Cold plate optimization design method and device
Through cold plate modeling, liquid-cooling simulation and multi-island genetic algorithm optimization design, the problem of rough cold plate design in immersion heat dissipation system is solved, the heat dissipation efficiency of cold plates is improved, and the refined development of liquid-cooling technology is adapted to the refined development of liquid-cooling technology.
Patent Information
- Application Number
- CN202410981458.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-22
- Publication Date
- 2024-11-12
- Estimated Expiration
- 2044-07-22
AI Technical Summary
In the existing immersion heat dissipation system, the cold plate design is too rough to effectively cope with the flow resistance and heat dissipation needs of high viscosity coolant, resulting in the inefficiency of the cold plate in directional centralized cooling.
Cold plate modeling tools and liquid-cooled simulation tools are used to generate cold plate models and simulation data, and combined with multi-island genetic algorithm to optimize the tooth height, tooth width, tooth spacing and coolant flow rate of cold plates to minimize cold plate cooling devices and immersion The temperature of the cooling device is generated and optimized for design parameter combinations to improve the heat dissipation efficiency of the cold plate.
The cooling and cooling efficiency of immersed liquid for directional centralized cooling of cold plates is significantly improved, ensuring that electronic equipment maintains a low working temperature under high loads, and adapting to the refined development needs of liquid cooling technology.
Smart Images

Figure CN118940432A_ABST
Abstract
Description
Technical Field
[0001] The present disclosure relates to the field of liquid cooling technology, and in particular to a cold plate optimization design method and device. Background Art
[0002] With the acceleration of digital transformation, the demand for data center and server computing power has surged, leading to increasingly prominent energy consumption and heat dissipation issues. Single-phase immersion systems in liquid cooling technology usually use cold plate radiators for directional centralized heat dissipation, and widely use spade-tooth cold plates designed by directly reusing pure cold plate servers. These designs are mostly based on the cooling characteristics of low-viscosity liquids (such as deionized water).
[0003] However, the current immersion cooling system uses a high-viscosity cooling liquid (such as fluorinated liquid, oil) to first flow through a cold plate to cool high-power devices such as the CPU and GPU, and then flows out of the cold plate and into the chassis to cool other lower-power devices such as hard drives and network cards.
[0004] Therefore, how to improve the heat dissipation efficiency of immersion liquid cooling for directional centralized cooling of the cold plate when optimizing the design of the spade-tooth cold plate of the immersion cooling system has become a problem that needs to be solved urgently. Summary of the invention
[0005] In view of this, the present disclosure provides a cold plate optimization design method and device to solve the problem of how to improve the heat dissipation efficiency of immersion liquid cooling for directional centralized cooling of the cold plate when optimizing the design of a spade-shaped cold plate of an immersion heat dissipation system.
[0006] In a first aspect, the present disclosure provides a cold plate optimization design method, the method comprising: using a cold plate modeling tool to generate a cold plate model set according to a sample set of initial design parameters of the cold plate;
[0007] Using a liquid cooling simulation tool, a liquid cooling simulation data set is generated according to the coolant type, coolant flow rate and cold plate model set, wherein the liquid cooling simulation data in the liquid cooling simulation data set includes: cold plate cooling device temperature, immersion cooling device temperature and system flow resistance;
[0008] Using a multi-island genetic algorithm, with the system pressure drop not lower than the pressure drop threshold as a constraint, the tooth height, tooth width, tooth spacing and coolant flow rate of the cold plate initial design parameter sample corresponding to the liquid cooling simulation data in the liquid cooling simulation data set as the optimization parameter variables, and minimizing the cold plate cooling device temperature and the immersion cooling device temperature in the liquid cooling simulation data as the first objective function, the optimal design parameter combination and the coolant flow rate of the optimal design parameter in the corresponding optimal design parameter combination are generated; the system pressure drop is determined based on the system flow resistance;
[0009] From the optimized design parameter combination, the optimized design parameter that meets the design requirements is selected as the target cold plate design parameter, and the target coolant flow rate corresponding to the target cold plate design parameter is determined according to the coolant flow rate of the optimized design parameter in the corresponding optimized design parameter combination.
[0010] The cold plate optimization design method provided in this embodiment can accurately evaluate the influence of different cold plate design parameters on the cold plate cooling device temperature, immersion cooling device temperature and system flow resistance through the liquid cooling simulation data generated by the liquid cooling simulation tool. The multi-island genetic algorithm is used for optimization to find the optimal design parameter combination corresponding to the target coolant flow rate that minimizes the cold plate cooling device temperature and the immersion cooling device temperature, and optimizes the determined target cold plate design parameters, thereby significantly improving the heat dissipation efficiency of the immersion liquid cooling of the cold plate directional centralized cooling, and ensuring that the cooled electronic equipment or system can maintain a low operating temperature even under high load.
[0011] In a second aspect, the present disclosure provides a cold plate optimization design device, the device comprising:
[0012] A model generation module, for generating a cold plate model set according to a sample set of initial design parameters of the cold plate by using a cold plate modeling tool;
[0013] A liquid cooling simulation module is used to generate a liquid cooling simulation data set using a liquid cooling simulation tool according to a coolant type, a coolant flow rate, and a cold plate model set. The liquid cooling simulation data in the liquid cooling simulation data set include: a cold plate cooling device temperature, an immersion cooling device temperature, and a system flow resistance;
[0014] A parameter optimization module is used to use a multi-island genetic algorithm, with the system pressure drop not being less than a pressure drop threshold as a constraint, with the tooth height, tooth width, tooth spacing and coolant flow rate of the cold plate initial design parameter sample corresponding to the liquid cooling simulation data in the liquid cooling simulation data set as optimization parameter variables, with minimizing the cold plate cooling device temperature and the immersion cooling device temperature in the liquid cooling simulation data as the first objective function, to generate an optimized design parameter combination and a coolant flow rate of the optimized design parameter in the corresponding optimized design parameter combination; the system pressure drop is determined based on the system flow resistance;
[0015] The parameter selection module is used to select the optimized design parameters that meet the design requirements from the optimized design parameter combination as the target cold plate design parameters, and determine the target coolant flow rate corresponding to the target cold plate design parameters according to the coolant flow rate of the optimized design parameters in the corresponding optimized design parameter combination.
[0016] In a third aspect, the present disclosure provides a computer device, comprising: a memory and a processor, the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the cold plate optimization design method of the first aspect or any corresponding embodiment thereof by executing the computer instructions.
[0017] In a fourth aspect, the present disclosure provides a computer-readable storage medium having computer instructions stored thereon, the computer instructions being used to enable a computer to execute the cold plate optimization design method of the first aspect or any corresponding embodiment thereof. BRIEF DESCRIPTION OF THE DRAWINGS
[0018] In order to more clearly illustrate the technical solutions of the specific embodiments of the present disclosure, the drawings required for use in the description of the specific embodiments will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present disclosure. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.
[0019] Figure 1a is an exemplary schematic diagram of an immersion liquid cooling server using a cold plate for directional centralized cooling according to an embodiment of the present disclosure;
[0020] Figure 1b is a flow chart of a cold plate optimization design method according to an embodiment of the present disclosure;
[0021] Figure 1c is a schematic diagram of a flow chart of modeling and simulation of a cold plate optimization design method according to an embodiment of the present disclosure;
[0022] Figure 1d is an exemplary schematic diagram of cold plate skiving tooth structural parameters according to an embodiment of the present disclosure;
[0023] Figure 1e is an exemplary schematic diagram of an optimized design parameter combination represented in the form of a Pareto curve obtained according to the cold plate optimization design method of an embodiment of the present disclosure;
[0024] Figure 2a is a flow chart of another cold plate optimization design method according to an embodiment of the present disclosure;
[0025] Figure 2b is a flow chart of an optimal solution for optimizing design parameters based on agent model training according to an embodiment of the present disclosure;
[0026] Figure 3 is a structural block diagram of a cold plate optimization design device according to an embodiment of the present disclosure;
[0027] Figure 4 It is a schematic diagram of the hardware structure of the computer device of the embodiment of the present disclosure. DETAILED DESCRIPTION
[0028] In order to make the purpose, technical solution and advantages of the embodiments of the present disclosure clearer, the technical solution in the embodiments of the present disclosure will be clearly and completely described below in conjunction with the drawings in the embodiments of the present disclosure. Obviously, the described embodiments are part of the embodiments of the present disclosure, rather than all the embodiments. Based on the embodiments in the present disclosure, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present disclosure.
[0029] In recent years, with the acceleration of the digital transformation process, the data generated has shown an exponential explosive growth, and the demand for computing power of data centers and servers has also continued to rise. In order to cope with the energy consumption and heat dissipation problems brought about by high-performance computing tasks and huge data processing, server liquid cooling technology has developed rapidly. Compared with traditional air cooling technology, liquid cooling technology has the advantages of low energy consumption, high heat dissipation, and low noise. It has obvious advantages in scenarios such as strong server computing power and high deployment density. The efficient heat dissipation of liquid cooling technology can effectively improve the efficiency and stability of server use, keep the hardware equipment within the normal operating temperature range, and improve the computing efficiency of the server.
[0030] Liquid cooling technology mainly consists of immersion cooling and cold plate cooling. Depending on whether the coolant undergoes phase change, these two liquid cooling methods can be divided into single-phase and two-phase methods. Single-phase immersion and single-phase cold plate technologies are currently widely used in data centers due to their advantages of lower cost, higher reliability, and easier maintenance. For single-phase immersion liquid cooling, in order to solve the problem of uneven heat dissipation, many studies have proposed directional and centralized cooling methods. Specifically, during the liquid circulation process, after the cold liquid enters the liquid cooling tank, it first flows through devices with higher heat dissipation requirements, such as CPU, GPU, etc., and then flows through devices with lower heat dissipation requirements in turn, thereby improving heat dissipation energy efficiency.
[0031] This design is currently widely used in designing cold plates on high-power devices such as CPUs and GPUs. The cold liquid first flows through the cold plate, and then flows through other devices after it flows out of the cold plate. Currently, for single-phase cold plates, in-line spade-tooth cold plates are widely used because of their simple structure and good heat dissipation effect.
[0032] In a specific example, see Figure 1a , Figure 1a The following is an exemplary schematic diagram of a server using cold plate directional centralized cooling and immersion liquid cooling in the related art. The main components in the server system include CPU, memory, hard disk and power supply. Figure 1aIn the example shown, since the CPU has a relatively high heat dissipation requirement, a cold plate is designed on top of the CPU. When the cold liquid enters the chassis from the refrigerant heat exchange unit, it first flows through the two cold plates, takes away the heat from the CPU, and then flows out of the cold plate into the chassis to immerse the remaining components with relatively low heat dissipation requirements, such as memory, hard disk, etc., thereby achieving the aforementioned directional and concentrated cooling of high-demand heat dissipation components, and then cooling the remaining devices.
[0033] In the process of using this cold plate for directional centralized cooling in a single-phase immersion liquid cooling system, the more widely used approach at this stage is to directly reuse the cold plate used in pure cold plate servers. Under certain working conditions, it is possible to fine-tune the cold plate flow channel to determine the final design solution while meeting the heat dissipation effect and flow resistance requirements.
[0034] However, as liquid cooling becomes more sophisticated, current optimization design methods for cold plates used in immersion are too crude, which is contrary to the development trend of liquid cooling. The reasons are as follows: 1. In the process of cold plate flow channel design for pure cold plate servers, the cooling medium used is generally deionized water, propylene glycol and other liquids. The viscosity of such liquids is relatively low. Therefore, in the process of designing the skived tooth flow channel, in order to improve the heat dissipation capacity of the cold plate, the skived teeth in the cold plate flow channel can be designed to be relatively dense. However, this design is not suitable for immersion systems, because the coolant used in immersion systems is generally fluorinated liquid or oil. The viscosity of such liquids is much higher than that of deionized water. If this design is still reused, there will be a problem of excessive flow resistance; in addition, when the liquid is different, in order to ensure the heat dissipation capacity and flow resistance at the same time, different requirements will be put forward for the height and thickness parameters of the skived teeth; 2. For pure cold plate servers, in the process of cold plate optimization design, the optimization target will only consider the temperature of the device bonded to the cold plate. After the cold plate radiator is used in the immersion system, in the optimization design process, in addition to considering the temperature of the device bonded to itself, the temperature of the other immersed components in the chassis should also be considered, that is, a new method needs to be proposed for the entire optimization process and the definition of the optimization target.
[0035] At present, relevant research institutions and enterprises are widely studying the use of cold plate radiators in single-phase immersion systems for directional centralized heat dissipation. The shovel-tooth cold plate design used is a direct reuse of the design of pure cold plate servers, that is, a cold plate design using plasma water as the cooling medium, or on this basis, some minor adjustments are made in combination with the characteristics of the coolant. There is currently no cold plate shovel-tooth optimization design method for high-power devices in immersion systems. The problem with this design is that although a cold plate design that meets the heat dissipation and flow resistance requirements can be obtained, it may not necessarily be the optimal design solution that balances the heat dissipation and flow resistance requirements in the comprehensive design of the immersion combined with the cold plate. At present, with the increasingly sophisticated development of liquid cooling technology, this rough design method cannot adapt to the development needs of liquid cooling technology. The current optimization design method for all shovel-tooth cold plates in the needle immersion system is still in a blank stage.
[0036] The disclosed embodiments provide a method for improving the heat dissipation efficiency of immersion liquid cooling for directional centralized cooling of the cold plate when optimizing the design of a shovel-toothed cold plate of an immersion heat dissipation system. The case proposes a cold plate optimization design method that can improve the heat dissipation efficiency of immersion liquid cooling for directional centralized cooling of the cold plate.
[0037] According to an embodiment of the present disclosure, an embodiment of a cold plate optimization design method is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.
[0038] In this embodiment, a cold plate optimization design method is provided, which can be used in the above-mentioned computer system, such as a mobile terminal such as a mobile phone, a tablet computer, a portable computer, etc., or a fixed terminal such as a desktop computer, etc. Figure 1b is a flow chart of a cold plate optimization design method according to an embodiment of the present disclosure, such as Figure 1b As shown, the process includes the following steps:
[0039] Step S101 , using a cold plate modeling tool, generating a cold plate model set according to a sample set of cold plate initial design parameters.
[0040] In this embodiment, the cold plate modeling tool can use thermal analysis software (ANSYS Icepak), thermal management design tools (Boyd SmartCFD), computer-aided design (CAD) software (such as SolidWorks, AutoCAD) and computational fluid dynamics (CFD) software. Cold plate modeling tools usually have powerful geometric modeling, meshing and physical field solving capabilities. The model framework can be determined by considering the design characteristics and functional requirements of the cold plate, and can include geometric shapes, size ranges, boundary conditions, etc.
[0041] The sample set of initial design parameters of the cold plate can be collected from various channels (such as design documents, experimental records, CAD drawings, etc.). These parameters may include the shape of the shovel teeth (such as angle, width, height), layout (such as number, spacing), material properties (such as thermal conductivity, specific heat capacity), working environment conditions (such as temperature, pressure), etc.
[0042] In some specific examples, a design parameter sample interval of the cold plate shovel teeth may be given, and then sampling may be performed according to the design parameter sample interval to generate an initial design parameter sample set of the cold plate.
[0043] Using the parametric design function in the cold plate modeling tool, the cold plate design parameters (such as the shape, size, layout, etc. of the shovel teeth) can be set as variable parameters. Afterwards, these parameters can be changed according to the sample set of the initial cold plate design parameters to generate a series of cold plate models.
[0044] In a specific example, considering the automatic modeling of the cold plate based on each cold plate initial design parameter sample in the cold plate initial design parameter sample set and outputting the cold plate model set, a macro command script for the 3D modeling software can be established. The function that needs to be implemented here is the automatic modeling of the cold plate skived tooth flow channel, and then assembling it into the existing server model.
[0045] Step S102, using a liquid cooling simulation tool, generating a liquid cooling simulation data set according to the coolant type, coolant flow rate and cold plate model set.
[0046] In this embodiment, the liquid cooling simulation data in the liquid cooling simulation data set includes: a cold plate cooling device temperature, an immersion cooling device temperature, and a system flow resistance.
[0047] The type of coolant can be determined based on the application scenario and requirements of the cold plate. The coolant flow rate can be determined based on the requirements and design specifications of the cooling system, and can be a coolant flow rate interval or more than one coolant flow rate corresponding to the design parameters of the cold plate.
[0048] The temperature of the cold plate cooling device refers to the temperature of the device in the cooled object that performs complex calculations, processes large amounts of data, or requires high-power operation. It is also the temperature of the device that uses a cold plate to dissipate heat, such as the temperature of the CPU and GPU.
[0049] The immersion cooling device temperature refers to the temperature of other cooled devices in the cooled object except the cold plate cooling device, that is, the temperature of the cooled device immersed in the coolant, such as the temperature of the memory, hard disk, etc.
[0050] System flow resistance refers to the resistance encountered by a fluid when it flows in a system. It is an important parameter to measure the difficulty of fluid flow and can include the inlet and outlet flow resistance of the system. The size of the system flow resistance is affected by many factors, including the properties of the fluid (such as viscosity, density, etc.), the flow state (laminar or turbulent), the geometry of the pipe or flow channel, the roughness of the inner wall, and the flow rate.
[0051] Since the cold plate models in the cold plate model set reflect the actual physical structure of the cold plate, liquid cooling simulation tools such as Star CCM+, Solidworks Flow Simulation, Ansys Fluent, etc. can be used to simulate the application of each cold plate model in the liquid cooling system, thereby generating liquid cooling simulation data corresponding to each cold plate model and obtaining a liquid cooling simulation data set corresponding to the cold plate model set.
[0052] In a specific example, it is necessary to perform simulation calculations for all cold plate initial design parameter sample point models, and output the simulation calculation results of using the cold plate model to cool the server model. Therefore, it is necessary to establish a macro script for liquid cooling simulation software to execute the meshing and simulation calculation of the server model, complete the simulation of the cold plate initial design parameter sample point model for each cold plate, and then extract the parameters that need to be paid attention to in the simulation calculation results. Specifically, based on the configuration of the selected server model, the parameters that need to be paid attention to are the CPU temperature, memory temperature and system flow resistance. Here, the memory temperature is selected as the standard for measuring the temperature of the immersed device. The main reason is that among the immersed devices, the heat dissipation of the memory device is relatively large, which belongs to the high-risk item of heat dissipation. Therefore, the memory temperature is selected as the target.
[0053] Based on the general programming language for code programming, Python can be used to integrate the macro commands of each part of the software together and execute them step by step. At the same time, it is necessary to integrate the sample point matrix of the initial design parameters of the cold plate mentioned in the above steps, and call the macro command mentioned in the above step S101 based on the sample point matrix of the initial design parameters of the cold plate, obtain multiple different sample point models of the initial design parameters of the cold plate, and then execute the macro script of the above simulation calculation. At this point, the automatic modeling and simulation process ends. Specifically, you can refer to Figure 1c , Figure 1c A schematic diagram of the modeling and simulation process of the cold plate optimization design method according to an embodiment of the present disclosure is shown.
[0054] Step S103, using a multi-island genetic algorithm, with the system pressure drop not being lower than the pressure drop threshold as a constraint, the tooth height, tooth width, tooth spacing and coolant flow rate of the cold plate initial design parameter sample corresponding to the liquid cooling simulation data in the liquid cooling simulation data set as optimization parameter variables, and minimizing the cold plate cooling device temperature and the immersion cooling device temperature in the liquid cooling simulation data as the first objective function, to generate an optimized design parameter combination and the coolant flow rate of the optimized design parameter in the corresponding optimized design parameter combination.
[0055] In this embodiment, the system pressure drop is the pressure consumed by the coolant to overcome the resistance of the pipeline or cooling equipment during the flow process, and can be determined based on the system flow resistance.
[0056] The above-mentioned execution entity can use the Multi-Island Genetic Algorithm (MIGA), with the system pressure drop not lower than the pressure drop threshold as a constraint, and minimizing the cold plate cooling device temperature and the immersion cooling device temperature in the liquid cooling simulation data as the first objective function. By dividing the optimization parameter variable population into multiple "islands", the genetic algorithm is independently run on each island, and individuals are regularly exchanged between islands to generate optimization results, thereby improving the diversity and global adaptability of the optimization results.
[0057] In a specific example, based on an optimization method such as a multi-island genetic algorithm, when selecting optimization parameter variables, the following can be selected: Figure 1d The height h of the cold plate shovel teeth, the width s of the shovel teeth, the spacing l between the shovel teeth, and the flow rate v in the cold plate shovel tooth structural parameter diagram shown in the figure are optimization parameter variables. According to the server configuration, the optimization target can be selected as the CPU device temperature of cold plate cooling and the memory temperature of immersion cooling. The reason is: for the devices immersed in the server, the heat flux density of the memory is the highest, which is a high-risk item for heat dissipation. Therefore, in the process of setting the objective function for the immersed components, the memory temperature is used as the standard. In addition, due to the high viscosity of the cooling liquid used for immersion, while considering the heat dissipation effect of the cold plate, the flow resistance also needs to be considered. Therefore, in the optimization process, the pressure drop parameter of the entire system is used as a constraint variable to ensure that the pressure will not be lower than a value according to actual needs.
[0058] It should be noted that there may be 2, 4 or even more CPUs in the server, so there will be multiple CPU temperatures; there will be multiple chips on each memory stick, each chip has a temperature, so there will be multiple temperatures on each memory stick. Therefore, it is necessary to determine which temperature parameter to select as the optimization target. An optional solution is as follows: (1) Select the average value of each temperature as the measurement standard; (2) Select the chip in the worst cooling environment as the measurement standard. If the worst case has a large temperature margin, then there will be no cooling risk in other working conditions.
[0059] After that, the dual-objective optimization of the cold plate cooling device temperature, the immersion cooling device temperature, and the system pressure drop can be performed to obtain the Pareto curve. The horizontal axis of the Pareto curve can show the cold plate cooling device temperature, and the vertical axis can show the immersion cooling device temperature. Figure 1e In the schematic diagram of the Pareto curve shown in FIG. 1 , the abscissa and ordinate respectively show the central processing unit (CPU) temperature and the double data rate synchronous dynamic random access memory (DIMM) temperature. It can be seen that the temperature change trends of the two are opposite, that is, the lower the temperature of the cold plate cooling device, the higher the temperature of the immersion cooling device, or vice versa.
[0060] Therefore, in the optimal design parameter combination represented by the Pareto curve, the design parameters that meet the design requirements are selected as the target cold plate design parameters.
[0061] Step S104, selecting an optimized design parameter that meets the design requirements as a target cold plate design parameter in the optimized design parameter combination, and determining a target coolant flow rate corresponding to the target cold plate design parameter in the coolant flow rate of the optimized design parameter in the corresponding optimized design parameter combination.
[0062] In this embodiment, the cold plate cooling device temperature and the immersion cooling device temperature corresponding to the optimized design parameter combination have opposite change trends, that is, the lower the cold plate cooling device temperature, the higher the immersion cooling device temperature, or the higher the cold plate cooling device temperature, the lower the immersion cooling device temperature.
[0063] The design requirements may be determined according to the actual needs of the user, for example, may include a pre-set allowable temperature range of a cold plate cooling device and an allowable temperature range of an immersion cooling device.
[0064] Therefore, the execution subject can select the optimized design parameter in the optimized design parameter combination as the target cold plate design parameter according to the design requirements, and determine the target coolant flow rate corresponding to the target cold plate design parameter in the coolant flow rate corresponding to the optimized design parameter.
[0065] The cold plate optimization design method provided in this embodiment can accurately evaluate the influence of different cold plate design parameters on the cold plate cooling device temperature, immersion cooling device temperature and system flow resistance through the liquid cooling simulation data generated by the liquid cooling simulation tool. The multi-island genetic algorithm is used for optimization to find the optimal design parameter combination corresponding to the target coolant flow rate that minimizes the cold plate cooling device temperature and the immersion cooling device temperature, and optimizes the determined target cold plate design parameters, thereby significantly improving the heat dissipation efficiency of the immersion liquid cooling of the cold plate directional centralized cooling, and ensuring that the cooled electronic equipment or system can maintain a low operating temperature even under high load.
[0066] In some optional implementations of the present embodiment, the sample set of initial design parameters of the cold plate is determined based on the following steps: obtaining the structural parameter interval of the cold plate shovel teeth, the structural parameter interval including: tooth height interval, tooth width interval, tooth spacing interval; obtaining the coolant inlet flow interval; using the experimental design method, according to the structural parameter interval of the cold plate shovel teeth and the coolant inlet flow interval, generating the sample set of initial design parameters of the cold plate, the initial design parameter samples of the cold plate in the sample set of initial design parameters of the cold plate including: tooth height, tooth width, tooth spacing and coolant inlet flow.
[0067] In this implementation, during the optimization design of the cold plate, the selected coolant is first determined, and the optimization variables are selected as follows: Figure 1d The cold plate shovel tooth structure parameter schematic diagram shows the height h (unit of measurement mm) of the cold plate shovel tooth, the width s (unit of measurement mm) of the shovel tooth, and the spacing l (unit of measurement mm) between the shovel teeth. At the same time, since the heat dissipation effect and flow resistance of the cold plate are also related to the flow rate of the liquid, the flow rate and the shovel tooth structure are also interrelated, and jointly determine the performance of the cold plate, the optimization parameters also select the flow rate m (unit L / min) at the liquid inlet pipe, and establish a cold plate design parameter sample set based on some experimental design methods, such as optimal Latin square sampling, uniform design or orthogonal design, and establish a cold plate design parameter sample set. It can be understood that the experimental design method can also be other experimental design methods, such as completely randomized design, randomized block design, etc., and the present disclosure is not limited to this.
[0068] In a specific example, the range of h can be set to [3, 7], the range of s can be set to [0.2, 0.6], the range of l can be set to [0.2, 0.6], and the range of m can be set to [2, 6]. Based on the optimal Latin square sampling, 4 factors and 40 cold plate initial design parameter samples are established, which are summarized in Table 1 below. Here, 40 cold plate initial design parameter sample points are corresponding, and simulation needs to be completed for these 40 cold plate design parameter sample points.
[0069] Table 1-Sample set table generated based on optimal Latin square sampling:
[0070]
[0071]
[0072] In some optional implementations of the present embodiment, a multi-island genetic algorithm is used, based on a liquid cooling simulation data set, with a constraint that the system pressure drop determined based on the system flow resistance is not less than a pressure drop threshold, with the tooth height, tooth width, tooth spacing and coolant flow rate of the cold plate initial design parameter sample corresponding to the liquid cooling simulation data as optimization parameter variables, and with minimizing the cold plate cooling device temperature and the immersion cooling device temperature in the liquid cooling simulation data as the first objective function, to generate an optimization result, including: taking the tooth height, tooth width, tooth spacing and coolant flow rate of the cold plate initial design parameter sample corresponding to the liquid cooling simulation data as optimization parameter variables, executing an initialization population on the optimization parameter variable set corresponding to the liquid cooling simulation data set to obtain multiple sub-populations, the optimization parameter variables including: the tooth height, tooth width, tooth spacing and coolant flow rate of the cold plate initial design parameter sample; performing independent standard genetic operations and inter-island migration operations on each sub-population to obtain a genetic sub-population, and the standard genetic operations including: selection, crossover and mutation difference; taking minimizing the cold plate cooling device temperature and the immersion cooling device temperature in the liquid cooling simulation data as the first objective function, calculating the fitness of the new optimization parameter variable according to the cold plate cooling device temperature and the immersion cooling device temperature of the liquid cooling simulation data corresponding to the new optimization parameter variable in the genetic sub-population; determining the system pressure drop based on the system flow resistance of the liquid cooling simulation data corresponding to the new optimization parameter variable in the genetic sub-population; using the following two constraints: the fitness is greater than the fitness threshold, and the system pressure drop is not less than the pressure drop threshold, guiding the iteration to perform independent standard genetic operations and inter-island migration operations on each sub-population, recording the optimal optimization parameter variables of each iteration and the cold plate cooling device temperature and the immersion cooling device temperature corresponding to the optimal optimization parameter variables, and the system pressure drop determined based on the system flow resistance, until the predetermined number of iterations is reached or the objective function meets the convergence condition, and the optimized design parameter combination and the coolant flow rate of the optimized design parameter in the corresponding optimized design parameter combination are obtained.
[0073] In this implementation method, the weighted sum of the cold plate cooling device temperature and the immersion cooling device temperature can be used as the objective function; the inverse of the objective function can be used as the fitness function; and the fitness function can be used to calculate the fitness of the cold plate cooling device temperature and the immersion cooling device temperature of the liquid cooling simulation data corresponding to the new optimization parameter variables in the genetic subpopulation.
[0074] In some optional examples, the multi-island genetic algorithm can also be updated. Specifically, the target cold plate design parameters can be used to manufacture the target cold plate; the target cold plate is used to perform immersion cooling tests to obtain the actual tested cold plate cooling device temperature and the immersion cooling device temperature; based on the difference between the actual tested cold plate cooling device temperature and the immersion cooling device temperature and the cold plate cooling device temperature and the immersion cooling device temperature of the liquid cooling simulation data corresponding to the target cold plate design parameters, the multi-island genetic algorithm is updated to update the optimized design parameter combination.
[0075] In this embodiment, a cold plate optimization design method is provided, which can be used for the above-mentioned execution subject. Figure 2a is a flow chart of a cold plate optimization design method according to an embodiment of the present disclosure, such as Figure 2a As shown, the process includes the following steps:
[0076] Step S201 : using a cold plate modeling tool to generate a cold plate model set according to a sample set of cold plate initial design parameters.
[0077] In this embodiment, the parametric design function in the cold plate modeling tool can be used to set the design parameters of the cold plate (such as the shape, size, layout, etc. of the shovel teeth) as variable parameters. Afterwards, these parameters can be changed according to the sample set of initial design parameters of the cold plate to generate a series of cold plate models.
[0078] Please refer to step S101 of the embodiment shown in FIG. 1 for details, which will not be described in detail here.
[0079] Step S202: using a liquid cooling simulation tool, generating a liquid cooling simulation data set according to the coolant type, coolant flow rate and cold plate model set.
[0080] In this embodiment, since the cold plate model in the cold plate model set reflects the actual physical structure of the cold plate, liquid cooling simulation tools, such as Star CCM+, Solidworks Flow Simulation, Ansys Fluent, etc., can be used to simulate the application of each cold plate model in the liquid cooling system, thereby generating liquid cooling simulation data corresponding to each cold plate model and obtaining a liquid cooling simulation data set corresponding to the cold plate model set.
[0081] Please refer to step S102 of the embodiment shown in FIG. 1 for details, which will not be described again here.
[0082] Step S203, using the proxy model to expand the cold plate initial design parameter samples and the corresponding liquid cooling simulation data.
[0083] In this embodiment, the proxy model is to collect data points by sampling the original model a limited number of times, and then build a simpler and less computationally expensive mathematical model based on these data points. This model can quickly predict the output of the original model at unsampled points, thereby performing efficient search and optimization in the design space.
[0084] The types of proxy models may include but are not limited to: polynomial regression, radial basis function, Kriging model, artificial neural network, support vector machine, etc. Those skilled in the art may write proxy model training code by themselves, or perform proxy model training based on the proxy model training module provided in the commercial software, with the cold plate initial design parameter sample as input and the liquid cooling simulation data corresponding to the cold plate initial design parameter sample obtained by simulation calculation as output.
[0085] In a specific example, for 4 optimization variables, a total of 40 sample points of the initial design parameters of the cold plate are formed. During the optimization process, the optimization will be based on these 40 sample points. Therefore, these 40 sample points are discrete from each other. In some optimization cases, if these sample points are sparse, there will be certain deviations in the optimization process. In this case, if you want to obtain very accurate optimization results, you need to increase the number of sample points. For example, for these 4 optimization variables, you can design 100 sample points, which will increase the simulation workload and the optimization process will take a lot of time. Based on the proxy model training method, the existing sample points are fitted, so that an accurate sample set can be obtained while greatly reducing the simulation work and reducing the simulation time.
[0086] Step S204, using a multi-island genetic algorithm, with the system pressure drop not being lower than the pressure drop threshold as a constraint, the tooth height, tooth width, tooth spacing and coolant flow rate of the cold plate initial design parameter sample corresponding to the liquid cooling simulation data in the liquid cooling simulation data set as optimization parameter variables, and minimizing the cold plate cooling device temperature and the immersion cooling device temperature in the liquid cooling simulation data as the first objective function, to generate an optimized design parameter combination and the coolant flow rate of the optimized design parameter in the corresponding optimized design parameter combination.
[0087] In this embodiment, the above-mentioned execution entity can use the Multi-Island Genetic Algorithm (MIGA), with the system pressure drop not lower than the pressure drop threshold as a constraint, and minimizing the cold plate cooling device temperature and the immersion cooling device temperature in the liquid cooling simulation data as the first objective function. By dividing the optimization parameter variable population into multiple "islands", the genetic algorithm is independently run on each island, and individuals are regularly exchanged between islands to generate optimization results, thereby improving the diversity and global adaptability of the optimization results.
[0088] Please refer to step S103 of the embodiment shown in FIG. 1 for details, which will not be described again here.
[0089] Step S205 , selecting an optimized design parameter that meets the design requirements from the optimized design parameter combination as the target cold plate design parameter, and determining the coolant flow rate corresponding to the target cold plate design parameter according to the coolant flow rate of the optimized design parameter in the optimized design parameter combination.
[0090] In this embodiment, the design requirements can be determined according to the actual needs of the user, for example, the allowable temperature range of the cold plate cooling device and the allowable temperature range of the immersion cooling device can be pre-set. Therefore, the above-mentioned execution subject can select the optimized design parameter in the optimized design parameter combination as the target cold plate design parameter according to the design requirements, and determine the target coolant flow rate corresponding to the target cold plate design parameter in the coolant flow rate corresponding to the optimized design parameter.
[0091] Please refer to step S103 of the embodiment shown in FIG. 1 for details, which will not be described again here.
[0092] The cold plate optimization design method provided in this embodiment expands the initial design parameter samples of the cold plate and the corresponding liquid cooling simulation data through the proxy model, and then uses the multi-island genetic algorithm for optimization, and further expands the optimal design parameter combination corresponding to the target coolant flow rate that minimizes the cold plate cooling device temperature and the immersion cooling device temperature, thereby further optimizing the determined target cold plate design parameters, and can significantly improve the heat dissipation efficiency of the immersion liquid cooling of the cold plate directional centralized cooling, ensuring that the cooled electronic equipment or system can maintain a low operating temperature even under high load.
[0093] In some optional implementations of the present embodiment, the proxy model can be trained by the following steps: taking the cold plate initial design parameter samples in the cold plate initial design parameter sample set as input, taking the liquid cooling simulation data corresponding to the cold plate initial design parameter samples as output, taking minimizing the difference between the discrete point set corresponding to the cold plate initial design parameter sample set and the predicted surface of the proxy model as the second objective function, training the proxy model, and outputting the proxy model when the fitting accuracy between the discrete point set corresponding to the cold plate initial design parameter sample set and the predicted surface of the proxy model reaches an accuracy threshold.
[0094] Here, during the training of the proxy model, it is confirmed whether to output the proxy model based on the fitting accuracy. If the accuracy is sufficient, the proxy model is output, which improves the expression accuracy of the proxy model, thereby improving the accuracy of the represented sample points, so that the subsequent optimization process can be performed within the expanded cold plate initial design parameter sample set obtained by the proxy model prediction and the corresponding liquid cooling simulation data set.
[0095] In some optional implementations of the present embodiment, the method also includes: using a cold plate modeling tool to generate a target cold plate model based on the target cold plate design parameters; using a liquid cooling simulation tool to generate liquid cooling simulation data of the target cold plate model based on the coolant type, coolant flow rate and target cold plate model; using a proxy model to determine the liquid cooling prediction data corresponding to the target cold plate design parameters; if the difference between the liquid cooling simulation data and the liquid cooling prediction data of the target cold plate model is less than a difference threshold, outputting the target cold plate design parameters as optimization results.
[0096] In this implementation, after determining the target cold plate design parameters, it is also necessary to perform simulation verification on the target cold plate design parameters. If the error between the liquid cooling simulation result and the predicted result is within a certain allowable range, the result is output, thereby optimizing the determined optimization result. This can significantly improve the heat dissipation efficiency of the immersion liquid cooling of the cold plate directional centralized cooling corresponding to the optimization result, ensuring that the cooled electronic equipment or system can maintain a low operating temperature even under high load.
[0097] In some optional implementations of the present embodiment, the method also includes: if the difference is greater than the difference threshold, updating the target cold plate design parameters and the liquid cooling simulation data corresponding to the target cold plate design parameters to the cold plate initial design parameter sample set and the liquid cooling simulation data set corresponding to the cold plate initial design parameter sample set, and using the updated cold plate initial design parameter sample set and liquid cooling simulation data set to retrain the proxy model.
[0098] In this implementation, if the error between the simulation result and the prediction result is too high, it is necessary to expand the target cold plate design parameters and the liquid cooling simulation data corresponding to the target cold plate design parameters to the cold plate initial design parameter sample set, return to the proxy model training step, and re-train the proxy model until the prediction accuracy of the target cold plate design parameters in the output optimization design parameter combination meets the requirements, and then output the final optimization result. In this way, with fewer cold plate initial design parameter samples and combined with the proxy model prediction, accurate optimization results can be obtained while reducing the simulation workload and saving simulation time, and the heat dissipation efficiency of the immersion liquid cooling of the cold plate directional centralized cooling corresponding to the final optimization result is improved, ensuring that the cooled electronic equipment or system can maintain a low operating temperature even under high load.
[0099] For example, in a specific example, if the proxy model prediction method is used, the number of sample points can be reduced to 30 or even 20 under the premise that the number of optimization variables is 4 parameters. At the same time, the value range of the optimization variables can be widened, because even if the sample points are sparsely distributed, as long as the training accuracy of the proxy model is sufficient, they can be fitted into a smooth spatial surface to achieve accurate expansion of sample points. For details, please refer to Figure 2b , Figure 2bA schematic flow chart of the optimal solution for optimizing design parameters based on agent model training is shown.
[0100] In some optional implementations of the present embodiment, the above-mentioned cold plate optimization design method also includes: using target cold plate design parameters to manufacture a target cold plate; using the target cold plate to perform immersion cooling tests to obtain actual tested cold plate cooling device temperatures and immersion cooling device temperatures; based on the difference between the actual tested cold plate cooling device temperatures and immersion cooling device temperatures and the cold plate cooling device temperatures and immersion cooling device temperatures of liquid cooling simulation data corresponding to the target cold plate design parameters, updating the multi-island genetic algorithm to update the optimization design parameter combination.
[0101] In this implementation, in order to update the multi-island genetic algorithm to update the optimization design parameter combination, the target cold plate design parameters can be used to manufacture the cold plate, and actual tests can be performed to verify the effectiveness of the optimization results. If necessary, the optimization algorithm or objective function can be adjusted based on the actual test results to further optimize the cold plate design.
[0102] The cold plate optimization design method in this implementation clarifies the steps of updating the multi-island genetic algorithm, thereby improving the effectiveness of the multi-island genetic algorithm, and further improving the effectiveness of the optimization design parameter combination updated by the multi-island genetic algorithm.
[0103] In this embodiment, a cold plate optimization design device is also provided, which is used to implement the above embodiments and preferred implementations, and the descriptions that have been made will not be repeated. As used below, the term "module" can implement a combination of software and / or hardware for a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, the implementation of hardware, or a combination of software and hardware, is also possible and conceivable.
[0104] This embodiment provides a cold plate optimization design device, such as Figure 3 As shown, including:
[0105] The model generation module 301 is used to generate a cold plate model set according to a sample set of cold plate initial design parameters using a cold plate modeling tool;
[0106] The liquid cooling simulation module 302 is used to generate a liquid cooling simulation data set using a liquid cooling simulation tool according to the coolant type, coolant flow rate and cold plate model set, wherein the liquid cooling simulation data in the liquid cooling simulation data set includes: cold plate cooling device temperature, immersion cooling device temperature and system flow resistance;
[0107] The parameter optimization module 303 is used to use a multi-island genetic algorithm, with the system pressure drop not being less than the pressure drop threshold as a constraint, with the tooth height, tooth width, tooth spacing and coolant flow rate of the cold plate initial design parameter sample corresponding to the liquid cooling simulation data in the liquid cooling simulation data set as optimization parameter variables, and with minimizing the cold plate cooling device temperature and the immersion cooling device temperature in the liquid cooling simulation data as the first objective function, to generate an optimized design parameter combination and a coolant flow rate of the optimized design parameter in the corresponding optimized design parameter combination; the system pressure drop is determined based on the system flow resistance;
[0108] The parameter selection module 304 is used to select the optimized design parameters that meet the design requirements from the optimized design parameter combination as the target cold plate design parameters, and determine the target coolant flow rate corresponding to the target cold plate design parameters according to the coolant flow rate of the optimized design parameters in the corresponding optimized design parameter combination.
[0109] In some optional implementations of this embodiment, the sample set of initial design parameters of the cold plate in the model generation module 301 is determined based on the following modules:
[0110] A parameter acquisition module is used to obtain the structural parameter range of the cold plate scraper teeth, and the structural parameter range includes: tooth height range, tooth width range, and tooth spacing range;
[0111] A flow acquisition module, used to obtain the coolant inlet flow range;
[0112] The parameter generation module is used to generate a cold plate initial design parameter sample set using an experimental design device according to the structural parameter range of the cold plate scraper teeth and the coolant inlet flow range. The cold plate initial design parameter samples in the cold plate initial design parameter sample set include: tooth height, tooth width, tooth spacing and coolant inlet flow.
[0113] In some optional implementations of this embodiment, the device further includes:
[0114] The parameter expansion module adopts a proxy model to expand the cold plate initial design parameter samples and corresponding liquid cooling simulation data in the cold plate initial design parameter sample set.
[0115] In some optional implementations of this embodiment, the proxy model in the parameter expansion module is trained using the following modules:
[0116] The model training module takes the cold plate initial design parameter samples in the cold plate initial design parameter sample set as input, takes the liquid cooling simulation data corresponding to the cold plate initial design parameter samples as output, takes minimizing the difference between the discrete point set corresponding to the cold plate initial design parameter sample set and the predicted surface of the proxy model as the second objective function, trains the proxy model, and outputs the proxy model when the fitting accuracy between the discrete point set corresponding to the cold plate initial design parameter sample set and the predicted surface of the proxy model reaches an accuracy threshold.
[0117] In some optional implementations of this embodiment, the device further includes:
[0118] A cold plate modeling module, used for generating a target cold plate model according to target cold plate design parameters by using a cold plate modeling tool;
[0119] The liquid cooling simulation module is further used to generate liquid cooling simulation data of the target cold plate model according to the coolant type, the coolant flow rate and the target cold plate model using the liquid cooling simulation tool;
[0120] A liquid cooling prediction module, used to determine liquid cooling prediction data corresponding to target cold plate design parameters using a proxy model;
[0121] The parameter output module is used to output the target cold plate design parameters as optimization results if the difference between the liquid cooling simulation data and the liquid cooling prediction data of the target cold plate model is less than a difference threshold.
[0122] In some optional implementations of this embodiment, the device further includes:
[0123] The model optimization module is used to update the target cold plate design parameters and the liquid cooling simulation data corresponding to the target cold plate design parameters to the cold plate initial design parameter sample set and the liquid cooling simulation data set corresponding to the cold plate initial design parameter sample set if the difference is greater than the difference threshold, and use the updated cold plate initial design parameter sample set and liquid cooling simulation data set to retrain the proxy model.
[0124] In some optional implementations of this embodiment, the parameter optimization module includes:
[0125] An initialization group module is used to use the tooth height, tooth width, tooth spacing and coolant flow rate of the cold plate initial design parameter sample corresponding to the liquid cooling simulation data as optimization parameter variables, execute the initialization group for the optimization parameter variable set corresponding to the liquid cooling simulation data set, and obtain multiple sub-groups, wherein the optimization parameter variables include: the tooth height, tooth width, tooth spacing and coolant flow rate of the cold plate initial design parameter sample;
[0126] The subpopulation operation module is used to perform independent standard genetic operations and inter-island migration operations on each subpopulation to obtain genetic subpopulations. Standard genetic operations include: selection, crossover and mutation;
[0127] A fitness calculation module is used to calculate the fitness of a new optimization parameter variable according to the cold plate cooling device temperature and the immersion cooling device temperature of the liquid cooling simulation data corresponding to the new optimization parameter variable in the genetic subpopulation, taking the minimization of the cold plate cooling device temperature and the immersion cooling device temperature in the liquid cooling simulation data as the first objective function;
[0128] A pressure drop determination module, for determining a system pressure drop based on a system flow resistance of liquid cooling simulation data corresponding to a new optimization parameter variable in a genetic subpopulation;
[0129] The result recording module is used to use the following two constraints: the fitness is greater than the fitness threshold, and the system pressure drop is not less than the pressure drop threshold, to guide the iteration to perform independent standard genetic operations and inter-island migration operations on each sub-population, and to record the optimal optimization parameter variables of each iteration and the cold plate cooling device temperature and immersion cooling device temperature corresponding to the optimal optimization parameter variables, as well as the system pressure drop determined based on the system flow resistance, until a predetermined number of iterations is reached or the objective function meets the convergence condition, and the optimized design parameter combination and the coolant flow rate of the optimized design parameter in the corresponding optimized design parameter combination are obtained.
[0130] The further functional description of each of the above modules and units is the same as that of the above corresponding embodiments and will not be repeated here.
[0131] The cold plate optimization design device in this embodiment is presented in the form of a functional unit, where the unit refers to an ASIC (Application Specific Integrated Circuit) circuit, a processor and memory that executes one or more software or fixed programs, and / or other devices that can provide the above functions.
[0132] See also Figure 4 , Figure 4 1 is a schematic diagram of the structure of a computer device provided by an optional embodiment of the present disclosure, and the computer device includes: one or more processors 10, a memory 20, and interfaces for connecting various components, including high-speed interfaces and low-speed interfaces. The various components are connected to each other using different buses for communication, and can be installed on a common mainboard or installed in other ways as needed. The processor can process instructions executed in the computer device, including instructions stored in or on the memory to display graphical information of the GUI on an external input / output device (such as a display device coupled to the interface). In some optional embodiments, if necessary, multiple processors and / or multiple buses can be used together with multiple memories and multiple memories. Similarly, multiple computer devices can be connected, and each device provides some necessary operations (for example, as a server array, a group of blade servers, or a multi-processor system).
[0133] The processor 10 may be a central processing unit, a network processor or a combination thereof. The processor 10 may further include a hardware chip. The hardware chip may be a dedicated integrated circuit, a programmable logic device or a combination thereof. The programmable logic device may be a complex programmable logic device, a field programmable gate array, a general purpose array logic or any combination thereof.
[0134] The memory 20 stores instructions executable by at least one processor 10, so that the at least one processor 10 executes the method shown in the above embodiment.
[0135] The memory 20 may include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required for at least one function; the data storage area may store data created according to the use of the computer device, etc. In addition, the memory 20 may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some optional embodiments, the memory 20 may optionally include a memory remotely arranged relative to the processor 10, and these remote memories may be connected to the computer device via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.
[0136] The memory 20 may include a volatile memory, such as a random access memory; the memory may also include a non-volatile memory, such as a flash memory, a hard disk or a solid state drive; the memory 20 may also include a combination of the above types of memory.
[0137] The computer device further includes an input device 30 and an output device 40. The processor 10, the memory 20, the input device 30 and the output device 40 may be connected via a bus or other means.
[0138] The input device 30 can receive input digital or character information, and generate key signal input related to the user settings and function control of the computer device, such as a touch screen, a keypad, a mouse, a track pad, a touch pad, an indicator bar, one or more mouse buttons, a trackball, a joystick, etc. The output device 40 may include a display device, an auxiliary lighting device (e.g., an LED) and a tactile feedback device (e.g., a vibration motor), etc. The above-mentioned display device includes but is not limited to a liquid crystal display, a light emitting diode, a display and a plasma display. In some optional embodiments, the display device can be a touch screen.
[0139] The embodiments of the present disclosure also provide a computer-readable storage medium. The above-mentioned method according to the embodiments of the present disclosure can be implemented in hardware, firmware, or can be implemented as a computer code that can be recorded in a storage medium, or can be implemented as a computer code that is originally stored in a remote storage medium or a non-temporary machine-readable storage medium and will be stored in a local storage medium and downloaded through a network, so that the method described herein can be stored in such software processing on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. Among them, the storage medium can be a magnetic disk, an optical disk, a read-only storage memory, a random access memory, a flash memory, a hard disk or a solid-state drive, etc.; further, the storage medium can also include a combination of the above-mentioned types of memory. It can be understood that a computer, a processor, a microprocessor controller, or programmable hardware includes a storage component that can store or receive software or computer code. When the software or computer code is accessed and executed by a computer, a processor, or hardware, the method shown in the above embodiment is implemented.
[0140] A part of the present disclosure may be applied as a computer program product, such as a computer program instruction, which, when executed by a computer, can call or provide the method and / or technical solution according to the present disclosure through the operation of the computer. Those skilled in the art should understand that the existence of computer program instructions in computer-readable media includes, but is not limited to, source files, executable files, installation package files, etc., and accordingly, the way in which computer program instructions are executed by a computer includes, but is not limited to: the computer directly executes the instruction, or the computer compiles the instruction and then executes the corresponding compiled program, or the computer reads and executes the instruction, or the computer reads and installs the instruction and then executes the corresponding installed program. Here, the computer-readable medium can be any available computer-readable storage medium or communication medium accessible to the computer.
[0141] Although the embodiments of the present disclosure have been described in conjunction with the accompanying drawings, those skilled in the art may make various modifications and variations without departing from the spirit and scope of the present disclosure, and such modifications and variations are all within the scope defined by the appended claims.
Claims
1. A cold plate optimization design method, It is characterized in that The method comprises: Using the cold plate modeling tool, a cold plate model set is generated according to a sample set of initial design parameters of the cold plate; Using a liquid cooling simulation tool, generating a liquid cooling simulation data set according to the coolant type, the coolant flow rate and the cold plate model set, wherein the liquid cooling simulation data in the liquid cooling simulation data set includes: a cold plate cooling device temperature, an immersion cooling device temperature and a system flow resistance; Utilizing a multi-island genetic algorithm, taking the system pressure drop not being lower than a pressure drop threshold as a constraint, taking the tooth height, tooth width, tooth spacing and coolant flow rate of the cold plate initial design parameter sample corresponding to the liquid cooling simulation data in the liquid cooling simulation data set as optimization parameter variables, and minimizing the cold plate cooling device temperature and the immersion cooling device temperature in the liquid cooling simulation data as a first objective function, generating an optimized design parameter combination and a coolant flow rate corresponding to the optimized design parameter in the optimized design parameter combination; the system pressure drop is determined based on the system flow resistance; From the optimized design parameter combination, the optimized design parameter that meets the design requirements is selected as the target cold plate design parameter, and the target coolant flow rate corresponding to the target cold plate design parameter is determined according to the coolant flow rate corresponding to the optimized design parameter in the optimized design parameter combination.
2. The method according to claim 1, It is characterized in that The sample set of initial design parameters of the cold plate is determined based on the following steps: Obtain the structural parameter range of the cold plate scraper teeth, the structural parameter range includes: tooth height range, tooth width range, and tooth spacing range; Get the coolant inlet flow range; An experimental design method is used to generate a cold plate initial design parameter sample set according to the structural parameter interval of the cold plate scraper teeth and the coolant inlet flow interval. The cold plate initial design parameter samples in the cold plate initial design parameter sample set include: tooth height, tooth width, tooth spacing and coolant inlet flow.
3. The method according to claim 1, It is characterized in that The method further comprises: The proxy model is adopted to expand the cold plate initial design parameter samples and the corresponding liquid cooling simulation data in the cold plate initial design parameter sample set.
4. The method according to claim 3, It is characterized in that The proxy model is trained by the following steps: The cold plate initial design parameter samples in the cold plate initial design parameter sample set are taken as input, the liquid cooling simulation data corresponding to the cold plate initial design parameter samples are taken as output, and minimizing the difference between the discrete point set corresponding to the cold plate initial design parameter sample set and the predicted surface of the proxy model is taken as the second objective function to train the proxy model. When the fitting accuracy between the discrete point set corresponding to the cold plate initial design parameter sample set and the predicted surface of the proxy model reaches an accuracy threshold, the proxy model is output.
5. The method according to claim 4, It is characterized in that The method further comprises: Using the cold plate modeling tool, generating a target cold plate model according to the target cold plate design parameters; Using the liquid cooling simulation tool, generating liquid cooling simulation data of a target cold plate model according to the coolant type, the coolant flow rate and the target cold plate model; Determining liquid cooling prediction data corresponding to the target cold plate design parameters using the proxy model; If the difference between the liquid cooling simulation data of the target cold plate model and the liquid cooling prediction data is less than a difference threshold, the target cold plate design parameters are output as optimization results.
6. The method according to claim 5, It is characterized in that The method further comprises: If the difference is greater than the difference threshold, the target cold plate design parameters and the liquid cooling simulation data corresponding to the target cold plate design parameters are updated to the cold plate initial design parameter sample set and the liquid cooling simulation data set corresponding to the cold plate initial design parameter sample set, and the updated cold plate initial design parameter sample set and liquid cooling simulation data set are used to retrain the proxy model.
7. The method according to claim 1, It is characterized in that The multi-island genetic algorithm is used to generate an optimization result according to the liquid cooling simulation data set, with the system pressure drop determined based on the system flow resistance not being less than the pressure drop threshold as a constraint, with the tooth height, tooth width, tooth spacing and coolant flow rate of the cold plate initial design parameter sample corresponding to the liquid cooling simulation data as optimization parameter variables, and with minimizing the cold plate cooling device temperature and the immersion cooling device temperature in the liquid cooling simulation data as the first objective function, including: Taking the tooth height, tooth width, tooth spacing and coolant flow rate of the cold plate initial design parameter sample corresponding to the liquid cooling simulation data as optimization parameter variables, executing the initialization group for the optimization parameter variable set corresponding to the liquid cooling simulation data set to obtain multiple sub-groups, the optimization parameter variables include: the tooth height, tooth width, tooth spacing and coolant flow rate of the cold plate initial design parameter sample; Performing independent standard genetic operations and inter-island migration operations on each sub-population to obtain genetic sub-populations, wherein the standard genetic operations include: selection, crossover and mutation; Taking minimizing the cold plate cooling device temperature and the immersion cooling device temperature in the liquid cooling simulation data as the first objective function, calculating the fitness of the new optimization parameter variable according to the cold plate cooling device temperature and the immersion cooling device temperature of the liquid cooling simulation data corresponding to the new optimization parameter variable in the genetic subpopulation; determining a system pressure drop based on the system flow resistance of the liquid cooling simulation data corresponding to the new optimized parameter variables in the genetic subpopulation; The following two constraints are adopted: the fitness is greater than the fitness threshold, and the system pressure drop is not less than the pressure drop threshold, guiding the iteration of performing independent standard genetic operations and inter-island migration operations on each subpopulation, recording the optimal optimization parameter variables of each iteration and the cold plate cooling device temperature and immersion cooling device temperature corresponding to the optimal optimization parameter variables, and the system pressure drop determined based on the system flow resistance, until a predetermined number of iterations is reached or the objective function meets the convergence condition, and obtaining the optimized design parameter combination and the coolant flow rate corresponding to the optimized design parameters in the optimized design parameter combination.
8. A cold plate optimization design device, It is characterized in that The device comprises: A model generation module, for generating a cold plate model set according to a sample set of initial design parameters of the cold plate using a cold plate modeling tool; A liquid cooling simulation module, for generating a liquid cooling simulation data set using a liquid cooling simulation tool according to the coolant type, the coolant flow rate and the cold plate model set, wherein the liquid cooling simulation data in the liquid cooling simulation data set includes: a cold plate cooling device temperature, an immersion cooling device temperature and a system flow resistance; a parameter optimization module, for generating an optimized design parameter combination and a coolant flow rate of the optimized design parameter in the corresponding optimized design parameter combination by using a multi-island genetic algorithm, with the system pressure drop not being lower than a pressure drop threshold as a constraint, with the tooth height, tooth width, tooth spacing and coolant flow rate of the cold plate initial design parameter sample corresponding to the liquid cooling simulation data in the liquid cooling simulation data set as optimization parameter variables, and minimizing the cold plate cooling device temperature and the immersion cooling device temperature in the liquid cooling simulation data as a first objective function; the system pressure drop is determined based on the system flow resistance; The parameter selection module is used to select the optimized design parameters that meet the design requirements from the optimized design parameter combination as the target cold plate design parameters, and determine the target coolant flow rate corresponding to the target cold plate design parameters according to the coolant flow rate corresponding to the optimized design parameters in the optimized design parameter combination.
9. A computer device, It is characterized in that include: A memory and a processor, wherein the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the cold plate optimization design method according to any one of claims 1 to 7 by executing the computer instructions.
10. A computer-readable storage medium, It is characterized in that The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a computer to execute the cold plate optimization design method according to any one of claims 1 to 7.
Citation Information
Patent Citations
Thermal simulation analysis method and device for liquid cooling energy storage system and storage medium
CN115203909A
Flow resistance optimization method of shoveling tooth cold plate and low-flow-resistance shoveling tooth cold plate
CN116306115A
Two-phase cold plate liquid cooling system and method
CN118250982A
Liquid cooled condensers for loop heat pipe like enclosure cooling
US20110277967A1
Multidisciplinary structural design optimization method for fuel assembly based on co-simulation
US20230252203A1
Cited By
Parameter optimization method and device for chemical cooling system
CN119717719A
Wave-shaped lithium battery cold plate multi-objective optimization method based on deep neural network
CN120068579A
Two-phase cold plate liquid cooling method and system
CN120373153A
Cooling processing method and device of server, electronic equipment and storage medium
CN120447710A
Server cooling method, device, electronic equipment and storage medium
CN120447710B
