Microchannel heat sink structure with coupled rib cavity combination and design method thereof

Abstract

The application discloses a micro-channel heat sink structure of a coupling rib cavity combination and a design method thereof, and relates to the technical field of micro-channel heat dissipation.The micro-channel heat sink structure comprises a micro-channel base body with a double-layer flow channel structure, and the double-layer flow channel structure comprises an inner micro flow channel, an outer micro flow channel, an upper micro flow channel and a lower micro flow channel; a plurality of coupling rib cavity combination structures are arranged on the bottom surface of the lower micro flow channel in a periodic arrangement; and a bottom heating surface is arranged outside the bottom surface of the lower micro flow channel, and the bottom heating surface is tightly combined with a heat sink solid area, so that heat is transmitted to cooling working medium in the lower micro flow channel through the coupling rib cavity combination structure.The application can significantly improve the heat dissipation capacity and the heating surface temperature uniformity under the condition of extreme heat flux density while maintaining controllable pressure drop.

Description

Technical Field

[0001] This invention relates to the field of microchannel heat dissipation technology, and in particular to a microchannel heat sink structure with coupled fins and cavity combination and its design method. Background Technology

[0002] With the rapid development of technologies such as artificial intelligence and high-performance computing, chips are constantly evolving towards higher integration and higher power consumption, leading to a continuous increase in heat flux density per unit area. Efficient heat dissipation has become a key factor restricting breakthroughs in chip technology. Currently, the average heat flux density of chips has reached 500 W·cm². -2 The heat flux density of its local hotspots will exceed 1000 W·cm⁻¹ -2 If such a high heat flux density cannot be dissipated in time, the chip temperature will rise rapidly. Excessive temperature will not only damage or accelerate the aging of internal electronic components and materials, leading to decreased reliability and stability of chip operation, but also shorten the chip's lifespan. Therefore, effective temperature control and enhanced heat dissipation are crucial for chip performance.

[0003] Existing microchannel cooling technologies first design the basic structural parameters of rectangular or trapezoidal microchannels and teardrop-shaped or rhomboid micro-needle ribs based on the overall low to medium heat flux density requirements of the chip. Then, by constructing multi-layer stacked channels or adopting a tree-like branched multi-inlet and multi-outlet layout, the relatively low-temperature coolant is directly diverted and guided to the core heat-generating area of ​​the chip to alleviate the temperature rise along the way. Periodically arranged concave cavity structures are etched on the sidewalls or bottom surfaces of the channels, and the backflow vortices and separation vortices generated when the fluid flows through them are used to enhance the mixing disturbance between the hot and cold fluids. The optimized microchannel cold plate is bonded to the chip through a thermal interface material or directly etched and integrated on the back of the chip to achieve efficient transfer and discharge of heat from inside the chip to the coolant, thereby completing the microchannel cooling function under the given operating conditions.

[0004] Current microchannel heat sink structures are designed and optimized for low to medium heat flux density conditions. When operating under extremely high heat flux conditions, the space for the development of the thermal boundary layer between the fluid and the wall is extremely compressed and quickly enters a saturated state. This makes it impossible for passive perturbation methods based on geometric optimization to break the thermally conductive boundary layer in the near-wall region, which has reached its physical limit. Consequently, the heat dissipation advantage of this microchannel heat sink structure is significantly weakened. Summary of the Invention

[0005] Therefore, it is necessary to provide a microchannel heat sink structure with coupled rib cavity combination and its design method to address the above-mentioned technical problems.

[0006] This invention provides a microchannel heat sink structure with coupled rib cavity assembly, comprising: A microchannel substrate with a dual-layer flow channel structure includes: an inner microchannel, an outer microchannel, an upper microchannel, and a lower microchannel; the cooling medium in the inner microchannel flows in from a first inlet located in the upper microchannel, flows through the lower microchannel, and flows out from a first outlet located in the upper microchannel; the cooling medium in the outer microchannel flows in from a second inlet and a third inlet located in the upper microchannel, and flows out from a second outlet and a third outlet located in the lower microchannel. The bottom surface of the lower microchannel is provided with a combination of multiple periodically arranged coupling rib cavity structures. The combination of coupling rib cavity structures is composed of alternating combinations of ribs protruding from the bottom surface of the lower microchannel and concave cavities, so as to form a periodic tapering-expanding cross section in the lower microchannel. The bottom outer surface of the lower microchannel is provided with a bottom heating surface for simulating the heat source of the chip. The bottom heating surface is in close contact with the heat sink solid area so that the heat can be transferred to the cooling working fluid in the lower microchannel through the coupling rib cavity combination structure.

[0007] Optionally, the ribs and cavities in the coupled rib-cavity combination structure have any one of the following geometric shapes in cross-section: rectangular, elliptical, or spindle-shaped.

[0008] Optionally, the rib height ranges from 400μm to 800μm, the rib width ranges from 100μm to 300μm, the cavity length ranges from 362μm to 762μm, and the cavity height ranges from 300μm to 900μm.

[0009] Optionally, the spacing between adjacent coupled rib cavity combination structures is 800 μm.

[0010] This invention provides a design method for a microchannel heat sink structure with coupled rib cavity assembly, comprising: Obtain the basic shape and initial structural parameters of the rib-cavity assembly. The initial structural parameters include: rib height, rib length, rib width, cavity length, cavity width, and cavity height. An orthogonal experimental design table corresponding to the rib cavity assembly was constructed using the Taguchi method to determine the contribution rate of each initial structural parameter to the performance index, and the initial structural parameters with a contribution rate greater than a preset threshold were taken as key structural parameters. By constructing an experimental scheme corresponding to the key structural parameters using the response surface method and conducting numerical simulations, a multivariate regression prediction model is obtained to quantitatively characterize the nonlinear mapping relationship between each key structural parameter and the heat transfer-flow performance of the heat sink. With the optimization objectives of maximizing the comprehensive performance evaluation factor, minimizing the pressure drop and minimizing the maximum temperature of the heating surface, a non-dominated sorting genetic algorithm with an elite strategy is used to solve the multivariate regression model for multi-objective optimization, and the Pareto optimal front solution set is obtained. Each solution in the Pareto optimal front solution set corresponds to a set of alternative parameter combinations for the microchannel heat sink structure. The Pareto optimal front solution set is analyzed by multi-attribute decision analysis using the entropy weight TOPSIS method to determine the relative fit of each candidate parameter combination. The candidate parameter combination with the highest relative fit is taken as the optimal compromise structural parameter combination. The optimal compromise structural parameter combination is used to design the microchannel heat sink structure with coupled rib cavity combination.

[0011] Optionally, an orthogonal experimental design table corresponding to the rib-cavity assembly is constructed using the Taguchi method to determine the contribution rate of each initial structural parameter to the performance index. Initial structural parameters with a contribution rate greater than a preset threshold are designated as key structural parameters, specifically including: The six initial structural parameters, namely rib height, rib length, rib width, cavity length, cavity width, and cavity height, are used as influencing factors, and multiple level values ​​are set for each influencing factor. A six-factor, five-level orthogonal experimental design table was constructed based on the influencing factors and their level values. Numerical simulations were performed based on the orthogonal experimental design table to determine the average Nusselt number, pressure drop, comprehensive performance evaluation factor, and signal-to-noise ratio of the highest temperature performance index of the heating surface for each group of experiments. The range and contribution rate are determined based on the average signal-to-noise ratio of each influencing factor at different levels. Initial structural parameters with contribution rates greater than a preset threshold are taken as key structural parameters. Key structural parameters include: rib height, rib width, cavity length, and cavity height.

[0012] Optionally, an experimental scheme corresponding to the key structural parameters is constructed using the response surface methodology and numerical simulation is performed to obtain a multiple regression prediction model for quantifying the nonlinear mapping relationship between each key structural parameter and the heat sink's heat transfer-flow performance. This model specifically includes: Based on key structural parameters, a response surface methodology was constructed using the Box-Behnken design method, where the range of values ​​for the key structural parameters was used as the design space, and repeated experiments were set at the center point. CFD numerical simulations were performed on each sample point in the response surface methodology to obtain the corresponding pressure drop, maximum temperature of the heating surface, and response values ​​of the comprehensive performance evaluation factor. Based on the response values, a quadratic polynomial multiple regression prediction model is constructed by fitting the least squares method to characterize the relationship between key structural parameters and response values. The significance of the quadratic multiple regression prediction model was tested by analysis of variance, and insignificant terms with significance greater than a set threshold were removed to obtain the final multiple regression prediction model.

[0013] Optionally, with the optimization objectives of maximizing the comprehensive performance evaluation factor, minimizing the pressure drop, and minimizing the maximum temperature of the heating surface, a non-dominated sorting genetic algorithm with an elitist strategy is used to solve the multivariate regression model for multi-objective optimization, obtaining the Pareto optimal frontier solution set, specifically including: Maximizing the comprehensive performance evaluation factor is taken as the first objective function, minimizing the pressure drop is taken as the second objective function, and minimizing the maximum temperature of the heating surface is taken as the third objective function; Initialize the parent population consisting of individuals corresponding to multiple sets of key structural parameters, and set the algorithm parameters such as the number of iterations, crossover probability, and mutation probability; In each iteration, the function values ​​of the first objective function, the second objective function, and the third objective function are determined according to the key structural parameters corresponding to each individual; and the individuals in the current population are sorted in a non-dominated order, dividing the individuals into different non-dominated levels, where individuals in the lower levels are not completely dominated by other individuals in the three objectives. For individuals within the same non-dominated level, their crowding distance in the target space consisting of the first objective function, the second objective function, and the third objective function is determined to measure the density of the individual distribution. Parent individuals with low non-dominated levels and large crowding distances are selected from the current population through a tournament selection strategy for crossover and mutation operations to obtain the offspring population. The parent and offspring populations are merged to obtain a mixed population. Non-dominated sorting and crowding distance calculation are performed on the mixed population. Based on the principle of non-dominated hierarchy from low to high and crowding distance within the same hierarchy from large to small, multiple individuals are selected to form the parent population of the new generation. Repeat the iterative process until the preset iteration termination condition is met, and finally take all individuals in the parent population whose non-dominated level is the first level as the Pareto optimal frontier solution set.

[0014] Optionally, multi-attribute decision analysis is performed on the Pareto optimal frontier solution set using the entropy-weighted TOPSIS method to determine the relative fit of each candidate parameter combination, and the candidate parameter combination with the highest relative fit is taken as the optimal compromise structure parameter combination, specifically including: Each solution in the Pareto optimal front solution set is taken as an alternative scheme, and the comprehensive performance evaluation factor, pressure drop and maximum heating surface temperature of each scheme are used as evaluation indicators to construct the original evaluation matrix. The original evaluation matrix is ​​dimensionless to construct a standardized decision matrix, in which the comprehensive performance evaluation factor is treated as a positive indicator, and the pressure drop and the maximum temperature of the heating surface are treated as negative indicators. The objective weights of each evaluation index are determined using the information entropy method, and a weighted standardized decision matrix is ​​constructed. Based on the weighted standardized decision matrix, the positive and negative ideal solutions of each evaluation index are determined, and the Euclidean distance from each alternative to the positive and negative ideal solutions is determined. The relative fit of each alternative is determined by Euclidean distance, and the parameter combination corresponding to the alternative with the highest relative fit is taken as the optimal compromise structural parameter combination.

[0015] The microchannel heat sink structure and its design method with coupled rib cavity combination provided in this invention have the following advantages compared with the prior art: This invention utilizes a coupled rib-cavity combination structure, consisting of alternating raised ribs and recessed cavities, to force the formation of a periodic tapering-expanding cross-section within the flow channel. When the cooling medium flows through the tapering section, the contraction of the flow cross-section causes a sharp increase in flow velocity, resulting in strong scouring of the wall surface and effectively thinning the thermally conductive boundary layer near the wall. Upon entering the expanding section, the sudden expansion of the cross-section induces large-area backflow vortices and separation vortices, forcing the high-temperature stagnant fluid adhering to the wall surface to detach and violently mix with the low-temperature fluid in the mainstream core region.

[0016] The unique vortex dynamics mechanism generated by this coupled rib cavity structure can significantly disrupt and penetrate the near-wall thermal boundary layer that tends to saturate under extremely high heat flux conditions. This fundamentally overcomes the limitation that traditional smooth walls or single geometric morphology disturbances cannot break through the boundary layer saturation, thereby significantly improving the heat dissipation capacity and heating surface temperature uniformity under extreme heat flux density conditions while maintaining controllable pressure drop. Attached Figure Description

[0017] Figure 1 This is a schematic diagram of a microchannel heat sink structure with coupled rib cavity assembly provided in one embodiment. Figure 2 This is a schematic diagram of a microchannel heat sink structure with coupled rib cavity assembly provided in one embodiment. Figure 2 (a) in the diagram represents a rectangular ribbed cavity structure. Figure 2 (b) in the diagram is an elliptical ribbed structure. Figure 2 (c) in the diagram represents a spindle-shaped rib cavity structure; Figure 3 This diagram illustrates the effect of the rib cavity assembly shape on the performance of a microchannel heat sink structure with coupled rib cavity assembly, as provided in one embodiment. Figure 3 In the figure, (a) represents the effect of the rib cavity assembly shape on performance Δ. PImpact diagram Figure 3 (b) shows the effect of the shape of the rib cavity assembly on performance. Nu avg Impact diagram Figure 3 (c) in the figure represents the effect of the rib cavity assembly shape on performance. PEC Impact diagram Figure 3 In the figure, (d) represents the effect of the rib cavity assembly shape on performance. δ T Impact diagram; Figure 4 This diagram illustrates the effect of factor levels on the performance of a microchannel heat sink structure with coupled rib cavity assembly provided in one embodiment. Figure 4 (a) represents the effect of factor levels on performance. Nu avg Impact diagram Figure 4 In the figure (b), the factor level affects the performance Δ. P Impact diagram Figure 4 (c) represents the effect of factor levels on performance. PEC Impact diagram Figure 4 In the figure (d), the factor level affects the performance. T max Impact diagram; Figure 5 This is a schematic diagram illustrating the contribution of factor levels to various performance characteristics of a microchannel heat sink structure with coupled rib cavity assembly provided in one embodiment. Figure 6 Δ is provided in one embodiment as a microchannel heat sink structure with coupled rib cavity assembly. P Comparison chart of predicted and actual values; Figure 7 One embodiment provides a microchannel heat sink structure with coupled rib cavity assembly. T max Comparison chart of predicted and actual values; Figure 8 One embodiment provides a microchannel heat sink structure with coupled rib cavity assembly. PEC Comparison chart of predicted and actual values; Figure 9 The Pareto optimal frontier curve is shown for a microchannel heat sink structure with coupled rib cavity combination provided in one embodiment. Figure 10 This is a schematic diagram showing the relative fit of various schemes of a microchannel heat sink structure with a coupled rib cavity combination provided in one embodiment; Figure 11 This is a comparison of cross-sectional velocity contour maps of a microchannel heat sink structure with coupled rib cavity assembly provided in one embodiment. Figure 11 (a) in the figure is the cross-sectional velocity contour plot of the smooth channel. Figure 11(b) in the figure is the cross-sectional velocity contour plot of the initial rib cavity structure. Figure 11 (c) in the figure is the cross-sectional velocity contour plot of the optimized rib cavity structure; Figure 12 This is a comparison of the heating surface temperature cloud diagrams of a microchannel heat sink structure with coupled rib cavity assembly provided in one embodiment. Figure 12 (a) in the figure is a temperature contour map of the heating surface of the smooth channel. Figure 12 (b) in the figure is the temperature contour map of the heating surface of the initial rib cavity structure. Figure 12 (c) in the figure is the temperature cloud map of the heating surface of the optimized rib cavity structure. Detailed Implementation

[0018] 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.

[0019] In traditional electronic device heat dissipation methods, natural convection is difficult to meet the heat dissipation requirements of higher heat flux density due to its low heat transfer coefficient; forced air cooling and conventional liquid cooling thermal management technologies can enhance heat transfer to a certain extent, but they usually rely on external power devices, resulting in high system complexity and energy consumption; jet impact and spray cooling systems have strong heat transfer capabilities, but the nozzle arrangement is complex and difficult to adapt to packaging integration in narrow spaces.

[0020] Compared to traditional heat dissipation technologies, microchannel heat dissipation technology can be directly integrated into the device substrate, offering advantages such as short heat transfer paths, high heat exchange efficiency, and compact structure. Therefore, it shows promising application prospects in the thermal management of high heat flux density electronic devices. Furthermore, by optimizing the design of microchannels, such as adjusting inlet flow rate, cooling medium, and microchannel surface roughness, the heat dissipation performance of microchannel heat sinks can be significantly improved. For the heat dissipation requirements of chips with multiple heat sources and non-uniform heat flow, further improvements in heat transfer performance can be achieved by rationally designing the structural parameters of microchannels / micro-needles, adopting multi-layered flow channels and multi-inlet / multi-outlet layouts, and incorporating cavities, thereby increasing the heat transfer area and enhancing flow disturbance.

[0021] This invention provides a microchannel heat sink structure with coupled rib cavity assembly, comprising: A microchannel substrate with a dual-layer flow channel structure includes: an inner microchannel, an outer microchannel, an upper microchannel, and a lower microchannel. The cooling medium in the inner microchannel flows in through a first inlet located in the upper microchannel, passes through the lower microchannel, and exits through a first outlet located in the upper microchannel. The cooling medium in the outer microchannel flows in through a second inlet and a third inlet located in the upper microchannel, and exits through a second outlet and a third outlet located in the lower microchannel.

[0022] The bottom surface of the lower microchannel has a combination of multiple periodically arranged coupling rib cavity structures. The combination of coupling rib cavity structures is composed of alternating combinations of ribs protruding from the bottom surface of the lower microchannel and concave cavities, so as to form a periodic tapering-expanding cross section in the lower microchannel.

[0023] The bottom outer surface of the lower microchannel is provided with a bottom heating surface for simulating the heat source of the chip. The bottom heating surface is in close contact with the heat sink solid area so that the heat can be transferred to the cooling working fluid in the lower microchannel through the coupling rib cavity combination structure.

[0024] Preferably, the ribs and cavities in the coupled rib-cavity combination structure have any one of the following geometric shapes in cross-section: rectangular, elliptical, or spindle-shaped.

[0025] Preferably, the rib height ranges from 400μm to 800μm, the rib width ranges from 100μm to 300μm, the cavity length ranges from 362μm to 762μm, and the cavity height ranges from 300μm to 900μm.

[0026] Preferably, the spacing between adjacent coupled rib cavity combination structures is 800 μm.

[0027] Based on the same inventive concept, embodiments of the present invention also provide a design method for a microchannel heat sink structure with coupled rib cavity assembly, the method comprising: Obtain the basic shape and initial structural parameters of the rib-cavity assembly. The initial structural parameters include: rib height, rib length, rib width, cavity length, cavity width, and cavity height.

[0028] An orthogonal experimental design table corresponding to the rib cavity assembly was constructed using the Taguchi method to determine the contribution rate of each initial structural parameter to the performance index, and the initial structural parameters with a contribution rate greater than a preset threshold were taken as key structural parameters.

[0029] By constructing an experimental scheme corresponding to the key structural parameters using the response surface methodology and conducting numerical simulations, a multivariate regression prediction model is obtained to quantify the nonlinear mapping relationship between each key structural parameter and the heat transfer-flow performance of the heat sink.

[0030] With the optimization objectives of maximizing the comprehensive performance evaluation factor, minimizing the pressure drop, and minimizing the maximum temperature of the heating surface, a non-dominated sorting genetic algorithm with an elitist strategy is used to solve the multivariate regression model for multi-objective optimization, and the Pareto optimal front solution set is obtained. Each solution in the Pareto optimal front solution set corresponds to a set of alternative parameter combinations for the microchannel heat sink structure.

[0031] The Pareto optimal front solution set is analyzed by multi-attribute decision analysis using the entropy weight TOPSIS method to determine the relative fit of each candidate parameter combination. The candidate parameter combination with the highest relative fit is taken as the optimal compromise structural parameter combination. The optimal compromise structural parameter combination is used to design the microchannel heat sink structure with coupled rib cavity combination.

[0032] Preferably, an orthogonal experimental design table corresponding to the rib-cavity assembly is constructed using the Taguchi method to determine the contribution rate of each initial structural parameter to the performance index, and initial structural parameters with a contribution rate greater than a preset threshold are designated as key structural parameters, specifically including: Six initial structural parameters—rib height, rib length, rib width, cavity length, cavity width, and cavity height—were used as influencing factors, and multiple level values ​​were assigned to each factor. A six-factor, five-level orthogonal experimental design table was constructed based on these influencing factors and their level values. Numerical simulations were performed using the orthogonal experimental design table to determine the signal-to-noise ratio of the average Nusselt number, pressure drop, comprehensive performance evaluation factor, and highest heating surface temperature performance index for each experimental group.

[0033] The range and contribution rate are determined based on the average signal-to-noise ratio of each influencing factor at different levels. Initial structural parameters with contribution rates greater than a preset threshold are taken as key structural parameters. Key structural parameters include: rib height, rib width, cavity length, and cavity height.

[0034] Preferably, an experimental scheme corresponding to the key structural parameters is constructed using the response surface methodology and numerical simulation is performed to obtain a multiple regression prediction model for quantifying the nonlinear mapping relationship between each key structural parameter and the heat sink's heat transfer-flow performance. Specifically, this includes: Based on key structural parameters, a response surface methodology (CSMA) test scheme was constructed using the Box-Behnken design method. The design space was defined by the range of values ​​for the key structural parameters, and repeated tests were conducted at the center point. CFD numerical simulations were performed on each sample point in the CSMA test scheme to obtain the corresponding pressure drop, maximum heating surface temperature, and response values ​​for the comprehensive performance evaluation factor.

[0035] Based on the response values, a quadratic multinomial multiple regression prediction model was constructed using the least squares method to characterize the relationship between key structural parameters and response values. The significance of the quadratic multiple regression prediction model was tested using analysis of variance, and insignificant terms with significance greater than a set threshold were removed to obtain the final multiple regression prediction model.

[0036] Preferably, with the optimization objectives of maximizing the comprehensive performance evaluation factor, minimizing the pressure drop, and minimizing the maximum temperature of the heating surface, a non-dominated sorting genetic algorithm with an elitist strategy is used to solve the multivariate regression model for multi-objective optimization, obtaining the Pareto optimal frontier solution set, specifically including: The first objective function is to maximize the comprehensive performance evaluation factor, the second objective function is to minimize the pressure drop, and the third objective function is to minimize the maximum temperature of the heating surface. A parent population consisting of individuals corresponding to multiple sets of key structural parameters is initialized, and algorithm parameters such as the number of iterations, crossover probability, and mutation probability are set.

[0037] In each iteration, the function values ​​of the first objective function, the second objective function, and the third objective function are determined according to the key structural parameters corresponding to each individual. The individuals in the current population are then sorted into different non-dominated levels, where individuals at lower levels are not completely dominated by other individuals in any of the three objectives.

[0038] For individuals within the same non-dominated level, their crowding distance in the target space consisting of the first objective function, the second objective function, and the third objective function is determined to measure the density of the individual distribution. Parent individuals with low non-dominated levels and large crowding distances are selected from the current population through a tournament selection strategy for crossover and mutation operations to obtain the offspring population.

[0039] The parent and offspring populations are merged to obtain a mixed population. The mixed population is then sorted by non-dominated hierarchy and crowding distance is calculated. Based on the principle of increasing non-dominated hierarchy and decreasing crowding distance within the same hierarchy, multiple individuals are selected to form the next generation of parent populations. This iterative process is repeated until a preset termination condition is met. All individuals in the final parent population with a non-dominated hierarchy of level one are then used as the Pareto optimal front solution set.

[0040] Preferably, the Pareto optimal frontier solution set is subjected to multi-attribute decision analysis using the entropy-weighted TOPSIS method to determine the relative fit of each candidate parameter combination, and the candidate parameter combination with the highest relative fit is taken as the optimal compromise structure parameter combination, specifically including: Each solution in the Pareto optimal front set is treated as an alternative. An original evaluation matrix is ​​constructed using the comprehensive performance evaluation factor, pressure drop, and maximum heating surface temperature corresponding to each alternative as evaluation indicators. The original evaluation matrix is ​​then dimensionless to construct a standardized decision matrix, where the comprehensive performance evaluation factor is treated as a positive indicator, and the pressure drop and maximum heating surface temperature are treated as negative indicators. The objective weights of each evaluation indicator are determined using the information entropy method, and a weighted standardized decision matrix is ​​constructed.

[0041] Based on the weighted standardized decision matrix, the positive and negative ideal solutions for each evaluation index are determined, and the Euclidean distance from each alternative to the positive and negative ideal solutions is also determined. The relative fit of each alternative is determined based on the Euclidean distance, and the parameter combination corresponding to the alternative with the highest relative fit is taken as the optimal compromise structural parameter combination.

[0042] A specific embodiment of the present invention is provided: 1. Characteristics of microchannel heat sink structures with coupled ribbed cavities.

[0043] Leveraging the complementary advantages of fins and cavities, a novel jet microchannel structure with coupled fin-cavity combinations was designed based on traditional jet microchannels. This structure aims to improve temperature uniformity and heat transfer performance while balancing pressure drop. Considering that the bottom of the microchannel is in direct contact with the heat source substrate, resulting in greater heat dissipation than the upper channels, a fin-cavity combination is introduced near the bottom of the channel to balance pump power loss while enhancing heat transfer. Figure 1 As shown, this is a microchannel heat sink with coupled ribbed cavities (taking a rectangular ribbed cavity assembly as an example), with a heat flux density of 700 W·cm² on the bottom heating surface. -2 The heat sink solid region is made of copper, and the cooling medium is deionized water.

[0044] Since the shape of the rib cavity assembly affects heat dissipation performance, three novel rib cavity assembly structures were designed: rectangular, elliptical, and spindle-shaped rib cavity structures, such as... Figure 2 As shown. The main structural parameters include the rib height ( h 1) Rib width ( a 1) Rib width ( b 1) Cavity height ( h 2) Cavity length ( a 2) Cavity width ( b 2) Fix the distance between two adjacent rib cavities ( l s The value is 800μm.

[0045] 2. Analysis of the heat dissipation characteristics of the rib cavity combination microchannel.

[0046] To mitigate the impact of rib shape on microchannel heat sink flow and heat transfer, ribs of different shapes were designed with identical cross-sectional areas. The fixed rib height was set to 900 μm, the rib width to 200 μm, the cavity height to 900 μm, and the cavity width to 100 μm. Only the rib length and cavity length were varied to ensure that combinations of ribs with different shapes had the same cross-sectional area. Table 1 lists the detailed geometric dimensions.

[0047] Impinging jets are a flow problem with unique flow characteristics, often exhibiting turbulent flow accompanied by backflow. Compared to standard... k - ε Model, RNG k - ε The model can take into account effective turbulent transport under different Reynolds numbers, and its simulation effect is the best. It can provide more accurate results when simulating impinging jets.

[0048] Therefore, RNG is adopted. k - ε Numerical simulations were conducted using the model. To comprehensively analyze the heat transfer and flow properties of the double-layer jet microchannel, this invention introduces a comprehensive performance evaluation factor. PEC To evaluate overall performance:

[0049] .

[0050] Table 1. Dimensions of Rib-Cavity Assembly Figure 3 The enhanced heat transfer characteristics of the three ribbed cavity combination structures were compared. As shown in the figure, for the microchannel heat sink with coupled ribbed cavity combinations, the periodic contraction-expansion of the flow channel caused by the ribbed cavity region effectively diverts and separates the fluid, enhancing local energy dissipation and resulting in a significantly higher pressure drop level than that of smooth microchannels. Specifically, the Δ... P The maximum values ​​indicate that the rectangular rib cavity has a stronger blocking effect on the mainstream, and stronger local vortex dissipation and drag. Under the same mass flow rate conditions, the three rib cavity combinations have the highest values. Nu avg Both are higher than the heat sink of a smooth microchannel, indicating that the coupled finned cavity combination can effectively enhance the convective heat transfer process within the channel, improve the heat transfer efficiency from the solid wall to the cooling medium, and thus reduce the overall temperature level of the heating surface. Simultaneously, combined with... Figure 3 As shown in (d), the temperature uniformity of the heating surface is significantly improved after the coupled fin cavity combination. This indicates that the coupled fin cavity combination can effectively alleviate temperature rise and make the temperature distribution more uniform.

[0051] from Figure 3 As shown in (c), although the finned cavity combination improves heat transfer performance and temperature distribution, the resulting increase in pressure drop weakens the advantages brought by the enhanced heat transfer to some extent, thus compromising the microchannels of the coupled finned cavity combination. PEC The overall performance is lower than that of smooth microchannels. This indicates that the application of ribbed cavity combinations in bilayer jet microchannels requires a comprehensive trade-off between enhanced heat transfer and flow resistance. Different ribbed cavity structures exhibit significant differences in the degree of enhanced heat transfer and resistance loss. Therefore, further optimization of the ribbed cavity combinations is needed to minimize pressure drop while improving heat dissipation capacity. Since the spindle shape exhibits the highest overall performance among the three ribbed cavity combinations, subsequent analysis and optimization of the structural parameters of the microchannel heat sink coupled with the spindle-shaped ribbed cavity combination will be conducted.

[0052] 3. Parameter optimization design of rib cavity combination structure based on Taguchi method.

[0053] Here, the Taguchi method was used to optimize the structural parameters of the rib-cavity assembly, and numerical simulations were performed based on the orthogonal experimental design table. To further optimize the design, the simulation results were analyzed to determine the significance ranking of the effects of each structural parameter on the average Nusselt number, pressure drop, overall performance, and maximum temperature of the heating surface of the microchannel.

[0054] 3.1 Taguchi Experiment Design and Scheme Construction.

[0055] The Taguchi method is an efficient, rapid, and economical design approach. By utilizing orthogonal arrays for experimental design, it can independently evaluate the effects of each influencing factor while eliminating interference from other factors. Furthermore, this method can analyze the influence patterns and contribution rates of each factor on the target parameter through systematic Design of Experiments (DOE), thereby determining the optimal combination of levels for each factor in the discrete space. Compared to other optimization methods, the Taguchi method is efficient and robust, saving time and resources, and yielding stable and reliable design results.

[0056] Therefore, the Taguchi method was chosen for this research. The signal-to-noise ratio (SNR) was introduced into the Taguchi method analysis. SNR The signal-to-noise ratio (SNR) is used to evaluate the robustness of a system; the higher the SNR, the better the performance of this evaluation metric.

[0057] There are three different types of quality characteristics for the target design: high potential, low potential, and low potential. The goal of this invention is to achieve higher heat transfer performance with a lower pressure drop. Therefore, the high potential in equation (1) will be used as a comprehensive performance evaluation factor. PEC ) and average Nusselt number ( Nu avg In equation (2), the small value will be used for the pressure drop (Δ). P ) and the highest temperature of the heating surface ( T max ).

[0058] (1) (2) In the formula, n It is the number of experiments; y i These are the results of each experiment.

[0059] Furthermore, since there are multiple different influencing factors between any two combinations, the parameter average signal-to-noise ratio (SNR) is defined. SNR avg Data processing is performed. SNR avg It is the average signal-to-noise ratio of the same impact factor at the same level, and the formula is: (3) In the formula, i It is the impact factor of the research; j It is the level of the factor; n It represents the number of cases and the corresponding impact factor levels studied in the orthogonal experimental table.

[0060] Using the range of each influencing factor ( R The factor's influence is represented by (), and its definition is as follows: (4) To clearly demonstrate its significance, a contribution rate ( ) is defined. CR )as follows: (5) In the formula, m This represents the number of influence factors.

[0061] To ensure the comparability of different schemes under the same boundary conditions and to complete the multi-parameter combination study within limited computational resources, the inlet mass flow rate was selected as the median value of 2.952 g·s⁻¹ within a typical range. -1 This was conducted as a representative working condition. Six different structural parameters were selected, A: rib height ( h 1) B: Rib length ( a 1) C: Rib width ( b 1) D: Cavity length ( a 2) E: Cavity width ( b 2) F: Cavity height ( h 2) Furthermore, considering five levels for each influencing factor, the orthogonal factors and levels are shown in Table 2. The ribs are distributed on the bottom surface of the lower channel, and the height is calculated based on the bottom surface. Table 3 presents the six-factor, five-level orthogonal experimental design, selecting... L 25 (5) 6 Orthogonal arrays reduce the number of trials from 5 for full factorial trials. 6 The number of operations has been reduced from 15,625 to the current 25, greatly reducing computing resources and time.

[0062] Table 2 Orthogonal Factors and Levels Table 3 L 25 (5) 6 Orthogonal experimental design table 3.2 Significance analysis of the impact of geometric parameters on performance.

[0063] Figure 4 The microchannel heat sink is given Nu avg Δ P , PEC as well as T max The graph shows the average signal-to-noise ratio (SNR) variation curves at different levels of the six structural parameters (factors). The steeper the slope of the curve, the more significant the effect of that factor on the response; a higher SNR value corresponds to better performance.

[0064] Depend on Figure 4 As shown in (a), with the increase of factor A... Nu avg The curve shows a significant monotonically increasing trend with the steepest slope, indicating that rib height significantly affects... Nu avg The dominant factor. Factor B on Nu avg The effect is manifested as increasing rib length. Nu avg The curve shows a slow upward trend, indicating that increasing the fin length can also improve heat transfer capacity, but its effect is weaker than that of fin height. Factors C and E affect... Nu avg The effects of factors D and F all show a trend of first increasing and then decreasing, with a peak at the middle level. This indicates that appropriately increasing the fin width and cavity width can effectively improve heat transfer performance, but there is an optimal range. Exceeding this range and continuing to increase the width will actually lead to a decrease in heat transfer capacity. Nu avg The impact exhibits fluctuating changes. In summary, when Nu avg When the maximum value is obtained, the optimal parameter combination is A5B5C4D3E3F1.

[0065] Depend on Figure 4 From (b) we can see that Δ P The overall curve change trend and Nu avg Broadly the opposite. Each factor affects Δ P The impact analysis is as follows: Δ P The pressure drop is highly sensitive to factor A; as the fin height increases, the pressure drop increases significantly, and this increase gradually rises. This is because increasing the fin height significantly reduces the fluid flow area, thereby increasing the fluid flow resistance. Factor C also affects Δ... P This had a significant impact. The wider the fin, the greater the pressure drop, because the increased fin width enhances fluid turbulence, leading to increased flow resistance. The signal-to-noise ratios of factors D and F increased significantly with increasing level, indicating that appropriately increasing the cavity length and height can effectively reduce the pressure drop. In summary, making Δ P The lowest optimal parameter combination is A1B2C1D5E1F3.

[0066] Depend on Figure 4 As can be seen from (c) in the text, factor C is... PEC The most significant factor, its impact on PEC The trend of the influence curve and its effect on Nu avg The trend is opposite to that of Δ. P The trends are roughly the same. This indicates that while increasing the fin width can improve heat transfer capacity, its pressure drop also increases accordingly. PEC The signal-to-noise ratio (SNR) decreases sharply as the level of factor A increases, indicating that the pressure drop caused by increasing the fin height exceeds the improvement in heat transfer capacity, leading to... PEC Decrease. As the number of levels for factor D and factor F increases, PEC The upward trend indicates that increasing the cavity length and cavity height can effectively improve... PEC Factor B and Factor E on PEC The curve showing a small change in influence indicates that the rib length and cavity width have little effect on... PEC The impact is relatively weak. Therefore, PEC The optimal parameter combination for the maximum value is A1B2C1D5E3F5.

[0067] Depend on Figure 4 From (d) we can see that, T max The curve change trend and Nu avg The curves change roughly the same way, which is because... Nu avg This represents the overall heat transfer capacity of the microchannel heat sink; the stronger the overall heat transfer capacity, the lower the heating surface temperature. T max It is related not only to the overall heat exchange capacity, but also to the local heat exchange intensity, therefore T max and Nu avg The curve trends are generally consistent, but there are also differences. As can be seen from the graph, T max The optimal parameter combination when the minimum is reached is A5B5C4D3E3F4.

[0068] Figure 5 The contribution rate of each factor level to each performance level was used to further quantify the influence of each factor. As shown in the figure, for... Nu avg The most significant influencing factor is the rib height, among which CR It was 34.6%; followed by rib width, CR It was 22.0%; the influence of cavity width and cavity height was relatively small. CR All are 7.8%. Therefore, each structural parameter affects... Nu avg The significance of the impact is ranked as follows: h 1> b 1> a 2> a 1> b 2> h 2.

[0069] For Δ P The rib height and rib width are also extremely sensitive. CR The figures are 35.3% and 31.0% respectively; Factor D and Factor F affect Δ P The contributions of factors B and E were 13.4% and 10.4% respectively, indicating that the influence of cavity length and cavity height on pressure drop is at a moderate level; factors B and E have a moderate influence on Δ. P The contributions of fin length and cavity width are relatively low, at 7.2% and 2.7% respectively, indicating that the influence of fin length and cavity width on voltage drop is not significant. Therefore, for Δ P The parameters with the greatest impact are ranked as follows: h 1> b 1> a 2> h 2> a 1> b 2.

[0070] for PEC Factor C contributed 37.7% to this, further indicating that rib width is a significant factor. PEC The most significant factor. Rib height. PEC Its contribution rate is also very high. CR The contribution rate was 31.6%. Factors D and F contributed 11.1% and 14.6% respectively, indicating that cavity length and height can effectively improve overall performance. Factors B and E... PEC The contribution rates were the smallest, at 3.3% and 1.8% respectively, indicating that the rib width and cavity width had a relatively weak impact on the overall performance. PEC The parameters with the greatest impact are ranked as follows: b 1> h 1> h 2> a 2> a 1> b 2.

[0071] Finally, for T max The rib height had the most significant impact; followed by the rib width, which contributed 23.7%. The cavity width and cavity height also had a significant impact. T max The impact of these factors was relatively small, with contribution rates of 8.8% and 6.8%, respectively. Therefore, the ranking of the factors' significant impact on these factors is related to their respective effects on the overall significance of the factors. Nu avg They are the same, both being: h 1> b 1> a 2> a 1> b 2> h 2.

[0072] Therefore, combining Figure 4 and Figure 5 It is evident that the rib height, rib width, cavity length, and cavity height have a significant impact on the heat transfer performance of the microchannel heat sink, while the rib length and cavity width have a relatively low impact. To conserve computational resources, the rib length and cavity width will be fixed in subsequent studies, and optimization will only be performed on the other four structural parameters. Figure 4 As shown in (d), when the rib length is 462 μm and the cavity width is 100 μm... PEC Therefore, in subsequent studies, the length of the rib was fixed at 462 μm and the width of the cavity was 100 μm.

[0073] 4. Research on multi-objective optimization based on response surface methodology and genetic algorithm.

[0074] The Taguchi method can be used to analyze the influence of multiple parameters on the performance of microchannel heat sinks and obtain the optimal combination of structural parameters for a single performance objective. However, it is difficult to obtain the correlation between heat transfer and flow performance and structural parameters, and when the optimization process involves multiple optimization objectives, the Taguchi method is difficult to satisfy all objectives. The ideal design objective of a microchannel heat sink is to achieve the strongest heat transfer performance while having the lowest pressure loss. Therefore, this invention adopts a multi-objective optimization method combining response surface methodology and genetic algorithm to optimize the structure of the rib cavity combination.

[0075] 4.1 Experimental design and numerical calculation based on response surface methodology.

[0076] The key point here is to select the rib height ( h 1) Rib width ( b 1) Cavity length ( a 2) and cavity height ( h 2) Four key design parameters, with PEC Δ P ,and T maxTo optimize the objectives, this study investigates the nonlinear relationships between various design variables. Box-Behnken design (BBD) is used to construct the response surface methodology. As analyzed earlier, a fin height of 900 μm results in severe flow channel blockage, significant flow resistance and pressure loss, and poor overall performance, constituting an extreme condition. Therefore, the fin height range is adjusted to 400 μm–800 μm, while other structural parameters remain unchanged. The response surface methodology design is shown in Table 4.

[0077] Table 4 Response Surface Design Table The experiment at the selected center point was repeated 5 times, resulting in 29 experimental groups using Design Expert software. Based on the sample points generated by BBD, the microchannel heat sink was numerically simulated using Fluent. Finally, the response values ​​were obtained through detailed calculations and data processing, as shown in Table 5.

[0078] Table 5 Response Surface Experimental Design and Results 4.2 Establishment and Fitting Analysis of Response Surface Model.

[0079] This section utilizes simulated data for regression analysis, fitting a response surface polynomial model using the least squares method, and conducting significance testing through analysis of variance. In the analysis of variance, the coefficient of determination is adjusted. R 2 The coefficient of determination (CCD) measures how well a model fits the sample data and adjusts for the number of independent variables. A value closer to 1 indicates a better fit. R 2 The coefficient of determination reflects the model's predictive ability on unknown data. To demonstrate that the model has robust predictive accuracy, it is used to indicate that the model has predictive accuracy. R 2 With Adjusted Determination Coefficient R 2 The difference should be small (usually less than 0.2). When a certain model term's P A value less than 0.05 indicates a significant impact on the response variable at a 95% confidence level, thus making the model term highly relevant. When constructing a mathematical model, terms with no significant impact can be simplified and excluded, while those with significant impact should be retained for further analysis.

[0080] Based on the data in Table 5, the microchannel heat sink Δ was obtained through fitting. P The regression equation is shown in equation (6). The analysis of variance results for the pressure drop show... PA value <0.0001 indicates a high degree of match between the mathematical model and the simulation data. For Δ P model items h 1. b 1. a 2. h 2. h 1 b 1. h 1 h 2. b 1 a 2. b 1 h 2 and b 12 of P The values ​​are all below 0.05, indicating they are significant model terms. Figure 6 For Δ P The chart comparing predicted and actual values ​​shows that the data points are located at... y = x Near the reference line, it indicates that the established model exhibits a high degree of fit and predictive accuracy for the response variable. After calculation and analysis, the determination coefficient of the pressure drop was obtained. R 2 The coefficient of determination is 0.9793, adjusted to [value missing]. R 2 The coefficient of determination is 0.9678. R 2 The coefficient of determination is 0.9325, adjusted accordingly. R 2 Coefficient of determination for prediction R 2 The difference is much less than 0.2, which also indicates that the fitted quadratic regression model has a strong correlation.

[0081] (6) T max The results of the analysis of variance show P A value <0.0001 indicates a high degree of match between the mathematical model and the simulation data. This was determined through calculation and analysis. T max coefficient of determination R 2 The coefficient of determination is 0.9849, adjusted. R 2 The coefficient of determination is 0.9751. R 2 It is 0.9378. Combining the coefficient of determination and... Figure 7 middle T max The comparison between predicted and actual values ​​shows that the established quadratic regression model has high accuracy in fitting and predicting the maximum temperature of the heating surface. Its regression equation is shown in equation (7).

[0082] (7) PEC The regression equation is shown in equation (8). The analysis of variance results show that the model corresponds to... P A value less than 0.0001 indicates that the regression model is highly significant. For PEC model items h 1. b 1. a 2. h 2. b 1 h 2. h 12 and b 12 of P The values ​​are all below 0.05, indicating significant model terms. Further calculation and analysis revealed the model's coefficient of determination. R 2 The coefficient of determination is 0.9872, adjusted accordingly. R 2 The coefficient of determination is 0.9776. R 2 The coefficient of determination was 0.9482, and all three were close to 1, while the predictive coefficient of determination was also high. R 2 With Adjusted Determination Coefficient R 2 The difference is much less than 0.2, indicating that the model has not overfitted. Combined with... Figure 8 middle PEC The comparison chart between predicted and actual values ​​shows that the model has high fitting accuracy and a very high degree of agreement with the actual data, verifying that the regression model has good applicability.

[0083] (8) 4.3 Parameter optimization solution based on multi-objective genetic algorithm.

[0084] NSGA-II is a typical multi-objective optimization algorithm capable of handling multiple conflicting optimization objectives simultaneously. Its core idea is to maintain the quality and diversity of the solution set based on non-dominated sorting and crowding distance.

[0085] In the structural optimization of ribbed heat sinks, enhancing heat transfer and reducing flow resistance are usually mutually restrictive. Therefore, the objectives of this invention are threefold: to maximize... PEC Minimize Δ P , T max . PEC Corresponding to the overall performance of the system, Δ P Corresponding to resistance and energy consumption, Tmax Corresponding thermal reliability. Based on the above regression model, the results of the multi-objective formula fitted by RSM are introduced into NSGA-II for multi-objective optimization. The multi-objective optimization model is shown in equation (9). Select the rib height ( h 1) Rib width ( b 1) Cavity length ( a 2) Cavity height ( h 2) As decision variables, the ranges of the four decision variables are 400μm~800μm, 100μm~300μm, 362μm~762μm, and 300μm~900μm, respectively. The initial population size is set to 50, the maximum number of iterations is 500, the crossover probability is 0.8, and the mutation probability is 0.2.

[0086] (9) In the formula, y 1( x )for PEC function; y 2( x ) is Δ P function; y 3( x ) is the Tmax function; x iL and x iU The first i The lower and upper limits of each factor. i =1, 2, 3, 4, where the lower and upper limits of the factors are the lower and upper limits of the range of variation of the four decision variables.

[0087] Entropy weighting is used to determine the weights of evaluation indicators. TOPSIS (Technique for Order Preference by Similarity to an Ideal Solution) is a classic multi-attribute decision analysis method used to rank and select the best among finite solutions. Its core idea is to evaluate the relative merits of each solution by measuring its relative distance from the ideal optimal solution and the ideal worst solution. The closer a solution is to the ideal solution and the farther it is from the negative ideal solution, the better its overall performance. This invention combines entropy weighting with TOPSIS to evaluate the performance of microchannel heat sinks. The steps of finding a compromise solution using the entropy weighted TOPSIS algorithm are as follows:

[0088] (1) Represent the initial data as the original evaluation matrix.

[0089] (10) In the formula, This is the original evaluation matrix; aij For the first i The first of the alternative options j Each evaluation indicator value, i =1, 2, ..., m ; j =1, 2, ..., n ; m Indicates the total number of alternative options; n This indicates the total number of evaluation indicators.

[0090] (2) Construct a standardized decision matrix.

[0091] The matrix is ​​made dimensionless, and each index is normalized. The dimensionless formulas corresponding to each index type are shown in equations (11) and (12).

[0092] Positive indicators: (11) Contrarian indicator: (12) In the formula, b ij These are the elements in the standardized decision matrix.

[0093] (3) Determine the entropy weight of the evaluation index.

[0094] (13) (14) (15) (16) Construct a weighted decision matrix.

[0095] (17) (4) Calculate the ideal solution A * and negative ideal solution A The calculation method is as follows: (18) (19) (5) Calculate the positive ideal solution for each alternative solution. A * and negative ideal solution A The distance is given by the following formula: (20) (twenty one) (6) Calculate the relative fit between each alternative and the ideal solution. The closer the value is to 1, the better the overall performance. The formula is as follows:

[0096] (twenty two) The Pareto optimal frontier and its decision point selection were obtained using MATLAB software, as follows: Figure 9 As shown in the figure, the red dots represent the Pareto optimal solution set, and the blue dots represent the optimal compromise solution. The figure reveals a conflict among the three objective functions; each point on the Pareto front represents a trade-off. Since decision-makers need to obtain the optimal compromise solution based on specific requirements in practical applications, this invention employs the entropy-weighted TOPSIS method to sort the Pareto optimal solution set, thereby determining the optimal compromise solution.

[0097] Figure 10 The relative fit of each solution is given. O i For the first i The relative fit of each solution is a factor, with a higher value indicating a better solution. As shown in the graph, group 40 has the highest value, meaning this solution is closer to the ideal solution and is therefore optimal. The optimal compromise solution obtained using the TOPSIS method corresponds to... Figure 9 The blue marker indicates that, under this parameter combination (rib height 638 μm, rib width 164 μm, cavity length 412 μm, cavity height 900 μm), the comprehensive performance evaluation factor of the microchannel heat sink is 1.06, the pressure drop is 16.13 kPa, and the highest temperature of the heating surface is 340.19 K.

[0098] 4.4 Optimization results and performance analysis.

[0099] To verify the accuracy of the Pareto optimal solution prediction results, the optimized structural parameters were substituted into Fluent software for CFD numerical simulation, and the simulated values ​​were compared with the predicted values. Detailed data and error analysis are shown in Table 6. As can be seen from Table 6, PEC Δ P and T max The relative errors between the CFD simulation values ​​and the predicted values ​​were 0.96%, 4.57%, and 0.05%, respectively, with each error not exceeding 5%. This result indicates that the predicted values ​​and CFD simulation values ​​are in good agreement, demonstrating good prediction accuracy.

[0100] Table 6 Comparison of Simulated Values ​​and Predicted Values To further analyze the optimal structure obtained through genetic algorithm and entropy-weighted TOPSIS optimization, the Pareto optimal solution was compared with the two structures mentioned above in terms of flow and heat transfer performance, compared to the smooth double-layer jet microchannel and the initial rib cavity combination. The results are shown in Table 7. The results show that the optimized microchannel heat sink has a significantly improved heat dissipation capacity compared to the smooth double-layer jet microchannel heat sink. Specifically, Nu avg It increased by 41.21%. T max It dropped by 20.09K. T b It dropped by 16.2K. Despite Δ P It increased by 128.07%, but PEC The 7% improvement indicates that overall heat transfer performance was enhanced at the cost of acceptable flow resistance. Secondly, compared to the initial spindle-shaped finned microchannel heat sink, the optimized microchannel heat sink showed a significant reduction in pressure drop (49.17%), although heat dissipation performance decreased slightly. Nu avg It decreased by 3.76%. T max and T b They increased by 1.77K and 1.56K respectively. However, due to the significant reduction in flow resistance, PEC It increased by 20.22%.

[0101] Table 7 Comparison of Pareto optimal solution and initial solution Figure 11 and Figure 12 The images show velocity and temperature distribution contour maps for the three microchannel structures.

[0102] Depend on Figure 11 As shown in (a), the fluid flow in a smooth channel is relatively uniform, with weak local disturbances, resulting in low flow resistance. However, due to the weak disturbances during fluid flow, the heat dissipation capacity is insufficient. Figure 12 In (a), the heating surface has a distinct high-temperature zone with uneven temperature distribution.

[0103] like Figure 11 As shown in (b), after the initial ribbed structure was added, the periodic contraction-expansion of the ribbed cavity in the channel led to significant local flow disturbances. Furthermore, a high-speed flow region appeared in the center of the middle channel, and the shear instability caused by the local velocity gradient generated eddies, enhancing fluid mixing and thus improving the local heat transfer capacity. Combined with... Figure 12 As shown in (b), its heat exchange performance is significantly improved and the temperature of the heating surface decreases overall.

[0104] Depend on Figure 11 As shown in (c), the optimized ribbed structure exhibits a more uniform fluid flow distribution, reduced local high-speed regions, and decreased local pressure loss. Simultaneously, combined with... Figure 12 As shown in (c), compared with the smooth channel, the optimized heating surface has a more uniform temperature distribution and the temperature is effectively controlled. This indicates that the optimized ribbed structure effectively balances heat transfer performance and flow resistance, resulting in the best overall performance.

[0105] The embodiments described above are merely examples of several implementations of the present invention, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these modifications and improvements all fall within the scope of protection of the present invention.

Claims

1. A microchannel heat sink structure with coupled rib-cavity assembly, characterized in that, include: A microchannel substrate with a dual-layer flow channel structure includes: an inner microchannel, an outer microchannel, an upper microchannel, and a lower microchannel; The cooling medium in the inner microchannel flows in from the first inlet in the upper microchannel, flows through the lower microchannel and flows out from the first outlet in the upper microchannel; the cooling medium in the outer microchannel flows in from the second inlet and the third inlet in the upper microchannel and flows out from the second outlet and the third outlet in the lower microchannel, respectively. The bottom surface of the lower microchannel is provided with a combination of multiple periodically arranged coupling rib cavity structures. The combination of coupling rib cavity structures is composed of alternating combinations of ribs protruding from the bottom surface of the lower microchannel and concave cavities, so as to form a periodic tapering-expanding cross section in the lower microchannel. The lower microchannel has a bottom heating surface on its outer side for simulating a chip heat source. The bottom heating surface is in close contact with the heat sink solid area to transfer heat to the cooling medium in the lower microchannel via the coupling rib cavity assembly structure.

2. The microchannel heat sink structure with coupled rib cavity assembly as described in claim 1, characterized in that, The ribs and cavities in the coupled rib-cavity combination structure have any one of the following geometric shapes in cross-section: rectangular, elliptical, or spindle-shaped.

3. The microchannel heat sink structure with coupled rib cavity assembly as described in claim 1, characterized in that, The height of the rib ranges from 400μm to 800μm, the width of the rib ranges from 100μm to 300μm, the length of the cavity ranges from 362μm to 762μm, and the height of the cavity ranges from 300μm to 900μm.

4. The microchannel heat sink structure with coupled rib cavity assembly as described in claim 1, characterized in that, The spacing between adjacent coupled rib cavity combination structures is 800 μm.

5. A design method for a microchannel heat sink structure based on a coupled rib cavity combination as described in any one of claims 1-2, characterized in that, include: Obtain the basic shape and initial structural parameters of the rib-cavity assembly, wherein the initial structural parameters include: rib height, rib length, rib width, cavity length, cavity width, and cavity height; An orthogonal experimental design table corresponding to the rib cavity assembly was constructed using the Taguchi method to determine the contribution rate of each initial structural parameter to the performance index, and the initial structural parameters with a contribution rate greater than a preset threshold were taken as key structural parameters. By constructing an experimental scheme corresponding to the key structural parameters using the response surface methodology and conducting numerical simulations, a multivariate regression prediction model is obtained to quantify the nonlinear mapping relationship between each key structural parameter and the heat sink's heat transfer-flow performance. With the optimization objectives of maximizing the comprehensive performance evaluation factor, minimizing the pressure drop and minimizing the maximum temperature of the heating surface, a non-dominated sorting genetic algorithm with an elite strategy is used to solve the multivariate regression model for multi-objective optimization, and the Pareto optimal front solution set is obtained. Each solution in the Pareto optimal front solution set corresponds to a set of alternative parameter combinations for the microchannel heat sink structure. The Pareto optimal front solution set is subjected to multi-attribute decision analysis using the entropy weight TOPSIS method to determine the relative fit of each candidate parameter combination. The candidate parameter combination with the highest relative fit is taken as the optimal compromise structural parameter combination. The optimal compromise structural parameter combination is used to design the microchannel heat sink structure of the coupled rib cavity combination.

6. The design method of a microchannel heat sink structure with coupled rib cavity assembly as described in claim 5, characterized in that, The orthogonal experimental design table corresponding to the rib-cavity assembly is constructed using the Taguchi method to determine the contribution rate of each initial structural parameter to the performance index, and the initial structural parameters with a contribution rate greater than a preset threshold are designated as key structural parameters. Specifically, this includes: The six initial structural parameters, namely rib height, rib length, rib width, cavity length, cavity width, and cavity height, are used as influencing factors, and multiple level values ​​are set for each influencing factor. A six-factor, five-level orthogonal experimental design table was constructed based on the aforementioned influencing factors and their level values. Numerical simulations were performed based on the orthogonal experimental design table to determine the average Nusselt number, pressure drop, comprehensive performance evaluation factor, and signal-to-noise ratio of the highest temperature performance index of the heating surface for each group of experiments. The range and contribution rate are determined based on the average signal-to-noise ratio of each influencing factor at different levels. Initial structural parameters with contribution rates greater than a preset threshold are taken as key structural parameters. The key structural parameters include: rib height, rib width, cavity length, and cavity height.

7. The design method of a microchannel heat sink structure with coupled rib cavity assembly as described in claim 5, characterized in that, The process involves constructing an experimental scheme corresponding to the key structural parameters using the response surface methodology and conducting numerical simulations to obtain a multiple regression prediction model for quantifying the nonlinear mapping relationship between each key structural parameter and the heat sink's heat transfer-flow performance. Specifically, this model includes: Based on the key structural parameters, a response surface experimental design is constructed using the Box-Behnken design method, wherein the range of values ​​of the key structural parameters is used as the design space, and the center point is set for repeated experiments. CFD numerical simulations were performed on each sample point in the aforementioned response surface methodology to obtain the corresponding pressure drop, maximum heating surface temperature, and response values ​​of the comprehensive performance evaluation factor. Based on the response value, a quadratic polynomial multiple regression prediction model is constructed by fitting the least squares method to characterize the relationship between the key structural parameters and the response value. The significance of the quadratic multiple regression prediction model was tested by analysis of variance, and insignificant terms with significance greater than a set threshold were removed to obtain the final multiple regression prediction model.

8. The design method of a microchannel heat sink structure with coupled rib cavity combination as described in claim 5, characterized in that, The optimization objectives are to maximize the comprehensive performance evaluation factor, minimize the pressure drop, and minimize the maximum temperature of the heating surface. A non-dominated sorting genetic algorithm with an elitist strategy is used to solve the multivariate regression model for multi-objective optimization, yielding the Pareto optimal frontier solution set, specifically including: Maximizing the comprehensive performance evaluation factor is taken as the first objective function, minimizing the pressure drop is taken as the second objective function, and minimizing the maximum temperature of the heating surface is taken as the third objective function; Initialize a parent population consisting of individuals corresponding to multiple sets of the key structural parameters, and set the algorithm parameters for the number of iterations, crossover probability, and mutation probability; In each iteration, based on the key structural parameters corresponding to each individual, the function values ​​of the first objective function, the second objective function, and the third objective function are determined respectively; and the individuals in the current population are sorted in a non-dominated order, dividing the individuals into different non-dominated levels, wherein the individuals at the lower level are not completely dominated by other individuals in the three objectives; For individuals within the same non-dominated level, their crowding distance in the target space composed of the first objective function, the second objective function, and the third objective function is determined to measure the density of the individual distribution; parent individuals with low non-dominated levels and large crowding distances are selected from the current population through a tournament selection strategy for crossover and mutation operations to obtain the offspring population; The parent population and the offspring population are merged to obtain a mixed population. The mixed population is then sorted by non-dominated order and crowding distance is calculated. Based on the principle of non-dominated level from low to high and crowding distance within the same level from large to small, multiple individuals are selected to form the parent population of the new generation. Repeat the iterative process until the preset iteration termination condition is met, and take all individuals in the final parent population whose non-dominant level is the first level as the Pareto optimal front solution set.

9. The design method of a microchannel heat sink structure with coupled rib cavity assembly as described in claim 5, characterized in that, The step involves performing multi-attribute decision analysis on the Pareto optimal frontier solution set using the entropy-weighted TOPSIS method to determine the relative fit of each candidate parameter combination, and selecting the candidate parameter combination with the highest relative fit as the optimal compromise structure parameter combination. Specifically, this includes: Each solution in the Pareto optimal front solution set is taken as an alternative scheme, and the comprehensive performance evaluation factor, pressure drop and maximum heating surface temperature of each scheme are used as evaluation indicators to construct the original evaluation matrix. The original evaluation matrix is ​​dimensionless to construct a standardized decision matrix, in which the comprehensive performance evaluation factor is treated as a positive indicator, and the pressure drop and the maximum temperature of the heating surface are treated as negative indicators. The objective weights of each evaluation index are determined using the information entropy method, and a weighted standardized decision matrix is ​​constructed. Based on the weighted standardized decision matrix, the positive ideal solution and the negative ideal solution of each evaluation index are determined, and the Euclidean distance from each alternative to the positive ideal solution and the negative ideal solution is determined. The relative fit of each alternative scheme is determined based on the Euclidean distance, and the parameter combination corresponding to the alternative scheme with the highest relative fit is taken as the optimal compromise structure parameter combination.

Images