Control method and control system of data center liquid cooling heat dissipation system

Abstract

The invention discloses a control method and a control system of a data center liquid cooling heat dissipation system, and relates to the technical field of liquid cooling heat dissipation, the system realizes real-time state monitoring of server operation through a data acquisition and monitoring module, integrates multiple data such as temperature, flow velocity and pressure difference, forms a detailed comprehensive heat dissipation data set, and improves the heat dissipation efficiency. And a database is constructed for storage. And meanwhile, a cooling liquid flow velocity analysis control module of the system generates a server heat dissipation coefficient CN through the server operation environment parameter C1 and the server operation parameter C2, the server heat dissipation coefficient CN is compared with a preset heat dissipation threshold C for evaluation, and the cooling liquid flow velocity is adjusted in real time. Therefore, when the load of the server is increased, the cooling intensity can be automatically increased, the chip temperature is reduced, and stable operation of the server is kept; and when the load is reduced, the cooling intensity is reduced, the energy consumption is reduced, and the cooling efficiency and the energy utilization rate of the system are remarkably improved.

Description

Technical Field

[0001] The present invention relates to the technical field of liquid cooling and heat dissipation, and in particular to a control method and a control system for a liquid cooling and heat dissipation system in a data center. Background Art

[0002] As the scale of data centers continues to expand, the demand for data storage and computing has also increased dramatically, and energy consumption in the field of information technology has gradually become a core issue that needs to be addressed. In the efficient data processing subfield of the information technology field, heat dissipation is an important part of ensuring the normal operation of the server. In recent years, liquid cooling systems have gradually become a key means of data center cooling technology, especially cold plate liquid cooling systems, which are widely used in large server systems due to their high heat dissipation efficiency and low noise. The cold plate liquid cooling system directly contacts the heat source through the coolant, which can more efficiently remove the heat generated by the server and ensure that the data center can operate efficiently and stably.

[0003] In the Chinese invention application with application publication number CN118170226A, an immersion liquid cooling temperature control system, method, device and medium are disclosed, which relates to the field of server heat dissipation technology and is used to achieve heat dissipation of server motherboards. In view of the problem that traditional solutions cannot accurately control and cool down key heating components or local areas, an immersion liquid cooling temperature control system is provided, which combines the phased array technology currently commonly used in the antenna field with the immersion liquid cooling of the server motherboard. The phased array technology can freely and flexibly interfere with the effect of electromagnetic wave waveforms, and the original phased array technology interference with electromagnetic waves is changed to interference with vibration waves, and then the flow channel and flow rate of the coolant are affected by the vibration waves, so as to achieve flexible and precise control of the flow channel and flow rate of the coolant in the server. In this way, the targeted cooling demand of the local key heating components in the current intensive server motherboard is met, and a better cooling effect is achieved, which is conducive to ensuring the normal and stable operation of the server.

[0004] In combination with the prior art, the above application still has the following deficiencies:

[0005] At present, cold plate liquid cooling systems are widely used in data centers. Although they have high heat dissipation capabilities, they still have obvious regulation problems in actual operation. For example, when the server load fluctuates, the coolant flow rate is often difficult to adapt to the load changes in time, resulting in poor cooling efficiency. Fixed flow rate control modes are difficult to balance the relationship between temperature control and energy consumption. Especially in an environment with frequent load changes, improper coolant flow rate can easily lead to a decrease in cooling effect or an increase in energy consumption. At the same time, liquid cooling systems that lack adaptive regulation functions are also difficult to dynamically respond to changes in load during the heat dissipation process. These deficiencies limit the overall performance of the liquid cooling system. Summary of the invention

[0006] In view of the deficiencies in the prior art, the present invention provides a control method and control system for a liquid cooling and heat dissipation system in a data center, which solves the problems mentioned in the background technology.

[0007] To achieve the above objectives, the present invention is implemented through the following technical solutions: including a data acquisition and monitoring module, a state analysis module, a coolant flow rate analysis control module, a smooth switching control module and a feedback and optimization module;

[0008] The data acquisition and monitoring module is integrated with the integrated sensor group installed at the liquid inlet and outlet of several servers and cold plates to collect the comprehensive heat dissipation data set of each server and cold plate in real time, and build a database to store the collected comprehensive heat dissipation data set;

[0009] The state analysis module is used to construct the cold plate cooling algorithm formula and the immersion cooling algorithm formula, input the comprehensive heat dissipation data set respectively, and calculate to obtain the server operating environment parameter C1 and the server operating parameter C2;

[0010] The coolant flow rate analysis control module is used to comprehensively calculate the server operating environment parameter C1 and the server operating parameter C2 to obtain the server heat dissipation coefficient CN, preset the heat dissipation threshold C and the server heat dissipation coefficient CN, perform preliminary comparative evaluation, and generate a coolant flow rate switching control instruction;

[0011] The smooth switching control module is used to send a control signal through the controller to adaptively adjust the start of cold plate cooling after adjusting the coolant flow rate, and at the same time construct a smooth control algorithm formula to calculate and output the switching smoothness adjustment parameter Stra to start the smooth cooling mode;

[0012] The feedback and optimization module is used to construct a dynamic feedback formula after the coolant flow rate is switched, calculate the output operation effect feedback parameter Cfee, and preset the effect threshold X and the operation effect feedback parameter Cfee for secondary comparative evaluation. According to the secondary comparative evaluation results, the adaptive optimization formula is executed to calculate the output adaptive feedback control factor Cada for adaptive adjustment optimization.

[0013] Preferably, the data acquisition and monitoring module includes a data acquisition unit and a data storage unit;

[0014] The data acquisition unit is integratedly connected to the integrated sensor group installed at the liquid inlet and outlet ports of several servers and cold plates through a sensor interface to collect a comprehensive heat dissipation data set in real time;

[0015] The integrated sensor set includes server pipe monitoring tools, thermocouple sensors, digital temperature and humidity sensors, conductivity sensors, flow meters, differential pressure sensors, and vibration sensors;

[0016] The comprehensive heat dissipation data set includes server load L, chip temperature Tc, coolant inlet and outlet temperature difference △Tw, ambient temperature Ten, coolant conductivity Cco, coolant flow rate Ffl, liquid cooling system pressure difference Pdiff and server vibration frequency Vfreq;

[0017] The data storage unit is used to build a time series database by using the InfluxDB technology, and set the write port and the write port, connect the data acquisition unit to the write port, write the comprehensive heat dissipation data set collected in real time into the time series database, and arrange and store it in the order of collection time.

[0018] Preferably, the status analysis module includes an operating environment analysis unit and a server operation analysis unit;

[0019] The operating environment analysis unit is used to construct a cold plate cooling algorithm formula, extract a comprehensive heat dissipation data set through the write port of the time series database and input it into the cold plate cooling algorithm formula to calculate and output the server operating environment parameter C1;

[0020] The server operating environment parameter C1 is calculated and output by the following cold plate cooling algorithm formula;

[0021] C1=α1·L+β1·(Tc-Ten)+γ1·ΔTw+δ1·Cco+ε1·Pdiff;

[0022] In the formula, (Tc-Ten) represents the temperature difference between the chip temperature and the ambient temperature, α1 represents the server load control coefficient, β1 represents the temperature difference control coefficient between the chip temperature and the ambient temperature, γ1 represents the coolant inlet and outlet temperature difference control coefficient, δ1 represents the coolant conductivity control coefficient, ε1 represents the liquid cooling system pressure difference control coefficient, and the specific values ​​of α1, β1, γ1, δ1 and ε1 are set by the user.

[0023] Preferably, the server operation analysis unit is used to construct an immersion cooling algorithm formula, extract a comprehensive heat dissipation data set through a write port of a time series database and input it into the server operation algorithm formula, and calculate and output a server operation parameter C2;

[0024] The server operating parameter C2 is calculated and output by the following immersion cooling algorithm formula;

[0025] C2=α2·(L-Lth)+β2·(Tc-Tmax)+δ2·Ffl+ε2·Vfreq;

[0026] Wherein, Lth represents the load threshold, which is set based on server design and historical heat dissipation data; Tmax represents the upper limit of chip temperature; α2 represents the control coefficient of the difference between server load and load threshold Lth; β2 represents the control coefficient of the difference between chip temperature and upper limit of chip temperature; δ2 represents the control coefficient of coolant flow rate; ε2 represents the control coefficient of server vibration frequency; and the specific values ​​of α2, β2, δ2 and ε2 are set by users.

[0027] Preferably, the coolant flow rate analysis and control module comprises an adaptive analysis unit and a cooling flow rate control unit;

[0028] The adaptive analysis unit is used to perform comprehensive calculation based on the acquired server operating environment parameter C1 and server operating parameter C2 to obtain the server heat dissipation coefficient CN, and comprehensively analyze the operating status of the cold plate server at the current time t;

[0029] The server heat dissipation coefficient CN is calculated using the following algorithm formula;

[0030] CN(t)=[(C1·A1)+(C2·A2)]+A;

[0031] Wherein, CN(t) represents the server heat dissipation coefficient CN at the current time t, A1 and A2 represent the preset weight values ​​of the server operating environment parameter C1 and the server operating parameter C2 respectively, and C1+C2=1, and the specific value is set by the user;

[0032] The cooling flow rate control unit performs a preliminary comparative evaluation of the preset heat dissipation threshold C and the obtained server heat dissipation coefficient CN based on the standard value of the cold plate server operating power, and generates a coolant flow rate switching control instruction based on the preliminary comparative evaluation result. The specific evaluation contents are as follows;

[0033] When the server heat dissipation coefficient CN is ≥ twice the heat dissipation threshold C, it indicates that the cold plate server is in a load state, and at this time, a first coolant flow rate switching control instruction is sent to adjust the cold plate heat dissipation coolant flow rate to 100%;

[0034] When the server heat dissipation coefficient CN ≥ the heat dissipation threshold C, it indicates that the cold plate server is in a load state, and at this time, a second coolant flow rate switching control instruction is sent to adjust the cold plate heat dissipation coolant flow rate to 65%;

[0035] When the server heat dissipation coefficient CN is less than the heat dissipation threshold C, it indicates that the cold plate server is in a load state. At this time, a third coolant flow rate switching control instruction is sent to reduce the cold plate heat dissipation coolant flow rate to below 45%.

[0036] Preferably, the smooth switching control module is used for, when receiving the first switching logic and the second switching logic, the control system sends an instruction to the cooling management module to switch the coolant flow rate mode, performs the coolant flow rate switching, constructs a smooth control algorithm formula during the switching process, calculates the output smoothness adjustment parameter Stra, and smoothly controls the switching process;

[0037] The smoothness adjustment parameter Stra is calculated and obtained by the following smoothing control algorithm formula;

[0038]

[0039] In the formula, △t represents the time change, △Pdiff represents the pressure difference change of the liquid cooling system, △Ffl represents the coolant flow rate change, η represents the weight coefficient of the pressure difference change of the liquid cooling system, and λ represents the weight coefficient of the coolant flow rate change.

[0040] Preferably, the feedback and optimization module includes a dynamic feedback unit, a feedback analysis unit and an adaptive adjustment unit;

[0041] The dynamic feedback unit is used to continuously monitor the running effect of the server after the cooling mode is smoothly switched, and calculate and output the running effect feedback parameter Cfee by constructing a dynamic feedback formula to feedback the running effect after cooling;

[0042] The operation effect feedback parameter Cfee is calculated and obtained through the following dynamic feedback formula;

[0043]

[0044] Wherein, Ttarget represents the target chip temperature set by the system, Ptarget represents the target pressure of the cooling system, a1, a2 and a3 represent the chip temperature feedback adjustment coefficient, the inlet and outlet water temperature difference feedback adjustment coefficient and the pressure difference feedback adjustment coefficient respectively, and their specific values ​​are set and obtained by the user.

[0045] Preferably, the feedback analysis unit usually provides the optimal operating conditions and temperature range of the equipment through the supplier of the server or cooling system. Based on these data, combined with the heat dissipation capacity of the system, the preset effect threshold X and the obtained operating effect feedback parameter Cfee are used to perform a secondary comparative evaluation to determine the current cooling effect. The specific evaluation content is as follows;

[0046] When the running effect feedback parameter Cfee> effect threshold X, it means that the current cooling effect is abnormal, and the unit is adaptively adjusted at this time;

[0047] When the running effect feedback parameter Cfee≤effect threshold value X, it indicates that the cooling effect is normal. At this time, the current cooling mode is continued to be executed and continuously monitored.

[0048] Preferably, the adaptive adjustment unit is used to calculate and output the adaptive feedback control factor Cada through the constructed adaptive optimization formula when the cooling effect is abnormal during the secondary comparison evaluation, so as to dynamically adjust the cooling mode and cooling intensity;

[0049] The adaptive feedback control factor Cada is calculated by the following adaptive optimization formula:

[0050]

[0051] In the formula, A adaptove represents the weight factor, represents the weighted average value of the historical operation effect feedback parameter, Cfee(i) represents the operation effect feedback parameter at the i-th moment, wi represents the weight value of the operation effect feedback parameter at the i-th moment, and n represents the selected historical period.

[0052] A control method for a liquid cooling system in a data center comprises the following steps:

[0053] S1. By integrating and connecting with the integrated sensor groups installed at the liquid inlets and outlets of several servers and cold plates, the comprehensive heat dissipation data set of each server and cold plate is collected in real time, and a database is constructed to store the collected comprehensive heat dissipation data set;

[0054] S2, constructing a cold plate cooling algorithm formula and an immersion cooling algorithm formula, inputting the comprehensive heat dissipation data set respectively, and performing calculations to obtain a server operating environment parameter C1 and a server operating parameter C2;

[0055] S3, comprehensively calculating the server operating environment parameter C1 and the server operating parameter C2 to obtain the server heat dissipation coefficient CN, presetting the heat dissipation threshold C and the server heat dissipation coefficient CN, performing preliminary comparative evaluation, and generating a coolant flow rate switching control instruction;

[0056] S4. After the coolant flow rate is adjusted and switched, a control signal is sent through the controller to start the cold plate cooling in an adaptive manner, and a smooth control algorithm formula is constructed to calculate and output the switching smoothness adjustment parameter Stra to start the smooth cooling mode;

[0057] S5. Construct a dynamic feedback formula, calculate and output the operation effect feedback parameter Cfee, and preset the effect threshold X and the operation effect feedback parameter Cfee for secondary comparative evaluation. According to the secondary comparative evaluation result, execute the adaptive optimization formula to calculate the output adaptive feedback control factor Cada, and perform adaptive adjustment optimization.

[0058] The present invention provides a control method and control system for a liquid cooling system in a data center, which has the following beneficial effects:

[0059] (1) The system realizes real-time status monitoring of server operation through the data acquisition and monitoring module, integrating multiple sensors such as temperature, flow rate, and pressure difference to form a detailed comprehensive heat dissipation data set. At the same time, the system's coolant flow rate analysis and control module generates the server heat dissipation coefficient CN through the server operating environment parameter C1 and the server operating parameter C2, and compares and evaluates it with the preset heat dissipation threshold C to adjust the coolant flow rate in real time. In this way, when the server load increases, the cooling intensity can be automatically increased, the chip temperature can be reduced, and the server can be kept running stably; when the load decreases, the cooling intensity can be reduced to reduce energy consumption, significantly improving the system's cooling efficiency and energy utilization.

[0060] (2) The system's smooth switching control module effectively alleviates the impact of rapid changes in coolant flow rate on the server by dynamically calculating the smoothness adjustment parameter Stra, so that the coolant flow rate pattern remains stable during the switching process. This design can smoothly manage the coolant when the flow rate is adjusted, avoiding system instability or hardware loss caused by coolant flow rate and pressure fluctuations. Especially in cold plate liquid cooling systems, smooth coolant flow rate switching ensures the reliability and operating life of the equipment, meeting the data center's demand for a high-stability cooling system.

[0061] (3) The feedback and optimization module of the system outputs the operating effect feedback parameter Cfee through a dynamic feedback formula, performs a secondary comparative evaluation with the effect threshold X, automatically determines the cooling effect, and adaptively adjusts the cooling state. If an abnormal cooling effect is detected, the system can calculate the adaptive feedback control factor Cada through the adaptive adjustment unit and perform optimization operations to improve the self-regulation ability of the cooling system under load fluctuations and maintain a stable cooling effect. This module enables the cooling system to have intelligent feedback and self-optimization functions, can adapt to complex operating environments, and significantly enhance the reliability and intelligence of the data center liquid cooling system. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] Figure 1 This is a schematic diagram of a control system flow of a data center liquid cooling and heat dissipation system according to the present invention;

[0063] Figure 2 The present invention is a schematic diagram of the steps of a control method for a liquid cooling and heat dissipation system of a data center. DETAILED DESCRIPTION

[0064] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments 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.

[0065] Example 1

[0066] See also Figure 1 The present invention provides a control system for a liquid cooling system of a data center. To achieve the above purpose, the present invention is implemented through the following technical solutions: including a data acquisition and monitoring module, a state analysis module, a coolant flow rate analysis control module, a smooth switching control module and a feedback and optimization module;

[0067] The data acquisition and monitoring module is integrated with the integrated sensor group installed at the liquid inlet and outlet of several servers and cold plates to collect the comprehensive heat dissipation data set of each server and cold plate in real time, and build a database to store the collected comprehensive heat dissipation data set;

[0068] The state analysis module is used to construct the cold plate cooling algorithm formula and the immersion cooling algorithm formula, input the comprehensive heat dissipation data set respectively, and calculate to obtain the server operating environment parameter C1 and the server operating parameter C2;

[0069] The coolant flow rate analysis control module is used to comprehensively calculate the server operating environment parameter C1 and the server operating parameter C2 to obtain the server heat dissipation coefficient CN, preset the heat dissipation threshold C and the server heat dissipation coefficient CN, perform preliminary comparative evaluation, and generate a coolant flow rate switching control instruction;

[0070] The smooth switching control module is used to trigger when it is determined that the coolant flow rate needs to be switched, send a control signal through the controller to switch the coolant flow rate, and at the same time build a smooth control algorithm formula to calculate and output the switching smoothness adjustment parameter Stra to switch the coolant flow rate smoothly;

[0071] The feedback and optimization module is used to construct a dynamic feedback formula after the coolant flow rate is switched, calculate the output operation effect feedback parameter Cfee, and preset the effect threshold X and the operation effect feedback parameter Cfee for secondary comparative evaluation. According to the secondary comparative evaluation results, the adaptive optimization formula is executed to calculate the output adaptive feedback control factor Cada for adaptive adjustment optimization.

[0072] In this embodiment, the system acquires the heat dissipation data of each server and cold plate in real time through the integrated sensor through the data acquisition and monitoring module, ensures the comprehensiveness and accuracy of data acquisition, and provides reliable support for subsequent analysis. The state analysis module constructs the cold plate and immersion cooling algorithm formula, analyzes and generates the server operating environment parameter C1 and the server operating parameter C2, and then comprehensively calculates the server operating environment parameter C1 and the server operating parameter C2 to obtain the server heat dissipation coefficient CN, presets the heat dissipation threshold C and the server heat dissipation coefficient CN, performs preliminary comparative evaluation, and generates the coolant flow rate switching control instruction. When it is necessary to switch the coolant flow rate mode, the smooth switching control module realizes the smooth switching of the flow rate by constructing a smooth control algorithm and calculating the switching smoothness adjustment parameter Stra, ensuring that the temperature fluctuation is minimized and avoiding the system instability caused by the sudden change of the flow rate. Finally, the feedback and optimization module performs dynamic feedback evaluation after switching, analyzes the cooling effect and adaptively optimizes the dynamic adjustment of the control factor Cada by comparing the setting effect threshold X and the operating feedback parameter Cfee, thereby realizing the continuous optimization of the cooling process. Compared with the existing technology, the innovation of this system lies in the intelligent decision-making of coolant flow rate control in an adaptive manner, which greatly improves the system's response speed to server load fluctuations and energy efficiency management level, reduces the energy consumption of the cooling system, and achieves a more energy-saving and stable liquid cooling effect, effectively extending the service life of server equipment and further improving the operating efficiency and reliability of the data center.

[0073] Example 2

[0074] This embodiment is explained in Example 1, please refer to Figure 1 ,Specifically: the data acquisition and monitoring module includes a data acquisition unit and a data storage unit;

[0075] The data acquisition unit is integratedly connected to the integrated sensor group installed at the liquid inlet and outlet ports of several servers and cold plates through a sensor interface to collect a comprehensive heat dissipation data set in real time;

[0076] The integrated sensor set includes server pipe monitoring tools, thermocouple sensors, digital temperature and humidity sensors, conductivity sensors, flow meters, differential pressure sensors, and vibration sensors;

[0077] The comprehensive heat dissipation data set includes server load L, chip temperature Tc, coolant inlet and outlet temperature difference △Tw, ambient temperature Ten, coolant conductivity Cco, coolant flow rate Ffl, liquid cooling system pressure difference Pdiff and server vibration frequency Vfreq;

[0078] The data storage unit is used to build a time series database by using the InfluxDB technology, and set the write port and the write port, connect the data acquisition unit to the write port, write the comprehensive heat dissipation data set collected in real time into the time series database, and arrange and store it in the order of collection time.

[0079] In this embodiment, the data acquisition and monitoring module in the system collects and processes key heat dissipation data of the server and cold plate in real time through the cooperation of the data acquisition unit and the data storage unit. Through the integrated sensor group, the system can comprehensively monitor multiple key parameters such as server load, chip temperature, inlet and outlet temperature difference of coolant, ambient temperature, etc., to form a comprehensive heat dissipation data set. These data are stored in a time series database through InfluxDB technology to ensure that they are recorded and managed in chronological order, making the system more accurate in tracking, analyzing and optimizing data. This module achieves higher real-time performance and data accuracy, and can capture more detailed heat dissipation parameters. Through this efficient data acquisition and storage mechanism, the system can accurately identify problems of insufficient or excessive heat dissipation, optimize the performance of the cooling system, and significantly improve the energy efficiency of the data center and the stability of the system.

[0080] Example 3

[0081] This embodiment is explained in Example 2. Please refer to Figure 1 ,Specifically: the state analysis module includes an operation environment analysis unit and a server operation analysis unit;

[0082] The operating environment analysis unit is used to construct a cold plate cooling algorithm formula, extract a comprehensive heat dissipation data set through the write port of the time series database and input it into the cold plate cooling algorithm formula to calculate and output the server operating environment parameter C1;

[0083] The server operating environment parameter C1 is calculated and output by the following cold plate cooling algorithm formula;

[0084] C1=α1·L+β1·(Tc-Ten)+γ1·ΔTw+δ1·Cco+ε1·Pdiff;

[0085] In the formula, (Tc-Ten) represents the temperature difference between the chip temperature and the ambient temperature, α1 represents the server load control coefficient, β1 represents the temperature difference control coefficient between the chip temperature and the ambient temperature, γ1 represents the coolant inlet and outlet temperature difference control coefficient, δ1 represents the coolant conductivity control coefficient, ε1 represents the liquid cooling system pressure difference control coefficient, and the specific values ​​of α1, β1, γ1, δ1 and ε1 are set by the user.

[0086] The server operation analysis unit is used to construct an immersion cooling algorithm formula, extract a comprehensive heat dissipation data set through the write port of the time series database and input it into the server operation algorithm formula to calculate and output the server operation parameter C2;

[0087] The server operating parameter C2 is calculated and output by the following immersion cooling algorithm formula;

[0088] C2=α2·(L-Lth)+β2·(Tc-Tmax)+δ2·Ffl+ε2·Vfreq;

[0089] Wherein, Lth represents the load threshold, which is set based on server design and historical heat dissipation data; Tmax represents the upper limit of chip temperature; α2 represents the control coefficient of the difference between server load and load threshold Lth; β2 represents the control coefficient of the difference between chip temperature and upper limit of chip temperature; δ2 represents the control coefficient of coolant flow rate; ε2 represents the control coefficient of server vibration frequency; and the specific values ​​of α2, β2, δ2 and ε2 are set by users.

[0090] In this embodiment, the system extracts comprehensive heat dissipation data from the time series database in real time based on the cold plate cooling algorithm formula through the operating environment analysis unit, calculates the server operating environment parameter C1, accurately reflects the server operating environment, and enables the cold plate cooling mode to adapt to dynamically changing load requirements; the server operation analysis unit calculates the server operating parameter C2 based on the immersion cooling algorithm formula, thereby accurately judging the load level, temperature difference and coolant flow rate, and adapting to the heat dissipation requirements of the server under high load. By calculating the server operating environment parameter C1 and the server operating parameter C2 respectively, the system can flexibly select the appropriate cooling mode, and intelligently switch the coolant flow rate when the load suddenly changes or the temperature exceeds the limit, to ensure the balance between cooling efficiency and energy consumption.

[0091] Example 4

[0092] This embodiment is explained in Example 3, please refer to Figure 1 ,Specifically: the coolant flow rate analysis and control module includes an adaptive analysis unit and a cooling flow rate control unit;

[0093] The adaptive analysis unit is used to perform comprehensive calculation based on the acquired server operating environment parameter C1 and server operating parameter C2 to obtain the server heat dissipation coefficient CN, and comprehensively analyze the operating status of the cold plate server at the current time t;

[0094] The server heat dissipation coefficient CN is calculated using the following algorithm formula;

[0095] CN(t)=[(C1·A1)+(C2·A2)]+A;

[0096] Wherein, CN(t) represents the server heat dissipation coefficient CN at the current time t, A1 and A2 represent the preset weight values ​​of the server operating environment parameter C1 and the server operating parameter C2 respectively, and C1+C2=1, and the specific value is set by the user;

[0097] The cooling flow rate control unit performs a preliminary comparative evaluation of the preset heat dissipation threshold C and the obtained server heat dissipation coefficient CN based on the standard value of the cold plate server operating power, and generates a coolant flow rate switching control instruction based on the preliminary comparative evaluation result. The specific evaluation contents are as follows;

[0098] When the server heat dissipation coefficient CN is ≥ twice the heat dissipation threshold C, it indicates that the cold plate server is in a load state, and at this time, a first coolant flow rate switching control instruction is sent to adjust the cold plate heat dissipation coolant flow rate to 100%;

[0099] When the server heat dissipation coefficient CN ≥ the heat dissipation threshold C, it indicates that the cold plate server is in a load state, and at this time, a second coolant flow rate switching control instruction is sent to adjust the cold plate heat dissipation coolant flow rate to 65%;

[0100] When the server heat dissipation coefficient CN is less than the heat dissipation threshold C, it indicates that the cold plate server is in a load state. At this time, a third coolant flow rate switching control instruction is sent to reduce the cold plate heat dissipation coolant flow rate to below 45%.

[0101] In this embodiment, the system calculates the server heat dissipation coefficient CN based on the server operating environment parameter C1 and the server operating parameter C2 through an adaptive analysis unit, thereby comprehensively evaluating the current operating status of the cold plate server. The cooling flow rate control unit uses the server heat dissipation coefficient CN as a basis, combined with the preset heat dissipation threshold C, to perform three-level flow rate switching control: automatically increase the coolant flow rate to 100% under high load conditions to ensure sufficient heat dissipation capacity; adjust the flow rate to 65% under medium load to balance energy consumption and heat dissipation requirements; and reduce the flow rate to below 45% under low load conditions to achieve higher energy utilization. Compared with the traditional constant flow rate cooling solution, this module introduces an adaptive control mechanism that enables the cooling system to flexibly adjust the flow rate as the server load changes, so that precise cooling and the lowest energy consumption can still be achieved in an environment with load fluctuations.

[0102] Example 5

[0103] This embodiment is explained in Example 4. Please refer to Figure 1Specifically: the smooth switching control module is used for sending a command to switch the coolant flow rate mode to the cooling management module when receiving the first switching logic and the second switching logic, switching the coolant flow rate, constructing a smooth control algorithm formula during the switching process, calculating the output smoothness adjustment parameter Stra, and smoothly controlling the switching process;

[0104] The smoothness adjustment parameter Stra is calculated and obtained by the following smoothing control algorithm formula;

[0105]

[0106] In the formula, △t represents the time change, △Pdiff represents the pressure difference change of the liquid cooling system, △Ffl represents the coolant flow rate change, η represents the weight coefficient of the pressure difference change of the liquid cooling system, and λ represents the weight coefficient of the coolant flow rate change.

[0107] In this embodiment, the smooth switching control module of the system ensures that the coolant flow rate conversion is smooth by intelligently controlling the coolant flow rate switching process. When receiving the first switching logic or the second switching logic, the module immediately sends a switching instruction to the cooling management module, and calculates the smoothness adjustment parameter Stra based on the constructed smooth control algorithm formula. In the formula, the time change Δt, the pressure difference change ΔPdiff of the liquid cooling system, the coolant flow rate change ΔFfl and their corresponding weight coefficients are combined to ensure that the pressure and flow rate adjustment of the system during the switching process are gradually transitioned, thereby avoiding sudden fluctuations in temperature and pressure. The smooth switching control module greatly reduces the instability of the system during the coolant flow rate switching process, and prevents hardware failures or performance degradation caused by sudden changes in pressure or abnormal flow rate. Through smooth control, the switching process is safer, the heat dissipation efficiency is maximized, and the system energy efficiency and equipment service life are significantly improved. This improvement enables the system to seamlessly switch the coolant flow rate mode under complex load conditions, ensuring the continuous and efficient operation of the data center while reducing energy consumption and maintenance costs.

[0108] Example 6

[0109] This embodiment is explained in Example 5. Please refer to Figure 1 ,Specifically: the feedback and optimization module includes a dynamic feedback unit, a feedback ,analysis unit and an adaptive adjustment unit;

[0110] The dynamic feedback unit is used to continuously monitor the running effect of the server after the cooling mode is smoothly switched, and calculate and output the running effect feedback parameter Cfee by constructing a dynamic feedback formula to feedback the running effect after cooling;

[0111] The operation effect feedback parameter Cfee is calculated and obtained through the following dynamic feedback formula;

[0112]

[0113] Wherein, Ttarget represents the target chip temperature set by the system, Ptarget represents the target pressure of the cooling system, a1, a2 and a3 represent the chip temperature feedback adjustment coefficient, the inlet and outlet water temperature difference feedback adjustment coefficient and the pressure difference feedback adjustment coefficient respectively, and their specific values ​​are set and obtained by the user.

[0114] The feedback analysis unit usually provides the optimal operating conditions and temperature range of the equipment through the supplier of the server or cooling system. Based on these data, combined with the heat dissipation capacity of the system, the preset effect threshold X and the obtained operating effect feedback parameter Cfee are used for secondary comparative evaluation to determine the current cooling effect. The specific evaluation contents are as follows;

[0115] When the running effect feedback parameter Cfee> effect threshold X, it means that the current cooling effect is abnormal, and the unit is adaptively adjusted at this time;

[0116] When the running effect feedback parameter Cfee≤effect threshold value X, it indicates that the cooling effect is normal. At this time, the current cooling mode is continued to be executed and continuously monitored.

[0117] The adaptive adjustment unit is used to calculate and output the adaptive feedback control factor Cada through the constructed adaptive optimization formula when the cooling effect is abnormal in the secondary comparative evaluation, so as to dynamically adjust the cooling mode and cooling intensity;

[0118] The adaptive feedback control factor Cada is calculated by the following adaptive optimization formula:

[0119]

[0120] In the formula, A adaptove represents the weight factor, represents the weighted average value of the historical operation effect feedback parameter, Cfee(i) represents the operation effect feedback parameter at the i-th moment, wi represents the weight value of the operation effect feedback parameter at the i-th moment, and n represents the selected historical period.

[0121] In this embodiment, the feedback and optimization module of the system realizes real-time monitoring, performance evaluation and automatic optimization of the cooling system through a dynamic feedback unit, a feedback analysis unit and an adaptive adjustment unit. First, after the cooling mode is smoothly switched, the dynamic feedback unit continuously monitors the operation effect of the server, calculates the operation effect feedback parameter Cfee using a dynamic feedback formula, and evaluates the cooling effect of the system. Then, the feedback analysis unit compares the feedback result with the target chip temperature Ttarget, the target pressure Ptarget and the preset effect threshold X set by the system to determine whether the current cooling is normal. If Cfee exceeds the effect threshold X, the system will start adaptive optimization through the adaptive adjustment unit, construct an adaptive feedback control factor Cada, and dynamically adjust the cooling mode and intensity according to the historical data and the current data to ensure the continuous optimization of the heat dissipation effect. The introduction of this module greatly improves the intelligence and adaptive ability of the cooling system. Compared with the passive adjustment of the traditional cooling system, the feedback and optimization module can actively monitor the system operation status, and optimize and adjust in time when an abnormality is found to ensure that the cooling effect is always in the best state. This adaptive optimization mechanism based on the combination of historical data and real-time feedback not only effectively improves the cooling efficiency, but also reduces the energy waste and hardware loss caused by overcooling or insufficient heat dissipation.

[0122] Example 7

[0123] See also Figure 1 and Figure 2 , a control method for a data center liquid cooling system, comprising the following steps:

[0124] S1. By integrating and connecting with the integrated sensor groups installed at the liquid inlets and outlets of several servers and cold plates, the comprehensive heat dissipation data set of each server and cold plate is collected in real time, and a database is constructed to store the collected comprehensive heat dissipation data set;

[0125] S2, constructing a cold plate cooling algorithm formula and an immersion cooling algorithm formula, inputting the comprehensive heat dissipation data set respectively, and performing calculations to obtain a server operating environment parameter C1 and a server operating parameter C2;

[0126] S3, comprehensively calculating the server operating environment parameter C1 and the server operating parameter C2 to obtain the server heat dissipation coefficient CN, presetting the heat dissipation threshold C and the server heat dissipation coefficient CN, performing preliminary comparative evaluation, and generating a coolant flow rate switching control instruction;

[0127] S4. After the coolant flow rate is adjusted and switched, a control signal is sent through the controller to start the cold plate cooling in an adaptive manner, and a smooth control algorithm formula is constructed to calculate and output the switching smoothness adjustment parameter Stra to start the smooth cooling mode;

[0128] S5. Construct a dynamic feedback formula, calculate and output the operation effect feedback parameter Cfee, and preset the effect threshold X and the operation effect feedback parameter Cfee for secondary comparative evaluation. According to the secondary comparative evaluation result, execute the adaptive optimization formula to calculate the output adaptive feedback control factor Cada, and perform adaptive adjustment optimization.

[0129] While the embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that many changes, modifications, substitutions and variations can be made to the embodiments without departing from the principles and spirit of the invention.

Claims

1. A control system for a liquid cooling system in a data center, characterized in that: It includes data acquisition and monitoring module, state analysis module, coolant flow rate analysis control module, smooth switching control module and feedback and optimization module; The data acquisition and monitoring module is integratedly connected with the integrated sensor group installed at the liquid inlet and outlet of several servers and cold plates to collect the comprehensive heat dissipation data set of each server and cold plate in real time, and build a database to store the collected comprehensive heat dissipation data set; The state analysis module is used to construct a cold plate cooling algorithm formula and an immersion cooling algorithm formula, input the comprehensive heat dissipation data set respectively, and calculate to obtain the server operating environment parameter C1 and the server operating parameter C2; The coolant flow rate analysis control module is used to comprehensively calculate the server operating environment parameter C1 and the server operating parameter C2 to obtain the server heat dissipation coefficient CN, preset the heat dissipation threshold C and the server heat dissipation coefficient CN, perform preliminary comparative evaluation, and generate a coolant flow rate switching control instruction; The smooth switching control module is used to send a control signal through the controller to adaptively adjust the start of cold plate cooling after adjusting the coolant flow rate, and at the same time construct a smooth control algorithm formula to calculate and output the switching smoothness adjustment parameter Stra to start the smooth cooling mode; The feedback and optimization module is used to construct a dynamic feedback formula after the coolant flow rate is switched, calculate the output operation effect feedback parameter Cfee, and preset the effect threshold X and the operation effect feedback parameter Cfee for secondary comparative evaluation, and according to the secondary comparative evaluation result, execute the adaptive optimization formula to calculate the output adaptive feedback control factor Cada, and perform adaptive adjustment optimization.

2. The control system of a data center liquid cooling system according to claim 1, characterized in that: The data acquisition and monitoring module includes a data acquisition unit and a data storage unit; The data acquisition unit is integratedly connected with the integrated sensor group installed at the liquid inlet and outlet of several servers and cold plates through the sensor interface to collect the comprehensive heat dissipation data set in real time; The integrated sensor group includes a server pipe monitoring tool, a thermocouple sensor, a digital temperature and humidity sensor, a conductivity sensor, a flow meter, a differential pressure sensor, and a vibration sensor; The comprehensive heat dissipation data set includes server load L, chip temperature Tc, coolant inlet and outlet temperature difference ΔTw, ambient temperature Ten, coolant conductivity Cco, coolant flow rate Ffl, liquid cooling system pressure difference Pdiff and server vibration frequency Vfreq; The data storage unit is used to build a time series database by using InfluxDB technology, and set a write port and a write port, connect the data acquisition unit to the write port, write the comprehensive heat dissipation data set collected in real time into the time series database, and arrange and store it in order of collection time.

3. The control system of a data center liquid cooling system according to claim 1, characterized in that: The state analysis module includes an operating environment analysis unit and a server operation analysis unit; The operating environment analysis unit is used to construct a cold plate cooling algorithm formula, extract a comprehensive heat dissipation data set through the write port of the time series database and input it into the cold plate cooling algorithm formula to calculate and output the server operating environment parameter C1; The server operating environment parameter C1 is calculated and output by the following cold plate cooling algorithm formula; C1=α1·L+β1·(Tc-Ten)+γ1·ΔTw+δ1·Cco+ε1·Pdiff; In the formula, (Tc-Ten) represents the temperature difference between the chip temperature and the ambient temperature, α1 represents the server load control coefficient, β1 represents the temperature difference control coefficient between the chip temperature and the ambient temperature, γ1 represents the coolant inlet and outlet temperature difference control coefficient, δ1 represents the coolant conductivity control coefficient, ε1 represents the liquid cooling system pressure difference control coefficient, and the specific values ​​of α1, β1, γ1, δ1 and ε1 are set by the user.

4. The control system of a data center liquid cooling system according to claim 3, characterized in that: The server operation analysis unit is used to construct an immersion cooling algorithm formula, extract a comprehensive heat dissipation data set through the write port of the time series database and input it into the server operation algorithm formula to calculate and output the server operation parameter C2; The server operating parameter C2 is calculated and output by the following immersion cooling algorithm formula; C2=α2·(L-Lth)+β2·(Tc-Tmax)+δ2·Ffl+ε2·Vfreq; Wherein, Lth represents the load threshold, which is set based on server design and historical heat dissipation data; Tmax represents the upper limit of chip temperature; α2 represents the control coefficient of the difference between server load and load threshold Lth; β2 represents the control coefficient of the difference between chip temperature and upper limit of chip temperature; δ2 represents the control coefficient of coolant flow rate; ε2 represents the control coefficient of server vibration frequency; and the specific values ​​of α2, β2, δ2 and ε2 are set by users.

5. The control system of a data center liquid cooling system according to claim 4, characterized in that: The coolant flow rate analysis and control module includes an adaptive analysis unit and a cooling flow rate control unit; The adaptive analysis unit is used to perform comprehensive calculation to obtain the server heat dissipation coefficient CN based on the acquired server operating environment parameter C1 and server operating parameter C2, and comprehensively analyze the operating status of the cold plate server at the current time t; The server heat dissipation coefficient CN is calculated by the following algorithm formula: CN(t)=[(C1·A1)+(C2·A2)]+A; Wherein, CN(t) represents the server heat dissipation coefficient CN at the current time t, A1 and A2 represent the preset weight values ​​of the server operating environment parameter C1 and the server operating parameter C2 respectively, and C1+C2=1, and the specific value is set by the user; The cooling flow rate control unit performs a preliminary comparative evaluation of the preset heat dissipation threshold C and the obtained server heat dissipation coefficient CN based on the standard value of the cold plate server operating power, and generates a coolant flow rate switching control instruction based on the preliminary comparative evaluation result. The specific evaluation content is as follows; When the server heat dissipation coefficient CN is ≥ twice the heat dissipation threshold C, it indicates that the cold plate server is in a load state, and at this time, a first coolant flow rate switching control instruction is sent to adjust the cold plate heat dissipation coolant flow rate to 100%; When the server heat dissipation coefficient CN ≥ the heat dissipation threshold C, it indicates that the cold plate server is in a load state, and at this time, a second coolant flow rate switching control instruction is sent to adjust the cold plate heat dissipation coolant flow rate to 65%; When the server heat dissipation coefficient CN is less than the heat dissipation threshold C, it indicates that the cold plate server is in a load state. At this time, a third coolant flow rate switching control instruction is sent to reduce the cold plate heat dissipation coolant flow rate to below 45%.

6. The control system of a data center liquid cooling system according to claim 1, characterized in that: The smooth switching control module is used to control the coolant flow rate after the coolant flow rate is adjusted and switched, and the control system sends a coolant flow rate switching control instruction to the coolant management module to switch the coolant flow rate. During the switching process, a smooth control algorithm formula is constructed to calculate the output smoothness adjustment parameter Stra to smoothly control the switching process; The smoothness adjustment parameter Stra is calculated and obtained by the following smoothing control algorithm formula; In the formula, △t represents the time change, △Pdiff represents the pressure difference change of the liquid cooling system, △Ffl represents the coolant flow rate change, η represents the weight coefficient of the pressure difference change of the liquid cooling system, and λ represents the weight coefficient of the coolant flow rate change.

7. The control system of a data center liquid cooling system according to claim 6, characterized in that: The feedback and optimization module includes a dynamic feedback unit, a feedback analysis unit and an adaptive adjustment unit; The dynamic feedback unit is used to continuously monitor the running effect of the server after the cooling mode is smoothly switched, and calculate and output the running effect feedback parameter Cfee by constructing a dynamic feedback formula to feedback the running effect after cooling; The operation effect feedback parameter Cfee is calculated and obtained by the following dynamic feedback formula: Wherein, Ttarget represents the target chip temperature set by the system, Ptarget represents the target pressure of the cooling system, a1, a2 and a3 represent the chip temperature feedback adjustment coefficient, the inlet and outlet water temperature difference feedback adjustment coefficient and the pressure difference feedback adjustment coefficient respectively, and their specific values ​​are set and obtained by the user.

8. The control system of a data center liquid cooling system according to claim 7, characterized in that: The feedback analysis unit performs a secondary comparative evaluation by using the preset effect threshold value X and the obtained operation effect feedback parameter Cfee to determine the current cooling effect. The specific evaluation contents are as follows; When the running effect feedback parameter Cfee> effect threshold X, it means that the current cooling effect is abnormal, and the unit is adaptively adjusted at this time; When the running effect feedback parameter Cfee≤effect threshold value X, it indicates that the cooling effect is normal. At this time, the current cooling mode is continued to be executed and continuously monitored.

9. The control system of a data center liquid cooling system according to claim 7, characterized in that: The adaptive adjustment unit is used to calculate and output the adaptive feedback control factor Cada through the constructed adaptive optimization formula when the cooling effect of the secondary comparison evaluation is abnormal, so as to dynamically adjust the cooling control of the coolant; The adaptive feedback control factor Cada is calculated and obtained by the following adaptive optimization formula: In the formula, A adaptove represents the weight factor, represents the weighted average value of the historical operation effect feedback parameter, Cfee(i) represents the operation effect feedback parameter at the i-th moment, wi represents the weight value of the operation effect feedback parameter at the i-th moment, and n represents the selected historical period.

10. A control method for a liquid cooling and heat dissipation system in a data center, applied to a control system of a liquid cooling and heat dissipation system in a data center according to any one of claims 1 to 9, characterized in that: The following steps are involved: S1. By integrating and connecting with the integrated sensor groups installed at the liquid inlets and outlets of several servers and cold plates, the comprehensive heat dissipation data set of each server and cold plate is collected in real time, and a database is constructed to store the collected comprehensive heat dissipation data set; S2, constructing a cold plate cooling algorithm formula and an immersion cooling algorithm formula, inputting the comprehensive heat dissipation data set respectively, and performing calculations to obtain a server operating environment parameter C1 and a server operating parameter C2; S3, comprehensively calculating the server operating environment parameter C1 and the server operating parameter C2 to obtain the server heat dissipation coefficient CN, presetting the heat dissipation threshold C and the server heat dissipation coefficient CN, performing preliminary comparative evaluation, and generating a coolant flow rate switching control instruction; S4. After the coolant flow rate is adjusted and switched, a control signal is sent through the controller to start the cold plate cooling in an adaptive manner, and a smooth control algorithm formula is constructed to calculate and output the switching smoothness adjustment parameter Stra to start the smooth cooling mode; S5. Construct a dynamic feedback formula, calculate and output the operation effect feedback parameter Cfee, and preset the effect threshold X and the operation effect feedback parameter Cfee for secondary comparative evaluation. According to the secondary comparative evaluation result, execute the adaptive optimization formula to calculate the output adaptive feedback control factor Cada, and perform adaptive adjustment optimization.

Images