A shunt design method and shunt device of a through liquid cooling case

By grouping and controlling the structure and size of the flow branch cavity, the junction cavity, and the transition section of the flow branch through a clustering algorithm, the problem of limited coolant flow management is solved, and the coolant resource utilization and heat dissipation efficiency of the through-type liquid cooling chassis are improved.

CN115221803BActive Publication Date: 2026-05-01SOUTHWEST CHINA RES INST OF ELECTRONICS EQUIP
View PDF 2 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
SOUTHWEST CHINA RES INST OF ELECTRONICS EQUIP
Filing Date
2022-03-25
Publication Date
2026-05-01

AI Technical Summary

Technical Problem

Existing technologies cannot effectively manage limited coolant flow, resulting in low resource utilization in the heat dissipation design of high heat-consuming modules in through-type liquid-cooled chassis.

Method used

Clustering algorithms are used to group and classify modules, and the structure and size of the flow splitter, the junction box, and the transition section of the flow splitter branch are controlled to achieve on-demand flow splitting and reasonable allocation of coolant flow.

Benefits of technology

It improves the utilization rate of coolant resources, simplifies the design process, reduces the manufacturing difficulty, and achieves efficient satisfaction of the module's heat dissipation requirements.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN115221803B_ABST
    Figure CN115221803B_ABST
Patent Text Reader

Abstract

The application discloses a shunt design method and shunt device of a penetrating liquid cooling case, and belongs to the field of electronic equipment heat dissipation, and comprises the following steps: grouping and classifying modules of the penetrating liquid cooling case based on a clustering algorithm, and controlling the structure form and size of a shunt cavity, a confluence cavity and a shunt branch transition section to realize on-demand shunting. According to the heat consumption distribution of the modules in the case, the application reasonably performs flow distribution of the cooling liquid, and improves the utilization rate of the cooling liquid resources under the premise of satisfying the module heat dissipation.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the field of heat dissipation for electronic devices, and more specifically, to a heat distribution design method and heat distribution device for a through-type liquid-cooled chassis. Background Technology

[0002] Currently, airborne electronic equipment typically employs a modular design, with modules using standardized structures to improve design reusability, shorten design cycles, and reduce maintenance costs. With the development of electronic information technology, modules are becoming increasingly complex and integrated, resulting in higher heat dissipation and heat flux density. Traditional air-cooled and conductive chassis are no longer sufficient to address heat dissipation issues. In recent years, through-type liquid-cooled chassis have been gradually adopted as a solution for high-heat-dissipation modules.

[0003] A through-type liquid-cooled chassis supplies coolant to the modules through a distributor within the chassis, reducing the thermal resistance between the chip and the coolant and thus improving the chip's heat dissipation. Due to the scarcity of liquid cooling resources in aircraft, the available coolant flow rate is limited, typically supplied at 2L / (min·kW). Therefore, it is essential to employ a rational distribution design to precisely manage the coolant flow distribution while meeting heat dissipation design and manufacturing requirements, thereby improving the utilization rate of cooling resources. Summary of the Invention

[0004] The purpose of this invention is to overcome the shortcomings of the prior art and provide a flow distribution design method and flow distribution device for a through-type liquid-cooled chassis. Based on the heat dissipation distribution of the modules inside the chassis, the flow rate of the coolant is reasonably distributed, thereby improving the utilization rate of coolant resources while meeting the heat dissipation requirements of the modules.

[0005] The objective of this invention is achieved through the following solution:

[0006] A method for shunt design of a through-type liquid-cooled chassis, comprising the following steps:

[0007] Based on clustering algorithms, the modules of the through-type liquid cooling chassis are grouped and classified, and the structure and size of the flow distribution cavity, the flow junction cavity, and the transition section of the flow distribution branch are controlled to achieve on-demand flow distribution.

[0008] Furthermore, it includes sub-steps:

[0009] S1. Count the number of modules n in the liquid-cooled chassis and determine the heat dissipation of each module in the liquid-cooled chassis, forming a module heat dissipation sample set D = {q1, q2, ... q}. n} Calculate the total heat consumption of the liquid-cooled unit.

[0010] S2, calculate the total flow requirement Q of the liquid cooling chassis. s ;

[0011] S3, calculate the diameter d of the splitter cavity and the merger cavity. s ;

[0012] S4, based on the module heat dissipation distribution, group modules with similar heat dissipation values ​​together, using C as the group. i Represents module group, μ i For group C i The average heat dissipation of the middle module, C i According to the corresponding μ i Arranged from smallest to largest, all modules can be divided into (C1, C2, ... C... k ), k≤5;

[0013] S5, C per group i All modules within the module group are allocated the same traffic. Calculate the traffic allocation requirements for each group of modules, denoted as Q1, Q2…Q… k ;

[0014] S6, the flow channel structure of the transition section of all module branch roads in each group is the same;

[0015] S7, adjust C1, C2...C k-1 The diameters d1, d2…d of the transition section of the group module branch are... k-1 By analyzing the structure of the transition section and performing iterative calculations using CFD simulation, the diameter and structural form of the transition section of each module's branch circuit can be determined.

[0016] Further, in step S2, the total flow rate requirement Q of the liquid cooling unit is calculated according to the formula Q = 2L / (min·kW) × q. s .

[0017] Furthermore, in step S3, the liquid flow velocity at the inlet of the diversion chamber and the outlet of the confluence chamber is controlled to be no greater than 1 m / s, according to the formula... Calculate the diameter d of the splitter cavity and the manifold cavity. s .

[0018] Furthermore, in step S4, the specific partitioning method adopts the K-means clustering algorithm, using the squared error... The goal is to minimize the distribution of heat consumption samples in each group, making the distribution as close as possible to the heat consumption samples in each group.

[0019] Further, in step S5, according to formula Q i = 2L / (min×kW)×μ i Calculate the traffic allocation requirements for each group of modules.

[0020] Further, step S6 includes the sub-step: placing C kThe transition section of the branch flow path is designed as a straight orifice, and the liquid flow velocity within the transition section is controlled to be no greater than 0.3 m / s, based on the maximum average heat consumption of module group C. k Single module traffic Q k and formula Calculate C k The diameter d of the transition section of the module k .

[0021] Furthermore, in step S7, when performing CFD simulation and iterative calculation, the convergence criterion is that the deviation between the flow allocation value and the flow demand of each group of modules is less than 10%.

[0022] A flow distribution device for a through-type liquid-cooled chassis is designed using any of the design methods described above.

[0023] The beneficial effects of this invention include:

[0024] (1) The design method in this invention achieves on-demand flow splitting by controlling the structure and size of the splitting cavity, the junction cavity and the transition section of the splitting branch, which has the advantages of simple design and good manufacturability.

[0025] (2) Based on cluster analysis theory, this invention groups and classifies modules, controls the solution parameters of the flow split design within a certain range, and achieves reasonable allocation of coolant resources while ensuring design efficiency. Attached Figure Description

[0026] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0027] Figure 1 This is a flowchart of the method steps in an embodiment of the present invention;

[0028] Figure 2 This is a schematic diagram of the first structure of a through-type liquid-cooled chassis;

[0029] Figure 3 This is a schematic diagram of the second structure of a through-type liquid-cooled chassis;

[0030] Figure 4 This is a schematic diagram of the third structure of a through-type liquid-cooled chassis;

[0031] In the diagram, 1-chassis frame, 2-through liquid cooling module, 3-distribution plate, 4-liquid inlet, 5-liquid outlet, 6-module blind-fit liquid cooling connector, 7-distribution channel, 8-combination channel, 9-distribution branch transition section, 10-module internal channel, 301-distribution cavity, 302-combination cavity, 303-C1 group branch transition section, 304-C2 group branch transition section, 305-C3 group branch transition section. Detailed Implementation

[0032] All features disclosed in all embodiments of this specification, or steps in all methods or processes implied in the disclosure, may be combined and / or extended or substituted in any way, except for mutually exclusive features and / or steps. The following description is based on the appendix. Figures 1-4 The present invention will be further described in detail and fully. The technical concept, the technical problem solved, the working principle, the working process and the beneficial effects of the present invention will be explained in more detail and fully.

[0033] This invention provides a flow distribution design method for a through-type liquid-cooled chassis. The through-type liquid-cooled chassis mainly consists of a chassis frame, a flow distribution plate, and through-type liquid-cooling modules. The entire chassis has an inlet and an outlet on its side. The inlet connects to a flow distribution cavity in the flow distribution plate, and the outlet connects to a manifold cavity in the flow distribution plate. Both the flow distribution cavity and the manifold cavity are circular through-cavities of the same diameter. Blind-fit connectors for the through-type liquid-cooling modules are arranged on the flow distribution plate, and there are flow branch transition sections between the flow distribution cavity or manifold cavity and the blind-fit connectors of the modules. All cold plates of the through-type liquid-cooling modules are designed with the same flow channels and have the same flow resistance characteristic curve.

[0034] like Figure 2 As shown, the through-type liquid cooling chassis includes a chassis frame 1, a through-type liquid cooling module 2, a flow divider 3, a liquid inlet 4, and a liquid outlet 5.

[0035] like Figure 3 As shown, the through-type liquid cooling chassis is equipped with a module blind-plug liquid cooling connector 6, a branch flow channel 7, a convergence flow channel 8, a branch transition section 9, and an internal flow channel 10.

[0036] like Figure 4 As shown, the through-type liquid cooling chassis is provided with a flow distribution chamber 301, a flow combination chamber 302, a C1 group branch transition section 303, a C2 group branch transition section 304, and a C3 group branch transition section 305.

[0037] In this embodiment of the invention, customized flow splitting of the module is achieved by adjusting the flow channel structure and dimensional parameters of the splitting cavity, the merging cavity, and the transition section of the splitting branch, such as... Figure 1 As shown, the specific design steps are as follows:

[0038] Step S1: Count the number of modules n in the liquid-cooled chassis and determine the heat dissipation of each module in the liquid-cooled chassis, forming a module heat dissipation sample set D = {q1, q2, ... q}. n} Calculate the total heat consumption of the liquid-cooled unit.

[0039] Step S2: Calculate the total flow rate requirement Q of the liquid cooling unit using the formula Q = 2L / (min·kW) × q. s ;

[0040] Step S3: To reduce the head resistance loss of the fluid in the splitter and merger chambers, the liquid velocity at the inlet of the splitter chamber and the outlet of the merger chamber is usually controlled to be no greater than 1 m / s. Therefore, according to the formula... Calculate the diameter d of the splitter cavity and the manifold cavity. s ;

[0041] Step S4: Based on the heat dissipation distribution of the modules, group modules with similar heat dissipation values ​​together, using C as the grouping factor. i Represents module group, μ i For group C i The average heat dissipation of the middle module, C i According to the corresponding μ i Arranged from smallest to largest, all modules can be divided into (C1, C2, ... C... k Typically, k ≤ 5 is required;

[0042] The specific partitioning method can adopt the K-means method in clustering algorithms, using the squared error. The goal is to minimize the heat consumption samples of modules in each group so that the distribution of these samples is as close as possible.

[0043] Step S5: Each group C i All modules within the module are allocated the same traffic, according to formula Q. i = 2L / (min×kW)×μ i The traffic allocation requirements for each group of modules are calculated as Q1, Q2...Q k ;

[0044] Step S6: The flow channel structure of the transition section of all module branch paths within each group is the same. In order to reduce C k The flow resistance loss of the transition section is usually C k The transition section of the branch flow path is designed as a straight orifice, and the liquid flow velocity within the transition section is controlled to be no greater than 0.3 m / s, based on the maximum average heat consumption of module group C. k Single module traffic Q k and formula Calculate C k The diameter d of the transition section of the module k ;

[0045] Step S7: Adjust C1, C2…C k-1 The diameters d1, d2…d of the transition section of the group module branch are... k-1 The structure of the transition section is determined by CFD simulation and iterative calculation. The convergence criterion is that the deviation between the flow allocation value and the flow demand of each group of modules is less than 10%. This allows us to determine the diameter and structure of the transition section of the branch road of each group of modules.

[0046] Example 1

[0047] A method for shunt design of a through-type liquid-cooled chassis, comprising the following steps:

[0048] Based on clustering algorithms, the modules of the through-type liquid cooling chassis are grouped and classified, and the structure and size of the flow distribution cavity, the flow junction cavity, and the transition section of the flow distribution branch are controlled to achieve on-demand flow distribution.

[0049] Example 2

[0050] Based on Example 1, the following sub-steps are included:

[0051] S1. Count the number of modules n in the liquid-cooled chassis and determine the heat dissipation of each module in the liquid-cooled chassis, forming a module heat dissipation sample set D = {q1, q2, ... q}. n} Calculate the total heat consumption of the liquid-cooled unit.

[0052] S2, calculate the total flow requirement Q of the liquid cooling chassis. s ;

[0053] S3, calculate the diameter d of the splitter cavity and the merger cavity. s ;

[0054] S4, based on the module heat dissipation distribution, group modules with similar heat dissipation values ​​together, using C as the group. i Represents module group, μ i For group C i The average heat dissipation of the middle module, C i According to the corresponding μ i Arranged from smallest to largest, all modules can be divided into (C1, C2, ... C... k ), k≤5;

[0055] S5, C per group i All modules within the module group are allocated the same traffic. Calculate the traffic allocation requirements for each group of modules, denoted as Q1, Q2…Q… k ;

[0056] S6, the flow channel structure of the transition section of all module branch roads in each group is the same;

[0057] S7, adjust C1, C2...Ck-1 The diameters d1, d2…d of the transition section of the group module branch are... k-1 By analyzing the structure of the transition section and performing iterative calculations using CFD simulation, the diameter and structural form of the transition section of each module's branch circuit can be determined.

[0058] Example 3

[0059] Based on Example 2, in step S2, the total flow rate requirement Q of the liquid cooling unit is calculated according to the formula Q = 2L / (min·kW) × q. s .

[0060] Example 4

[0061] Based on Example 2, in step S3, the liquid flow velocity at the inlet of the diversion cavity and the outlet of the confluence cavity is controlled to be no greater than 1 m / s, according to the formula... Calculate the diameter d of the splitter cavity and the manifold cavity. s .

[0062] Example 5

[0063] Based on Example 2, in step S4, the specific partitioning method adopts the K-means clustering algorithm, using the squared error. The goal is to minimize the distribution of heat consumption samples in each group, making the distribution as close as possible to the heat consumption samples in each group.

[0064] Example 6

[0065] Based on Example 2, in step S5, according to formula Q i = 2L / (min×kW)×μ i Calculate the traffic allocation requirements for each group of modules.

[0066] Example 7

[0067] Based on Example 2, step S6 includes the sub-step: C k The transition section of the branch flow path is designed as a straight orifice, and the liquid flow velocity within the transition section is controlled to be no greater than 0.3 m / s, based on the maximum average heat consumption of module group C. k Single module traffic Q k and formula Calculate C k The diameter d of the transition section of the module k .

[0068] Example 8

[0069] Based on Example 2, in step S7, when performing CFD simulation and iterative calculation, the convergence criterion is that the deviation between the flow allocation value and the flow demand of each group of modules is less than 10%.

[0070] Example 9

[0071] A flow distribution device for a through-type liquid-cooled chassis is designed using any of the design methods described in Examples 1 to 8.

[0072] Example 10

[0073] The implementation of this invention is further explained below with reference to a typical liquid-cooled chassis with a flow distribution scheme: Multiple through-type liquid-cooled modules are installed in the liquid-cooled chassis, each with identical flow channels. The flow distribution of the liquid-cooled chassis is designed according to the heat dissipation of different modules. The specific steps are as follows:

[0074] Step S1: Count the number of modules n in the liquid cooling unit and determine the heat consumption of each module in the liquid cooling unit, forming a module heat consumption sample set D = {q1, q2, ... q}. n} Calculate the total heat consumption of the liquid-cooled unit.

[0075] Step S2: Calculate the total flow rate requirement Q of the liquid cooling unit using the formula Q = 2L / (min·kW) × q. s ;

[0076] Step S3: To reduce the head resistance loss of the fluid in the splitter and merger chambers, the liquid velocity at the inlet of the splitter chamber and the outlet of the merger chamber is usually controlled to be no greater than 1 m / s. Therefore, according to the formula... Calculate the diameter d of the splitter cavity and the manifold cavity. s Typically, for ease of manufacturing and processing, d s Round it upwards;

[0077] Step S4: Based on the heat dissipation distribution of the modules, group modules with similar heat dissipation values ​​together, using C as the grouping factor. i Represents module group, μ i For group C i The average heat dissipation of the middle module, C i According to the corresponding μ i Arranged from smallest to largest, all modules can be divided into (C1, C2, ... C... k Typically, k ≤ 5 is required, and to control the number of module groups, μ is usually required. i / μ i-1 ≥1.2;

[0078] For specific partitioning methods, the K-means clustering algorithm can be used, with the squared error as the basis of the algorithm. To minimize the heat consumption of the modules, the modules are divided into k groups, such that the heat consumption samples of the modules in each group are distributed as closely as possible.

[0079] Step S5: All modules within each group are allocated the same flow rate, according to formula Q.i = 2L / (min·kW)×μ i The traffic allocation requirements for each group of modules are calculated as Q1, Q2...Q k ;

[0080] Step S6: The flow channel structure of the transition section of all module branch paths within each group is the same. In order to reduce C k The flow resistance loss of the transition section is usually C k The transition section of the component flow branch is designed as a straight orifice, and the liquid velocity within the transition section is controlled to be no greater than 0.3 m / s, based on the maximum average heat consumption of module C. k Single module traffic Q k and formula Calculate C k The diameter d of the transition section of the group module branch is k ;

[0081] Step S7: Adjust C1, C2…C k-1 The diameters d1, d2…d of the transition section of the group module branch are... k-1 Regarding the structure of the transition section, a straight-hole structure is usually preferred. To prevent clogging caused by excessively small flow channel dimensions, the diameter of the transition section should not be less than 2mm. If adjusting the transition section diameter to 2mm still cannot meet the flow distribution requirements, a serpentine flow channel structure can be used, adjusting the flow distribution by changing the number of bends in the serpentine flow channel. Through CFD simulation and iterative calculation, the convergence criterion is that the deviation between the flow distribution value and the flow demand of each group of modules is less than 10%, thus determining the diameter and structure of the transition section of each group of flow branch modules.

[0082] All parts not covered in this invention are the same as or can be implemented using existing technologies.

[0083] The above technical solution is only one embodiment of the present invention. For those skilled in the art, based on the application methods and principles disclosed in the present invention, it is easy to make various types of improvements or modifications, and not limited to the methods described in the above specific embodiments of the present invention. Therefore, the methods described above are only preferred and are not restrictive.

[0084] In addition to the examples above, other embodiments may be obtained by those skilled in the art based on the above disclosure or by making modifications using knowledge or technology in related fields. The features of each embodiment may be interchanged or replaced. Modifications and changes made by those skilled in the art that do not depart from the spirit and scope of the present invention should be within the protection scope of the appended claims.

Claims

1. A flow distribution design method for a through-type liquid-cooled chassis, characterized in that, Including the following steps: Based on clustering algorithms, the modules of the through-type liquid-cooled chassis are grouped and classified, and the structural form and size of the distribution cavity, the manifold cavity, and the transition section of the distribution branch are controlled to achieve on-demand distribution; specifically, it includes the following sub-steps: S1. Count the number of modules n in the liquid-cooled chassis and determine the heat dissipation of each module in the liquid-cooled chassis to form a module heat dissipation sample set. Calculate the total heat consumption of the liquid-cooled unit. ; S2, calculate the total flow requirement of the liquid cooling chassis. ; S3, calculate the diameters of the splitter cavity and the merger cavity. ; S4, based on the module heat dissipation distribution, group modules with similar heat dissipation values ​​together, to... Represents module group, For group The average heat dissipation of the middle module will According to the corresponding Arranged from smallest to largest, all modules can be divided into: k≤5; S5, each group All modules within the group are allocated the same traffic. Calculate the traffic allocation requirements for each group of modules, as follows: ; S6, the flow channel structure of the transition section of all module branch flow paths in each group is the same; S7, Adjustment Diameter of the transition section of the group module branch circuit The diameter and structural form of the transition section of each module branch can be determined by CFD simulation and iterative calculation. In step S2, according to the formula Calculate the total flow rate requirement of the liquid cooling chassis. ; In step S3, the liquid flow velocity at the inlet of the diversion chamber and the outlet of the confluence chamber is controlled to be no greater than 1 m / s, according to the formula... Calculate the diameters of the splitter cavity and the manifold cavity. ; In step S4, the specific partitioning method adopts the K-means clustering algorithm, using the squared error. The goal is to minimize the heat consumption samples of modules in each group so that the distribution of these samples is as close as possible. In step S5, according to the formula Calculate the traffic allocation requirements for each group of modules; Step S6 includes the sub-step: to The transition section of the branch flow path is designed as a straight orifice, and the liquid flow velocity within the transition section is controlled to be no greater than 0.3 m / s, based on the maximum average heat consumption of the module. Single module traffic and formula Calculate Path of the transition section of the module .

2. The flow distribution design method for a through-type liquid-cooled chassis according to claim 1, characterized in that, In step S7, when performing CFD simulation and iterative calculation, the convergence criterion is that the deviation between the flow allocation value and the flow demand of each group of modules is less than 10%.

3. A flow distribution device for a through-type liquid-cooled chassis, characterized in that, It is designed by the design method described in any one of claims 1 to 2.

Citation Information

Patent Citations

  • Water-cooled parallel flow heat exchanger with multiple rows of flat pipes

    CN106440866A

  • A flow adjusting method and device for a liquid cooling system

    CN109598028A