Segmented control cold plate type liquid cooling structure supporting intelligent temperature control

Abstract

The invention relates to the technical field of heat dissipation, and discloses a subsection control cold plate type liquid cooling structure supporting intelligent temperature control, the subsection control cold plate type liquid cooling structure comprises a base, the left and right sides of the base are fixedly connected with supporting legs, the left and right sides of the top of the base are fixedly connected with supporting plates, and uniformly distributed cooling plates are mounted between the two supporting plates; computing equipment is arranged between every two adjacent cooling plates, a supporting frame is fixedly connected between the tops of the two supporting legs, and a fixing plate is fixedly connected to the top of the supporting frame. A cooling plate is divided into a plurality of independent cooling sections, each section covers computing equipment, meanwhile, a temperature data acquisition module in a central controller acquires the real-time temperature of each area, a data analysis and processing unit analyzes temperature data by using a deep learning algorithm, and a decision control module dynamically adjusts the opening degree of an electric valve according to the temperature data. Accurate heat dissipation is achieved, consumption of cooling resources is reduced during low load, and the purpose of saving energy is achieved.

Description

Technical Field

[0001] The present invention relates to the field of heat dissipation technology, and in particular to a segmented controlled cold plate type liquid cooling structure supporting intelligent temperature control. Background Art

[0002] With the rapid development of information technology, the demand for computing power in data centers, high-performance computing, and edge computing is growing exponentially. In these high-density computing environments, the heat generated by electronic equipment continues to rise. Efficient and precise heat dissipation technology has become the key to ensuring stable operation of equipment and extending its service life. Traditional air cooling technology has been unable to meet the needs of efficient heat dissipation, and liquid cooling technology has gradually gained favor due to its efficient thermal conductivity.

[0003] Existing liquid cooling systems are often unable to accurately control the temperature differences in different areas, resulting in energy waste and affecting the life of the equipment. The current common practice is to use a single controller to adjust the fluid speed in the entire cooling loop in an attempt to balance the thermal load of the entire machine. This control strategy ignores the different heat source distributions inside the equipment. When the workload changes dynamically, it is easy to cause local overheating or over-cooling, affecting the life of the equipment and causing energy waste. Summary of the invention

[0004] In view of the deficiencies in the prior art, the present invention provides a segmented controlled cold plate liquid cooling structure that supports intelligent temperature control, which solves the problem of local overheating or over-cooling, which affects the life of the equipment and causes energy waste.

[0005] To achieve the above objectives, the present invention is implemented through the following technical solutions: a segmented control cold plate liquid cooling structure supporting intelligent temperature control, including a base, legs are fixedly connected on the left and right sides of the base, support plates are fixedly connected on the left and right sides of the top of the base, evenly distributed cooling plates are installed between the two support plates, computing equipment is arranged between two adjacent cooling plates, a support frame is fixedly connected between the tops of the two legs, a fixed plate is fixedly connected to the top of the support frame, a central controller is fixedly connected at the center position of the top of the fixed plate, a return water pipe is fixedly connected to the front side of the left leg, a water supply pipe is fixedly connected to the rear side of the right leg, connecting pipes are installed on the left and right sides of the cooling plate, evenly distributed shunt pipes are fixedly connected to the opposite side of the water supply pipe and the return water pipe, delivery pipes are fixedly connected to the opposite ends of the connecting pipes on the left and right sides, and electric valves are installed between the two adjacent delivery pipes and the shunt pipes.

[0006] Preferably, evenly distributed grooves are formed on opposite sides of the two support plates, and the cooling plate is slidably connected to the grooves of the support plates.

[0007] Preferably, the connecting pipes on the left and right sides are located at diagonal positions on the left and right sides of the cooling plate, and the connecting pipes are fixedly connected to the cooling plate by clamps.

[0008] Preferably, the central controller is used to intelligently control the temperature of the computing device, and when the temperature of the computing device rises, the opening of the electric valve connecting the corresponding cooling plate and the connecting pipe is increased.

[0009] Preferably, the central controller reduces the opening of the electric valve when the computing device is under low load.

[0010] Preferably, the central controller includes a temperature monitoring module, a data analysis module, a deep learning algorithm module and a control decision module; The temperature monitoring module is used to obtain temperature data in the computing device in real time; The data analysis module is used to analyze and process the acquired temperature data to determine the heat load mode; The deep learning algorithm module is used to predict the heat load pattern in the future period based on historical data and real-time data; The control decision module is used to control the opening of each electric valve according to the thermal load prediction result.

[0011] Preferably, the temperature monitoring module includes a temperature sensor unit, a data acquisition unit and a data transmission unit. The temperature sensor unit is used to accurately collect real-time temperature data of each area. The data acquisition unit is used to collect temperature data acquired by the temperature sensor in real time. The data transmission unit is used to transmit the collected temperature data to the data analysis module. The temperature sensor units are evenly distributed inside the computing device.

[0012] Preferably, the data analysis module includes a data receiving unit, a data processing unit and a pattern recognition unit, the data receiving unit is used to receive temperature data from the temperature monitoring module, the data processing unit is used to analyze and calculate the temperature data using a specific algorithm, and the pattern recognition unit is used to identify the heat load pattern based on the processed data.

[0013] Preferably, the deep learning algorithm module includes a model generation unit, a data input unit, a model training unit and a prediction generation unit. The model generation unit is used to generate a model and initialize mechanical energy. The data input unit is used to receive historical data, real-time data and equipment working status information. The model training unit trains and optimizes the deep learning model based on the input data. The prediction generation unit is used to generate prediction results of future thermal loads.

[0014] Preferably, the control decision module includes a prediction result receiving unit, a scheme calculation unit and a valve control unit. The prediction result receiving unit is used to obtain the thermal load prediction result generated by the deep learning algorithm module, the scheme calculation unit is used to calculate the optimal fluid distribution plan according to the prediction result, and the valve control unit is used to send the calculated control instructions to each electric valve to achieve precise adjustment of the valve opening.

[0015] Working principle: The coolant starts from the water supply main and enters each cooling plate through evenly distributed diversion pipes. In the cooling plate, the coolant absorbs the heat generated by the computing equipment, and then flows back to the return water main through the connecting pipe and the delivery pipe, thereby realizing the transfer and dissipation of heat; During the whole process, the electric valve controls the flow of coolant, and the temperature sensor units evenly distributed inside the computing device collect temperature data of each area in real time. The data acquisition unit then collects this data, and then the data transmission unit transmits it to the data analysis module in the central controller; After the data receiving unit of the data analysis module receives the temperature data, the data processing unit performs analysis and calculation, and the pattern recognition unit recognizes the heat load pattern based on the processed data; The model generation unit of the deep learning algorithm module generates and initializes the model, the data input unit receives historical data, real-time data and equipment working status information, the model training unit trains and optimizes the deep learning model based on these input data, and the prediction generation unit generates the thermal load prediction results for a period of time in the future; After the prediction result receiving unit of the control decision module obtains the heat load prediction result, the solution calculation unit calculates the best fluid distribution plan according to the prediction result, and the valve control unit sends the calculated control instruction to the electric valve to achieve precise adjustment of the valve opening; When the temperature of the computing equipment rises, the central controller monitors this change in real time, and increases the opening of the corresponding electric valve according to the preset control strategy, increases the flow of coolant, achieves rapid heat dissipation, and avoids damage to the equipment caused by overheating. When the computing equipment is under low load, the central controller reduces the opening of the electric valve and reduces the flow of coolant to achieve energy conservation and rational use of resources. The evenly distributed grooves opened on the opposite sides of the two support plates provide a directional sliding track for the cooling plate, limiting its unnecessary movement, facilitating installation, disassembly and maintenance, ensuring close contact between the cooling plate and the computing equipment, and facilitating heat transfer. The connecting pipes at the diagonal positions on the left and right sides of the cooling plate are fixedly connected by clamps to promote coolant convection, make the coolant more evenly distributed in the cooling plate, improve heat dissipation efficiency, enhance connection stability, reduce leakage risks, reduce the occurrence rate of computing equipment failures, and extend its service life.

[0016] The present invention provides a segmented controlled cold plate liquid cooling structure supporting intelligent temperature control. It has the following beneficial effects: 1. The present invention is divided into multiple independent cooling sections by using cooling plates, each section covers the computing equipment. At the same time, the temperature data acquisition module in the central controller can accurately obtain the real-time temperature of each area through the temperature sensor unit. The data analysis and processing unit uses a deep learning algorithm to analyze the temperature data. The decision control module dynamically adjusts the opening of the electric valve accordingly to achieve precise heat dissipation, reduce cooling resource consumption at low load, and achieve energy saving.

[0017] 2. Through the independent design of each cooling section and the accurate prediction of future heat load changes by the heat load prediction unit, the present invention enables the central controller to adjust the cooling strategy in advance to avoid damage to the equipment caused by temperature fluctuations. Stable temperature control ensures that all components of the equipment are always at a suitable working temperature, reduces the occurrence of failures, and significantly extends the service life of the equipment.

[0018] 3. The present invention can flexibly adapt to changes in different computing tasks and workloads by supporting a segmented control cold plate liquid cooling system with intelligent temperature control, and can provide the best heat dissipation solution regardless of the computing intensity. This modular structure and intelligent central controller make troubleshooting and maintenance more convenient, reduce maintenance costs and downtime, and improve the overall reliability and availability of computing equipment. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 A three-dimensional diagram of a segmented controlled cold plate liquid cooling structure supporting intelligent temperature control according to the present invention; Figure 2 It is a front view schematic diagram of a segmented control cold plate type liquid cooling structure supporting intelligent temperature control of the present invention; Figure 3 It is a right side schematic diagram of the segmented control cold plate type liquid cooling structure supporting intelligent temperature control of the present invention; Figure 4 A schematic top view of a segmented controlled cold plate liquid cooling structure supporting intelligent temperature control according to the present invention; Figure 5 A schematic diagram of the system framework of the central controller of the present invention; Figure 6 This is a schematic diagram of a temperature monitoring module of a central controller of the present invention; Figure 7 A schematic diagram of a data analysis module of a central controller of the present invention; Figure 8 This is a schematic diagram of the deep learning algorithm module of the central controller of the present invention; Fig. 9 Schematic diagram of the control decision module of the central controller of the present invention.

[0020] Among them, 1. base; 2. support legs; 3. delivery pipe; 4. electric valve; 5. water supply pipe; 6. central controller; 7. fixed plate; 8. support frame; 9. return water pipe; 10. computing equipment; 11. cooling plate; 12. support plate; 13. connecting pipe; 14. diversion pipe. DETAILED DESCRIPTION

[0021] The technical solution of the present invention will be described clearly and completely below in conjunction with the accompanying drawings of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0022] Please see attached Figure 1 -Attached Figure 4 The embodiment of the present invention provides a segmented control cold plate liquid cooling structure supporting intelligent temperature control, including a base 1, legs 2 are fixedly connected to the left and right sides of the base 1, support plates 12 are fixedly connected to the left and right sides of the top of the base 1, a uniformly distributed cooling plate 11 is installed between the two support plates 12, a computing device 10 is arranged between two adjacent cooling plates 11, a support frame 8 is fixedly connected between the tops of the two legs 2, a fixed plate 7 is fixedly connected to the top of the support frame 8, a central controller 6 is fixedly connected to the center position of the top of the fixed plate 7, a return water pipe 9 is fixedly connected to the front side of the left leg 2, a water supply pipe 5 is fixedly connected to the rear side of the right leg 2, connecting pipes 13 are installed on the left and right sides of the cooling plate 11, a uniformly distributed shunt pipe 14 is fixedly connected to the opposite side of the water supply pipe 5 and the return water pipe 9, a delivery pipe 3 is fixedly connected to the opposite ends of the left and right connecting pipes 13, and an electric valve 4 is installed between the two adjacent delivery pipes 3 and the shunt pipe 14.

[0023] The coolant enters the cooling plate 11 from the water supply main pipe 5 through the shunt pipe 14, absorbs heat and then returns to the return water main pipe 9 through the connecting pipe 13 and the delivery pipe 3, thereby realizing heat transfer and dissipation. The flow rate of the coolant is controlled by the electric valve 4 to ensure the smooth flow of the coolant in the entire cycle, thereby realizing heat exchange, taking away the heat generated by the computing device 10, avoiding local heat accumulation, ensuring the stability and reliability of the equipment operation, enabling it to operate continuously and stably under high load conditions, and avoiding performance degradation or failure due to insufficient heat dissipation.

[0024] Please see attached Figure 2 The two support plates 12 have evenly distributed grooves on opposite sides, and the cooling plate 11 is slidably connected with the grooves of the support plates 12 .

[0025] The groove design provides a directional sliding track for the cooling plate 11, limiting its unnecessary movement, while supporting the cooling plate 11, making it easy to maintain and replace the cooling plate 11, ensuring close contact between the cooling plate 11 and the computing device 10, and facilitating the effective transfer of heat. This sliding connection method not only facilitates the installation and disassembly of the cooling plate 11, but also enables the cooling plate 11 to maintain a stable position during operation.

[0026] Please see attached Figure 2 The left and right connecting pipes 13 are located at diagonal positions on the left and right sides of the cooling plate 11, and the connecting pipes 13 are fixedly connected to the cooling plate 11 by clamps.

[0027] The diagonal position setting promotes the convection of the coolant, and the clamp fixation provides sufficient tightening force to prevent coolant leakage and reduced heat dissipation effect caused by loose connection, so that the coolant can be more evenly distributed in the cooling plate 11, thereby improving the heat dissipation efficiency and enhancing the stability of the connection, reducing the risk of leakage, reducing the occurrence rate of failure of the computing device 10, and extending the service life of the computing device 10.

[0028] Please see attached Figure 3 The central controller 6 is used to intelligently control the temperature of the computing device 10. When the temperature of the computing device 10 increases, the opening of the electric valve 4 connected to the corresponding cooling plate 11 and the connecting pipe 13 is increased.

[0029] The central controller 6 monitors the temperature changes in real time, and sends instructions to increase the opening of the electric valve 4 according to the preset control strategy, thereby increasing the supply of coolant and achieving rapid heat dissipation. It can respond to the increase in temperature of the computing device 10 in a timely and effective manner, and quickly reduce the temperature by increasing the valve opening and increasing the coolant flow, thereby avoiding damage to the equipment caused by overheating.

[0030] When the computing device 10 is under low load, the central controller 6 reduces the opening of the electric valve 4 .

[0031] When the computing device 10 is under low load, the valve opening is reduced, and the flow rate of the coolant is reduced, thereby achieving energy conservation and rational use of resources.

[0032] Please see attached Figure 5 -Attached Fig. 9 , the central controller 6 includes a temperature monitoring module, a data analysis module, a deep learning algorithm module and a control decision module; The temperature monitoring module is used to obtain temperature data in the computing device 10 in real time; The data analysis module is used to analyze and process the acquired temperature data to determine the heat load mode; The deep learning algorithm module is used to predict the heat load pattern in the future based on historical data and real-time data; The control decision module is used to control the opening of each electric valve 4 according to the thermal load prediction result.

[0033] Through the temperature sensor unit component, the temperature information of different parts of the computing device 10 is collected in real time, providing real-time basic data for the temperature control of the central controller 6, so that the central controller 6 can quickly respond to temperature changes. The data analysis module helps the central controller 6 to understand the thermal state of the computing device 10 more comprehensively and accurately. Then, the deep learning algorithm module uses deep learning technology and algorithms to train and learn a large amount of data to establish a prediction model. Finally, the control decision module generates corresponding control instructions based on the prediction results, and passes the instructions to the electric valve 4 to adjust its opening, thereby achieving precise control of the electric valve 4 and ensuring that the flow of the coolant can be adjusted in time according to changes in the heat load.

[0034] Please see attached Figure 6 The temperature monitoring module includes a temperature sensor unit, a data acquisition unit and a data transmission unit. The temperature sensor unit is used to accurately collect real-time temperature data of each area. The data acquisition unit is used to collect temperature data obtained by the temperature sensor in real time. The data transmission unit is used to transmit the collected temperature data to the data analysis module. The temperature sensor units are evenly distributed inside the computing device 10.

[0035] The temperature of each area of ​​the computing device 10 is monitored and data is transmitted. The temperature sensor unit senses the temperature, the data acquisition unit collects the data, and the data transmission unit transmits the data, thereby jointly completing the task of temperature monitoring.

[0036] Please see attached Figure 7 The data analysis module includes a data receiving unit, a data processing unit and a pattern recognition unit. The data receiving unit is used to receive the temperature data from the temperature monitoring module. The data processing unit is used to analyze and calculate the temperature data using a specific algorithm. The pattern recognition unit is used to identify the heat load pattern based on the processed data.

[0037] It can process and analyze the temperature data comprehensively and deeply, accurately identify the heat load mode, provide strong support for the control decision of the central controller 6, improve the utilization value of the data, and enable the central controller 6 to make reasonable control decisions based on accurate analysis results.

[0038] Please see attached Figure 8The deep learning algorithm module includes a model generation unit, a data input unit, a model training unit and a prediction generation unit. The model generation unit is used to generate a model and initialize mechanical energy. The data input unit is used to receive historical data, real-time data and equipment working status information. The model training unit trains and optimizes the deep learning model based on the input data. The prediction generation unit is used to generate prediction results of future heat loads.

[0039] By establishing a deep learning model, accurate prediction of future heat loads can be achieved, so that the central controller 6 can control the corresponding advance cooling plate 11 to prepare for heat dissipation and further optimize the heat dissipation effect.

[0040] Please see attached Fig. 9 The control decision module includes a prediction result receiving unit, a scheme calculation unit and a valve control unit. The prediction result receiving unit is used to obtain the heat load prediction result generated by the deep learning algorithm module. The scheme calculation unit is used to calculate the optimal fluid distribution plan according to the prediction result. The valve control unit is used to send the calculated control instructions to each electric valve to achieve precise adjustment of the valve opening.

[0041] It can quickly and accurately calculate the best control scheme based on the prediction results, and realize precise adjustment of the opening of the electric valve 4 to ensure the optimal heat dissipation effect, ensure the scientificity and effectiveness of the system's control decision, and realize precise control of the heat dissipation process.

[0042] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A segmented controlled cold plate liquid cooling structure supporting intelligent temperature control, comprising a base (1), characterized in that: The left and right sides of the base (1) are fixedly connected to legs (2); the left and right sides of the top of the base (1) are fixedly connected to support plates (12); a cooling plate (11) is evenly distributed between the two support plates (12); a computing device (10) is arranged between two adjacent cooling plates (11); a support frame (8) is fixedly connected between the tops of the two legs (2); a fixing plate (7) is fixedly connected to the top of the support frame (8); a central control is fixedly connected at the center of the top of the fixing plate (7). The cooling plate (11) comprises a cooling device (6), a return water pipe (9) fixedly connected to the front side of the left support leg (2), a supply water pipe (5) fixedly connected to the rear side of the right support leg (2), connecting pipes (13) installed on both the left and right sides of the cooling plate (11), evenly distributed shunt pipes (14) fixedly connected to the side opposite to the return water pipe (9) of the supply water pipe (5), and a delivery pipe (3) fixedly connected to the opposite ends of the connection pipes (13) on the left and right sides, and an electric valve (4) installed between two adjacent delivery pipes (3) and the shunt pipe (14).

2. The segmented controlled cold plate liquid cooling structure supporting intelligent temperature control according to claim 1 is characterized in that: Evenly distributed grooves are provided on opposite sides of the two support plates (12), and the cooling plate (11) is slidably connected to the grooves of the support plates (12).

3. The segmented controlled cold plate liquid cooling structure supporting intelligent temperature control according to claim 1 is characterized in that: The connecting pipes (13) on the left and right sides are located at diagonal positions on the left and right sides of the cooling plate (11), and the connecting pipes (13) and the cooling plate (11) are fixedly connected by clamps.

4. The segmented controlled cold plate liquid cooling structure supporting intelligent temperature control according to claim 1 is characterized in that: The central controller (6) is used to intelligently control the temperature of the computing device (10). When the temperature of the computing device (10) rises, the opening of the electric valve (4) connected to the corresponding cooling plate (11) and the connecting pipe (13) is increased.

5. The segmented controlled cold plate liquid cooling structure supporting intelligent temperature control according to claim 1 is characterized in that: The central controller (6) reduces the opening of the electric valve (4) when the computing device (10) is under low load.

6. The segmented controlled cold plate liquid cooling structure supporting intelligent temperature control according to claim 1 is characterized in that: The central controller (6) comprises a temperature monitoring module, a data analysis module, a deep learning algorithm module and a control decision module; The temperature monitoring module is used to obtain temperature data in the computing device (10) in real time; The data analysis module is used to analyze and process the acquired temperature data to determine the heat load mode; The deep learning algorithm module is used to predict the heat load pattern in the future period based on historical data and real-time data; The control decision module is used to control the opening of each electric valve (4) according to the thermal load prediction result.

7. The segmented controlled cold plate liquid cooling structure supporting intelligent temperature control according to claim 6 is characterized in that: The temperature monitoring module comprises a temperature sensor unit, a data acquisition unit and a data transmission unit. The temperature sensor unit is used to accurately collect real-time temperature data of each area. The data acquisition unit is used to collect temperature data acquired by the temperature sensor in real time. The data transmission unit is used to transmit the acquired temperature data to the data analysis module. The temperature sensor units are evenly distributed inside the computing device (10).

8. The segmented controlled cold plate liquid cooling structure supporting intelligent temperature control according to claim 6 is characterized in that: The data analysis module includes a data receiving unit, a data processing unit and a pattern recognition unit. The data receiving unit is used to receive temperature data from the temperature monitoring module. The data processing unit is used to analyze and calculate the temperature data using a specific algorithm. The pattern recognition unit is used to identify the heat load pattern based on the processed data.

9. The segmented controlled cold plate liquid cooling structure supporting intelligent temperature control according to claim 6 is characterized in that: The deep learning algorithm module includes a model generation unit, a data input unit, a model training unit and a prediction generation unit. The model generation unit is used to generate a model and initialize mechanical energy. The data input unit is used to receive historical data, real-time data and equipment working status information. The model training unit trains and optimizes the deep learning model based on the input data. The prediction generation unit is used to generate prediction results of future thermal loads.

10. The segmented controlled cold plate liquid cooling structure supporting intelligent temperature control according to claim 6, characterized in that: The control decision module includes a prediction result receiving unit, a scheme calculation unit and a valve control unit. The prediction result receiving unit is used to obtain the thermal load prediction result generated by the deep learning algorithm module. The scheme calculation unit is used to calculate the optimal fluid distribution plan according to the prediction result. The valve control unit is used to send the calculated control instructions to each electric valve to achieve precise adjustment of the valve opening.

Images