making it a multi-objective optimization problem. This invention considers the heat dissipation performance indicators as minimizing the overall average temperature of the liquid cooling plate and minimizing power consumption. The overall average temperature... expression:

[0091] .

[0092] in, This represents the area of ​​the liquid cooling plate.

[0093] Generally, to minimize the inlet and outlet pressure drop, power consumption must be minimized, which means reducing flow resistance. Therefore, the optimization objective is to minimize the power consumption of the liquid cooler, as expressed below:

[0094] .

[0095] The overall objective function is expressed as a weighted sum of two performance metrics, namely:

[0096] .

[0097] In the formula: and The initial values ​​for the two objective functions are used, and the objective functions are normalized by dividing by the initial values. , is the weighting coefficient, which is set to 0.9 in this invention.

[0098] S8. To ensure the topology optimization results have practical engineering significance, the volume fraction of the fluid domain in the liquid cooling plate was restricted, i.e., the constraint conditions were defined as follows:

[0099] .

[0100] In the formula: This is the volume fraction, which is the volume of the topology flow channel region relative to the total volume of the substrate. In this invention, the value is 50%.

[0101] S9. Based on the objective function, constraints, and design variables, construct a topology optimization model, which is expressed as follows:

[0102] .

[0103] S10. The two-dimensional model of the design domain is meshed using triangular elements; mesh independence verification is performed. Mesh independence verification in the topology optimization of liquid-cooled plate structures ensures the accuracy and reliability of the obtained design results, avoids differences in results due to different meshing methods, and improves the confidence level of the calculation results. This verification process helps confirm the robustness of the optimization results, meet industry standards, reduce design risks, and ensure that the results can be generalized to different application scenarios, while also meeting the requirements of scientific research rigor and interdisciplinary applications. The structure is studied using a mesh, and the objective function is... ,pass Figure 3 It can be seen that as the number of grid cells increases, the objective function value gradually decreases and then stabilizes. When the number of grid cells exceeds 40,000, the objective function value stabilizes at around 0.56, and the objective function error... Since the value is less than 0.001, 41799 is selected as the topology optimization calculation grid.

[0104] S11. Compile the topology optimization algorithm based on the COMSOL platform, and use the SNOPT algorithm to iteratively solve the topology optimization model. During the solution process, set the volume fraction to 50% and set the initial design variables. The value is 0.5, and the model iteration terminates when the objective function error value is lower than 0.5. Using optimization algorithms to optimize design variables Update the algorithm, aiming to minimize the objective function, and execute the solution process to observe the convergence during optimization.

[0105] S12, Studying different weighting coefficients and Reynolds number The impact on topology optimization structure was investigated to determine the optimal weighting coefficients and Reynolds number: Under the condition of a volume fraction of 50%, the Reynolds number was adjusted accordingly. For four cases: 50, 100, 150, and 200, the weighting coefficients are... Topology optimization was performed under 20 conditions (0.1, 0.3, 0.5, 0.7, and 0.9) to obtain a series of liquid-cooled plate structures, such as... Figure 4 As shown. In Figure 4 The diagram shows the velocity and temperature of the topology formed by different weighting coefficients and Reynolds numbers (the upper part is the velocity field, and the lower part is the temperature field). It reveals that, under a fixed Reynolds number, as the weighting coefficient increases, the flow channel gradually disperses and diffuses to all areas of the cooling plate, and the solid region also gradually disperses. The weighting coefficient has a significant impact on the heat transfer and flow of the liquid cooling plate. For example, when At this point, the flow resistance is minimal, the channel is very smooth, and the inlet and outlet channels are simple, consisting of only one curved channel. Therefore, the flow resistance is minimal in all designs. However, the temperature distribution is very uneven, with significantly higher temperatures in areas far from the fluid or channel, indicating poor local heat transfer. This is because the power dissipation weighting factor is heavily weighted towards reducing flow resistance, while lower flow resistance ensures a lower pressure difference between the inlet and outlet. However, this also leads to poor heat exchange performance of the target, resulting in an increase in the average and maximum temperatures inside the liquid cooling plate channel. ).when At this stage, the number of channels increases the most, and the channels become more intertwined and branched. More and more slender branches grow from the wide main channel, guiding the fluid to areas away from the inlet and main channel. Multiple curved channels are evenly distributed throughout the design area, cutting off the solid region, preventing heat accumulation in the solid region and heat transfer from the radiator, resulting in a more uniform temperature field. This improves the heat transfer performance of the liquid cooling plate, giving it a lower average temperature and a lower maximum temperature. However, excessive branch structures also increase flow resistance, leading to a greater pressure difference between the inlet and outlet. Therefore, the weighting factors explicitly shape the topology through a multi-objective function, thereby creating an optimal design that aligns with the desired objectives.

[0106] The magnitude of the Reynolds coefficient directly affects the change in inlet velocity. Figure 4 It was observed that, regardless of the value, as As the pressure difference between inlet and outlet increases, both the average and maximum temperatures decrease. hour, The changes have little impact on the topology; all consist of three main roads. Get bigger The changes did not significantly increase the number of branches in the topology, and the distribution of branches exhibited diversity. To ensure the mesh quantity and quality of the liquid cooling plate channels in subsequent studies, comparisons were made. Figure 4 After considering different structures, select hour( Figure 4 dashed box, The topology of the liquid cooling plate fluid channel structure was used as the basis for subsequent research. This topology has a uniform temperature distribution and the number of liquid cooling plate channel branches ensures that the pressure difference between the inlet and outlet is not too large, making it a relatively ideal topology.

[0107] S13. When post-processing the topology optimization results obtained in S12, a filter is used to smooth the optimized structure corresponding to the optimal parameters. The lower limit of the filter is set to 0.5 to obtain a topology with smoother boundaries.

[0108] S14. After processing the filtered topology in step S13, export the final data and save it as a text file for subsequent three-dimensional numerical simulation.

[0109] S15. Numerical simulation of the three-dimensional model was performed. Based on the optimized liquid cooling plate structure data, a three-dimensional model of the liquid cooling plate (TOP structure) was established. Correspondingly, three types of linear channel cold plates (LCP) and serpentine channel cold plates (SCP) with different fluid volume fractions were established and studied. The performance was compared with that of the topology-optimized liquid cooling plate (TOP). To study the influence of the average fluid volume fraction, two topological structures with average fluid volume fractions of 30% and 70% were also established. Figure 5 The image shows nine 3D fluid heat transfer models and liquid cooling plate structures. The models are simplified, with a battery embedded between two liquid cooling plates, each 2mm thick. The battery is a square battery with dimensions of 80mm x 120mm x 12mm. To simulate a realistic environment, 25°C water is used as the coolant, aluminum as the cooling plate material, and the external ambient temperature is 25°C. The surface heat transfer coefficient of the battery and cooling plates, excluding the heating surface, is... Ignoring thermal radiation.

[0110] S16. Based on the coolant fluid quality, cold plate material, and selected battery, set the relevant parameters; the battery density is 2055 kg / m³. 3 The average specific heat capacity is 1399 J / (kg·K), and the thermal conductivity is anisotropic, with the horizontal thermal conductivity being... Thermal conductivity in the thickness direction Set initial conditions, with the inlet condition set to a mass-velocity inlet and the inlet velocity as follows: The outlet condition is set to pressure outlet, with the default outlet pressure being 0 Pa; a volumetric heat source is used for battery heat generation simulation, and the battery operates at a constant heat flux j on the heating surface during 5C discharge. During this stage, the surface heat generation rate of the battery discharged at 5C is measured. It is used as a heat source for the cooling plate and is applied to the cooling plate.

[0111] S17. After completing S17, the three-dimensional model is meshed. This invention uses a triangular mesh to mesh the three-dimensional model. At the same time, in order to ensure the accuracy of the calculation, the number of meshes is refined in the fluid-structure boundary layer. Then, mesh independence analysis is performed. The purpose of mesh independence analysis is to ensure that the mesh used is refined enough to obtain an accurate solution that is independent of the mesh size. The specific operation is the same as step S11.

[0112] S19. Use COMSOL Multiphysics to perform fluid-structure interaction heat transfer analysis on the 3D model, solve the governing equations of the model, and finally obtain the temperature distribution, fluid pressure, and velocity distribution of the battery cooling plate. Figure 6 The velocity and pressure distributions at different topological cross-sections are shown. Figure 7 The temperature distribution across different topological sections.

[0113] exist Figure 6 In (a), the cross-section is a plane divided at half the height of the bottom liquid cooling plate. It is observed that the smaller the average fluid volume fraction, the greater the velocity distribution within the liquid cooling plate channel. As the average fluid volume fraction increases, the velocity distribution within the liquid cooling plate channel gradually becomes more uniform. When the average fluid volume fraction... In this invention and the conventional LCP structure, the velocity field of the liquid-cooled plate flow channel is mainly distributed in the inlet and outlet areas and adjacent areas, while the velocity field is significantly reduced far from the inlet and outlet areas. In the TOP structure, the high-velocity distribution area spreads from the inlet and outlet to all channels within the liquid-cooled plate flow channel, while in the LCP structure, the high-velocity area is mainly concentrated in the middle few linear regions. For the SCP structure, the velocity field distribution inside the structure with the three average fluid volume fractions is more uniform than that of the TOP and LCP structures, and all liquid-cooled plate flow channels maintain a high velocity.

[0114] The histogram of the maximum pressure difference for each structure is shown below. Figure 6 As shown in (b), the average fluid volume fraction after topology optimization The TOP structure has the smallest pressure drop at 17.595 Pa and the average fluid volume fraction. The SCP structure exhibits the largest pressure drop, at 100 Pa. For both TOP and LCP structures, the pressure drop of the TOP structure is lower than that of the LCP structure under all three average fluid volume fractions. For the TOP, LCP, and SCP structures, the pressure drop gradually decreases with increasing average fluid volume fraction. For the TOP and LCP structures, the pressure drop changes very little after the average fluid volume fraction exceeds 50%, while the SCP structure shows a significantly larger pressure drop than the other two structures, which is also evident in the velocity field. Therefore, the topology-optimized TOP structure has the smallest pressure difference, resulting in minimal flow resistance in the liquid-cooled plate channel and thus minimal pump power loss. The SCP structure, with its higher velocity field distribution, exhibits a larger pressure drop, making it unsuitable for heat-fluid coupling calculations.

[0115] The battery's maximum temperature is often a key factor affecting its lifespan. From... Figure 7(a) It can be intuitively seen that the highest temperature of the 5C exothermic square battery cell obtained from simulation calculations for different structures is controlled below 304 K, keeping the battery within a suitable temperature range. This proves that these structures can meet the heat dissipation requirements. Compared with LCP and SCP structures, the temperature distribution of the fluid in the flow channel of the TOP structure is more uniform, and the temperature distribution of the cold plate is also more uniform. Observation Figure 7 (b) In the analyzed structures, the maximum temperature of all three structures decreased with increasing average fluid volume fraction. The TOP structure showed the largest decrease in maximum temperature (0.51 K) as the average fluid volume fraction increased from 30% to 50%, while the LCP and SCP structures exhibited more gradual changes in maximum temperature. Comparing the maximum temperatures of the three structures, the TOP structure had the lowest maximum temperature, while the LCP structure had the highest, but the differences were all within 1 K. Figure 7 (c) It can be seen that the average temperature difference of the LCP structure is higher than that of the other two structures, but the change in the average fluid volume fraction has little effect on the average temperature. The average temperature of the TOP structure is around 301.5K, the LCP structure is around 302.2K and the SCP structure is around 301.95K.

[0116] The above results indicate that when choosing hour( Figure 4 When the liquid cooling plate fluid channel structure (with black frame) is the final structure of this invention, it can make the temperature distribution uniform, and the number of liquid cooling plate channel branches can prevent the pressure difference between the inlet and outlet from being too large. It combines the performance of liquid cooling plates in terms of temperature and pressure, and is a relatively ideal topology structure.

[0117] The specific embodiments of the present invention have been described in detail above with reference to the accompanying drawings. However, the present invention is not limited to the above embodiments. Within the scope of knowledge possessed by those skilled in the art, various changes can be made without departing from the spirit of the present invention.

Claims

1. A symmetrically arranged topological multi-channel single-cell liquid cooling plate, comprising a substrate, wherein topological channels are formed on the substrate, characterized in that, The topological flow channel is symmetrical about the centerline in the width direction, specifically including: A liquid inlet direct flow channel is located on the center line of the liquid cooling plate along the width direction and is used for liquid inlet. A liquid outlet direct channel is located on the center line of the liquid cooling plate along the width direction and is used for liquid outlet; Sub-channel, located between the inlet direct channel and the outlet direct channel, is used to transport the coolant flowing into the inlet direct channel to the outlet direct channel.

2. The symmetrically arranged multi-channel liquid cooling plate for a single-cell battery according to claim 1, characterized in that, The sub-channels include a first sub-channel and a second sub-channel; the first sub-channel is distributed on the outer periphery, and the second sub-channel is located in the area enclosed by the first sub-channel; the coolant flows into the inlet direct channel from the inlet port, and the coolant flowing into the inlet direct channel converges into the outlet direct channel through the first sub-channel and the second sub-channel, and is discharged through the outlet port of the outlet direct channel.

3. The symmetrically arranged multi-channel liquid cooling plate for a single-cell battery according to claim 1, characterized in that, The volume fraction of the topological flow channel region in the liquid cooling plate is 30%-70%.

4. A method for optimizing the topology flow channels of a liquid cooling plate for a symmetrically arranged multi-channel single-cell battery according to any one of claims 1-3, characterized in that, include: S1. Construct a single-cell heat generation model based on the Bernardi heat generation model; S2. Based on the contact surface between the battery and the liquid cooling plate, the dimensions of the liquid cooling plate substrate are designed using finite element analysis software, and a two-dimensional model is constructed by selecting a design domain on the substrate. S3. Given the assumptions; S4. Based on the assumptions, add governing equations; S5. Select interpolation functions to define the mappings for the reverse osmosis function, thermal conductivity function, density function, and specific heat capacity function of the design variables in the governing equations; S6. Use a Helmholtz density filter to filter the design variables, and then apply hyperbolic tangent projection to the filtered design variables to obtain the projected design variables. S7. Construct an objective function with the goal of minimizing the overall average temperature and power consumption of the liquid cooling plate; S8. Based on the volume fraction of the fluid domain in the liquid cooling plate, construct the constraint conditions; S9. Based on the objective function, constraints, and design variables, construct a topology optimization model; S10. Mesh the two-dimensional model of the design domain using triangular elements. S11. The topology optimization model is solved iteratively using an optimization algorithm. During the solution process, the volume fraction is set as a preset value, the initial design variables are set, and the termination condition for the topology optimization model iteration is set as the objective function error value being lower than a preset error threshold. The design variables are updated using the optimization algorithm, and the solution is executed with the objective function minimization as the objective. S12. Investigate the effects of different values ​​of weight coefficients and Reynolds number in the objective function on the topology optimization structure, in order to determine the optimal weight coefficients and Reynolds number; S13. When post-processing the topology optimization results obtained in S12, a filter is used to smooth the optimized structure corresponding to the optimal parameters; where the optimal parameters are the optimal weight coefficient and the Reynolds number.

5. The method for optimizing the topology flow channels of the symmetrically arranged multi-channel liquid cooling plate for a single-cell battery according to claim 4, characterized in that, The determination of the two penalty parameters involved in the reverse osmosis function, thermal conductivity function, density function, and specific heat capacity function of the control equations regarding the design variables is as follows: Set a fixed value for the second penalty parameter; Within a preset range, the topology of the liquid cooling plate obtained by taking different values ​​of the first penalty parameter is determined according to preset screening conditions, thereby determining the optimal value of the first penalty parameter. The first penalty parameter is a constant parameter in the reverse osmosis function, thermal conductivity function, and density function of the design variables; the second penalty parameter is a constant parameter in the specific heat capacity function of the design variables.

6. The method for optimizing the topology flow channels of the symmetrically arranged multi-channel liquid cooling plate for a single-cell battery according to claim 5, characterized in that, Using continuity and jagged regions as filtering criteria, the first penalty parameter value with the fewest discontinuous and jagged regions is selected as the optimal value.

Images