Server cooling method, device, electronic equipment and storage medium

Abstract

The present application discloses a cooling treatment method, device, electronic device, and storage medium for a server. The method comprises obtaining multiple candidate coolant samples, each having a different concentration; determining a target coolant based on the performance evaluation results of the multiple candidate coolant samples; determining a target component configuration scheme based on the numerical simulation results of the target coolant and multiple candidate component models; configuring a liquid cooling component within the server according to the target component configuration scheme; and cooling the server using the liquid cooling component and the target coolant. This scheme, through the evaluation and simulation of coolant with different concentrations and different component model configuration schemes, can select the optimal coolant concentration and the optimal component model, effectively improving the performance of the liquid cooling component and the cooling efficiency of the server, and ensuring the heat dissipation and cooling effect and operational stability of the server.

Description

Technical Field

[0001] The present application relates to the technical field of server cooling processing, and in particular to a server cooling processing method, device, electronic equipment and storage medium. Background Art

[0002] As the scale and density of data centers continue to increase, liquid cooling technology has gradually become a new trend in data center cooling due to its high efficiency, low noise and small space occupation. At present, traditional coolants may have problems with phase change or corrosion. In addition, the current flow channel design for coolant may not be able to adapt to all scenarios and the cooling strategy cannot be flexibly adjusted. This may lead to uneven heat dissipation or poor cooling effect in some areas in high-density server environments. Summary of the Invention

[0003] The present application provides a cooling processing method, device, electronic device and storage medium for a server, so as to at least solve the current problem of uneven heat dissipation or poor cooling effect in some areas.

[0004] The present application provides a cooling processing method for a server, comprising: obtaining a plurality of candidate coolant samples, each candidate coolant sample having a different concentration;

[0005] determining a target coolant based on performance evaluation results of a plurality of candidate coolant samples;

[0006] Determine the target component configuration scheme based on the numerical simulation results of the target coolant and multiple candidate component models;

[0007] Configure the liquid cooling components in the server according to the target component configuration plan;

[0008] The server is cooled by the liquid cooling components and the target coolant.

[0009] The present application also provides a cooling processing device for a server, comprising: an acquisition module, configured to acquire a plurality of candidate coolant samples, each candidate coolant sample having a different concentration;

[0010] a processing module, configured to determine a target coolant based on performance evaluation results of a plurality of candidate coolant samples;

[0011] The processing module is further used to determine a target component configuration scheme based on numerical simulation results of the target coolant and multiple candidate component models;

[0012] A configuration module, configured to configure the liquid cooling components in the server according to the target component configuration plan;

[0013] The processing module is further used to cool the server through the liquid cooling component and the target coolant.

[0014] The present application also provides an electronic device, comprising: a memory for storing a computer program; and a processor for implementing the steps of any of the above-mentioned server cooling processing methods when executing the computer program.

[0015] The present application also provides a computer-readable storage medium, in which a computer program is stored. When the computer program is executed by a processor, the steps of any of the above-mentioned server cooling processing methods are implemented.

[0016] The present application also provides a computer program product, including a computer program, which implements the steps of any of the above-mentioned server cooling processing methods when the computer program is executed by a processor.

[0017] Through this application, multiple candidate coolant samples are obtained, each with a different concentration; a target coolant is determined based on the performance evaluation results of the multiple candidate coolant samples; a target component configuration scheme is determined based on the numerical simulation results of the target coolant and multiple candidate component models; the liquid cooling components within the server are configured according to the target component configuration scheme; and the server is cooled using the liquid cooling components and the target coolant. In this scheme, by evaluating and simulating coolant with different concentrations and different component model configuration schemes, the optimal coolant concentration and optimal component model can be selected, effectively improving the performance of the liquid cooling components and the cooling efficiency of the server, ensuring the server's heat dissipation and cooling effect and operational stability. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the embodiments of the present application, the following is a brief introduction to the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0019] Figure 1 A process for cooling a server provided in an embodiment of the present application Figure 1 ;

[0020] Figure 2 A process for cooling a server provided in an embodiment of the present application Figure 2 ;

[0021] Figure 3 A structural diagram of a cooling processing device for a server provided in an embodiment of the present application;

[0022] Figure 4 A structural diagram of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0023] The following will be combined with the accompanying drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0024] It should be noted that, in the description of this application, the terms "comprises," "includes," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or device comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or elements inherent to such process, method, article, or device. The terms "first," "second," etc., in this application are used to distinguish similar objects, and are not used to describe a particular order or sequence.

[0025] It should be noted that in the embodiments of this application, words such as "exemplary" or "for example" are used to indicate examples, illustrations, or descriptions. Any embodiment or design described as "exemplary" or "for example" in the embodiments of this application should not be interpreted as being more preferred or advantageous over other embodiments or designs. Rather, the use of words such as "exemplary" or "for example" is intended to present the relevant concepts in a concrete manner.

[0026] With the rapid development of information technology, the scale and density of data centers continue to increase, and server thermal management has become a critical issue. Traditional air cooling technology, due to its inefficiency and high energy consumption, is gradually failing to meet the demands of modern high-density servers. Liquid cooling technology, with its high efficiency, low noise, and small footprint, is becoming a new trend in data center cooling. The application of nanoparticle-enhanced water-based coolants has further improved cooling efficiency, demonstrating the enormous potential of liquid cooling technology for cooling high-density servers.

[0027] In related technologies, although the current more traditional coolants such as water or ethylene glycol solutions have good thermal conductivity, they are prone to phase change or corrosion problems under extreme conditions. The new nanoparticle-enhanced coolant significantly improves the thermal conductivity coefficient of the coolant by adding nanomaterials such as aluminum oxide, while maintaining low viscosity and good chemical stability. Therefore, the coolant material can be flexibly selected to achieve cooling and heat dissipation of the server.

[0028] In the related art, the currently common server cooling system based on liquid cooling convection mainly monitors the server temperature in real time through a temperature monitoring module, and then generates liquid cooling control instructions based on the temperature data. The liquid cooling control module controls the liquid cooling module to perform cooling work through the liquid cooling control instructions. However, the related art does not describe in detail the coolant selection process and its thermal conductivity performance evaluation, and only relies on existing coolants for cooling operations. This may lead to less than ideal cooling efficiency in some high-density server environments. In addition, the related art does not mention the process of optimizing the flow channel design for different server structures and cooling requirements. Using only a fixed flow channel design may not be suitable for all application scenarios, resulting in uneven heat dissipation or poor cooling effect in some areas. Moreover, the related art mainly relies on simple temperature change monitoring and calculation to adjust the cooling strategy, which has limited response speed and cooling accuracy.

[0029] In order to solve the above problems, an embodiment of the present application provides a cooling processing method for a server. By evaluating and simulating coolant of different concentrations and different component model configuration schemes, the optimal coolant concentration and the optimal component model can be selected, thereby effectively improving the performance of the liquid cooling component and the cooling efficiency of the server, ensuring the heat dissipation and cooling effect and operational stability of the server.

[0030] In order to enable those skilled in the art to better understand the present application, the present application is further described in detail below with reference to the accompanying drawings and specific implementation methods.

[0031] like Figure 1 As shown, Figure 1 A flowchart of a cooling method for a server provided in an embodiment of the present application may include the following steps:

[0032] 101. Obtain multiple candidate coolant samples.

[0033] In the embodiment of the present application, in order to select the optimal coolant, a plurality of candidate coolant samples may be obtained first, each candidate coolant sample having a different concentration.

[0034] It should be noted that in the relevant technology, through testing, it can be concluded that although traditional coolants such as water or ethylene glycol solutions have good thermal conductivity, they are prone to phase change or corrosion problems under extreme conditions. The new nanoparticle-enhanced coolant significantly improves the thermal conductivity coefficient of the coolant by adding nanomaterials such as aluminum oxide, while maintaining a low viscosity and good chemical stability. However, the cooling effect of water-based coolants enhanced with aluminum oxide nanoparticles of different concentrations may also be different. Therefore, in order to select the coolant with the best cooling effect, you can first obtain multiple candidate coolant samples of different concentrations.

[0035] 102. Determine a target coolant based on performance evaluation results of multiple candidate coolant samples.

[0036] In an embodiment of the present application, after obtaining multiple candidate coolant samples, in order to select the optimal coolant, it is necessary to perform a performance evaluation on each candidate coolant sample, and then select the coolant sample with the best performance as the target coolant based on the performance evaluation results.

[0037] It should be noted that a performance evaluation is performed on each candidate coolant sample, and some parameters that can be used to represent the cooling and heat dissipation effect of the candidate coolant sample can be calculated as the performance evaluation result, such as thermal conductivity, specific heat capacity, convection heat transfer coefficient, etc. By comparing the above parameters of each candidate coolant sample, a target coolant can be selected, and the cooling and heat dissipation effect of the target coolant will be better than that of other coolants.

[0038] 103. Determine the target component configuration scheme based on the numerical simulation results of the target coolant and multiple candidate component models.

[0039] In an embodiment of the present application, after determining the optimal target coolant, it is also necessary to determine a component configuration scheme. Since the target coolant is in liquid form, cooling and dissipating heat in the server through the target coolant requires a radiator and a transmission pipeline. Therefore, different component configuration schemes may also affect the cooling and dissipation effect of the target coolant. In order to select the optimal component configuration scheme, multiple candidate component models can be obtained first, and the parameters of each component model (such as: pipe width, height, curvature, etc.) are different. Then, the multiple candidate component models are numerically simulated using the target coolant. That is, the target coolant is simulated to flow through each candidate component model to obtain a numerical simulation result for each candidate component model. The numerical simulation result can be used to represent the heat transfer of the candidate component model to the target coolant, thereby selecting the optimal target component configuration scheme. The target component configuration scheme may include: pipe width, height, curvature, radiator position, connection relationship, etc., which are not specifically limited.

[0040] 104. Configure the liquid cooling components in the server according to the target component configuration plan.

[0041] In an embodiment of the present application, once the optimal target component configuration scheme is determined, the corresponding liquid cooling component can be configured in the server according to the target component configuration scheme. The liquid cooling component may include a coolant delivery pipe, a radiator, and connectors, etc. When configuring the liquid cooling component in the server, the target component configuration scheme can be referred to so that the target coolant can flow through the coolant delivery pipe to the key heating area of ​​the server and comply with the optimal pipe path.

[0042] 105. The server is cooled by using a liquid cooling component and a target coolant.

[0043] In the embodiment of the present application, after the liquid cooling component is configured, the server can be cooled by the liquid cooling component and the target coolant. Specifically, the liquid cooling component can be used to make the target coolant flow to the heating area of ​​the server, thereby reducing the temperature of the server through the cooling and heat dissipation performance of the coolant, thereby achieving the cooling purpose of the server.

[0044] In the embodiment of the present application, by evaluating and simulating coolant of different concentrations and different component model configuration schemes, the optimal coolant concentration and the optimal component model can be selected, thereby effectively improving the performance of the liquid cooling component and the cooling efficiency of the server, and ensuring the heat dissipation and cooling effect and operational stability of the server.

[0045] like Figure 2 As shown, Figure 2 Another flow chart of a cooling method for a server provided in an embodiment of the present application, the method may include the following steps:

[0046] 201. Obtain multiple candidate coolant samples.

[0047] In the embodiment of the present application, for the description of step 201, please refer to the detailed description of step 101 in the above embodiment, and the embodiment of the present application will not be repeated.

[0048] 202. The performance of multiple candidate coolant samples is evaluated by the steady-state heat flow method to obtain the thermal conductivity coefficient of each candidate coolant sample.

[0049] In an embodiment of the present application, when performing performance evaluation on multiple candidate coolant samples, the thermal conductivity coefficient of each candidate coolant sample can be calculated. The thermal conductivity coefficient can reflect the ability of the coolant to conduct heat. The higher the thermal conductivity coefficient, the faster the heat is transferred from the heat source to the coolant. In other words, if the thermal conductivity coefficient of a candidate coolant sample is the highest, then it can be said that the efficiency of cooling the server through this candidate coolant sample is the highest. Therefore, the cooling performance of each candidate coolant sample can be evaluated by the thermal conductivity coefficient.

[0050] It should be noted that the steady-state heat flow method is a classic method for measuring thermal conductivity. Its principle is based on Fourier's one-dimensional steady-state heat conduction law. By establishing a stable temperature gradient on both sides of the sample, the heat flow passes vertically through the sample, thereby accurately measuring the thermal conductivity of the material. The core of the steady-state heat flow method is to establish a stable heat conduction state. The sample can be clamped between a hot plate and a cold plate. The hot plate heats and the cold plate cools, forming a stable temperature difference (ΔT). The heat flux density (i.e., the ratio of heat to area) passing through the sample is then measured using a heat flow sensor. The thermal conductivity coefficient is calculated using the following formula:

[0051]

[0052] in, is the thermal conductivity coefficient, For the heat transferred, is the heat transfer area, is the material thickness, is the temperature difference.

[0053] The steady-state heat flow method reduces the impact of transient effects by establishing a stable heat conduction state. The measurement results are highly accurate and are suitable for testing low thermal conductivity materials (such as insulation materials) and high thermal conductivity materials (such as metals and ceramics). In addition, based on Fourier's law, the calculation method is mature and the results are reliable.

[0054] In the embodiment of the present application, a steady-state heat flow test can be performed on each candidate coolant sample in turn, so as to obtain the heat Q, heat transfer area A, and material thickness corresponding to each candidate coolant sample by direct measurement. , and temperature difference , and then put the above parameters into the formula to calculate the thermal conductivity k corresponding to each candidate coolant sample.

[0055] 203. Determine the candidate coolant sample with the largest thermal conductivity as the target coolant.

[0056] In the embodiment of the present application, since the thermal conductivity coefficient can be used to reflect the ability of the coolant to conduct heat, that is, the higher the thermal conductivity coefficient, the faster the heat is transferred from the server to the coolant, and the higher the cooling efficiency, the thermal conductivity coefficients corresponding to the various candidate coolant samples can be compared, and the candidate coolant sample with the largest thermal conductivity coefficient can be determined as the target coolant.

[0057] In some embodiments, experimental tests have shown that when the aluminum oxide concentration is 3 g / L, the thermal conductivity of the coolant reaches a maximum value of approximately 0.65 W / m·K; therefore, an aluminum oxide nanoparticle-reinforced water-based coolant with an aluminum oxide concentration of 3 g / L can be selected as the target coolant.

[0058] 204. Obtain multiple pre-stored candidate component models.

[0059] In an embodiment of the present application, multiple candidate component models can be obtained, and the pipeline parameters of each candidate component model are different. The pipeline parameters may at least include: pipeline width and pipeline height. In addition, since the pipeline in the server may not be completely straight and may be curved, the pipeline parameters may also include curvature.

[0060] 205. Numerical simulation is performed on multiple candidate component models using the target coolant to obtain the heat exchange efficiency corresponding to each candidate component model.

[0061] In an embodiment of the present application, when the server is cooled by the target coolant, different component designs may cause the target coolant to flow to different locations of the server, or pass through different locations of the server, thereby achieving different cooling effects. In order to achieve the highest cooling efficiency, each candidate component model can be numerically simulated, and the heat exchange efficiency corresponding to each candidate component model can be calculated. The heat exchange efficiency can reflect the efficiency of the component model in removing the heat generated inside the server.

[0062] In some embodiments, a computer-aided engineering (CAE) simulation method is used to analyze the heat exchange efficiency of the target coolant under different component configuration designs. Specifically, ANSYS Fluent software can be used as a CAE tool to perform numerical simulations of the heat exchange efficiency of the target coolant under various preset candidate component models.

[0063] In some embodiments, the heat exchange efficiency can be reflected by the heat transfer amount. The higher the heat transfer amount, the greater the corresponding heat exchange efficiency. The heat transfer amount can be calculated by the following formula:

[0064]

[0065] in, is the amount of heat transferred, is the heat transfer coefficient of the coolant, is the heat transfer area, is the temperature difference, is the heat transfer path length.

[0066] In some embodiments, since the coolant transmission pipeline may not be straight and may have bends, when calculating the heat transfer path length, the actual curve length of the pipeline (calculated by tortuosity) rather than the straight distance can be used as part of the heat transfer path length.

[0067] 206. Determine the candidate component model with the highest heat exchange efficiency as the target component configuration solution.

[0068] In an embodiment of the present application, since the heat exchange efficiency can be used to reflect the efficiency of removing the heat generated inside the server, and the heat exchange efficiency is positively correlated with the heat transfer amount, that is, the greater the heat transfer amount, the higher the heat exchange efficiency, the faster the heat generated inside the server is removed, and the higher the cooling efficiency, the heat exchange efficiencies corresponding to each candidate component model can be compared, that is, the heat transfer amounts corresponding to each candidate component model can be compared, and the candidate component model with the highest heat exchange efficiency, that is, the largest heat transfer amount, is determined as the target component configuration scheme.

[0069] In some embodiments, after experimental testing, when the pipe width is 5mm, the height is 3mm, and the curvature is 0.2, the heat transfer The maximum is about 80 W, at which point the heat exchange efficiency reaches its highest, indicating that this pipe design can most effectively remove the heat generated inside the server.

[0070] In some embodiments, numerical simulation using ANSYS Fluent software can not only accurately evaluate the heat exchange efficiency of different candidate component models, but also help identify potential design flaws and improvement directions. By determining the target component configuration scheme through simulation experiments, the heat exchange efficiency can be maximized while reducing flow resistance and energy consumption, ensuring the efficient operation of the cooling system.

[0071] 207. Determine the length of the coolant pipeline based on the target component configuration plan and the volume of the server to be cooled.

[0072] In the embodiment of the present application, the liquid cooling assembly may include a coolant pipe. Since the server may have multiple heat-generating areas, cooling of one area is not sufficient. In addition, the server may have multiple heat-generating areas. Therefore, a pipe of a certain length is required to transport the target coolant. The length of the coolant pipe is at least related to the pipe width, the pipe height, and the volume of the server to be cooled, that is, the volume to be covered (the volume of the effective area or component to be cooled). Therefore, the coolant pipe length can be calculated using the following formula:

[0073]

[0074] in, is the coolant pipe length, is the volume to be covered (volume to be cooled), is the pipe width, is the pipe height.

[0075] 208. Determine the surface area of ​​the radiator based on the heat and temperature data of the server.

[0076] In the embodiment of the present application, the liquid cooling assembly may further include a radiator, which is a device. A larger surface area of ​​the radiator means that more heat can be transferred through convection and radiation, thereby improving the overall heat dissipation efficiency. The larger the surface area of ​​the radiator, the more heat can be transferred per unit time, and thus the temperature inside the server can be more effectively reduced. Therefore, the required radiator surface area can be determined based on the heat and temperature data of the server. The radiator surface area can be specifically calculated using the following formula:

[0077]

[0078] in, is the radiator surface area, is the heat generated from the server, is the effective temperature difference, is the heat dissipation coefficient.

[0079] It's important to note that the effective temperature difference (ETD) refers to the difference between the radiator's surface temperature and the ambient temperature. It's a key indicator of a radiator's performance. A larger ETD means the radiator can transfer heat to the environment more quickly, improving heat dissipation efficiency. Furthermore, the heat dissipation coefficient isn't a fixed value; it's affected by a variety of factors, including the radiator's material, surface treatment, and air flow. A higher ETD indicates a greater heat exchange capacity between the radiator and the environment.

[0080] 209. Configure the liquid cooling components in the server based on the length of the coolant pipes and the surface area of ​​the radiator.

[0081] In an embodiment of the present application, after calculating the length of the coolant pipe and the surface area of ​​the radiator, the coolant pipe and the radiator can be configured. In addition, the liquid cooling assembly can also include a connector that can firmly connect the coolant pipe to the radiator and the cooling inlet and outlet inside the server.

[0082] In some embodiments, after experimental testing, for a standard server module, a coolant pipe of about 10 meters is required to ensure that the coolant can be evenly distributed in all key heating areas, and the surface area of ​​the radiator should be at least 0.5 square meters to ensure sufficient heat exchange area.

[0083] 210. Pressure test the liquid cooling components using the target coolant and monitor coolant leakage in real time.

[0084] In the embodiments of the present application, after configuring the liquid cooling assembly, a preliminary pressure test and leak check is performed on the liquid cooling assembly to detect whether the liquid cooling assembly leaks coolant under a certain pressure. Specifically, a pressure pump can be used to inject the target coolant into the liquid cooling assembly, gradually increasing the pressure to exceed the normal operating pressure, and then maintaining this pressure for a certain period of time to monitor coolant leakage in real time.

[0085] In some embodiments, a pressure pump may be used to inject the target coolant into the liquid cooling assembly, and the pressure may be gradually increased to 1.5 times the normal working pressure, and the pressure may be maintained for 15 minutes, during which time the coolant may be continuously observed for leaks.

[0086] 211. If the coolant leakage condition indicates that no leakage has occurred, the server is cooled using the liquid cooling assembly and the target coolant.

[0087] In the embodiment of the present application, if no leakage is detected, it means that the current liquid cooling component is intact and not damaged, and can withstand a certain pressure and can be used normally. Therefore, the server can be cooled by the liquid cooling component and the target coolant.

[0088] In some embodiments, by accurately calculating the required coolant pipe length and radiator surface area, it is possible to ensure that the coolant in the server is evenly distributed and effectively covers all key heating areas. By performing pressure testing and leak checks before normal use, the sealing of the entire liquid cooling assembly can be guaranteed, ensuring the safety of the server's cooling process.

[0089] 212. Dynamically adjust the coolant flow plan of the target coolant in the liquid cooling component based on the collected server temperature data.

[0090] In an embodiment of the present application, during the process of cooling the server by the target coolant in the liquid cooling assembly, the temperature data of the server will change. Specifically, the temperature of the server will gradually drop and gradually approach the normal temperature. Then, during the temperature change process, the coolant flow scheme of the target coolant can also be adaptively adjusted. For example, when cooling is just beginning, the temperature of the server is very high, so it needs to be cooled quickly first. The flow rate of the target coolant can be increased first. When the temperature of the server is getting closer and closer to the normal temperature, the flow rate of the target coolant can be appropriately reduced. In other words, the coolant flow scheme of the target coolant in the liquid cooling assembly can be dynamically adjusted according to the temperature data of the server. The coolant flow scheme may include parameters such as the flow rate and flow direction of the target coolant.

[0091] In some embodiments, the purpose of cooling the server is to ensure that the temperature of the server is within a normal range, which can also be understood as the difference between the server and the standard temperature value is within a certain range. Therefore, the cooling liquid flow scheme of the target coolant in the liquid cooling component is dynamically adjusted through the collected temperature data of the server. Specifically, it may include: collecting the real-time temperature value of the server according to a preset period; and dynamically adjusting the cooling liquid flow scheme of the target coolant in the liquid cooling component according to the temperature difference between the real-time temperature value and the standard temperature value so that the temperature difference is within a preset temperature difference range.

[0092] In some embodiments, the preset period can be a custom duration, such as 3 seconds, 5 seconds, etc. After the real-time temperature value is collected, the temperature difference between it and the standard temperature value can be calculated. The standard temperature value can be a preset normal temperature value that the server needs to reach. Then, the coolant flow scheme can be adjusted according to the temperature difference, which can be expressed by the following formula:

[0093]

[0094] in, is the target coolant flow output, Is the difference between the real-time temperature value and the standard temperature value, It is a proportional term that can directly generate a control output according to the current temperature difference. The larger the temperature difference, the larger the output of the proportional term, thereby quickly responding to temperature deviations. is the integral term, It can represent the integral of the temperature difference signal from the initial time 0 to time t, which can eliminate the steady-state error of the system. By accumulating the past temperature difference values, even if the temperature difference is small but persists, the integral term will gradually increase the control output until the steady-state error is eliminated. is the differential term, It is the derivative of the temperature difference signal with respect to time, reflecting the rate of change of the temperature difference. It can predict the future temperature difference trend and adjust the control output in advance according to the rate of change of the temperature difference, thereby improving the dynamic response of the system and reducing overshoot and oscillation. 、 、 These are the proportional, integral, and differential gain coefficients, respectively. In the adaptive dynamic adjustment algorithm, the controller monitors the behavior of the server and cooling components in real time and dynamically adjusts these gain coefficients to maintain optimal performance. Adjustments are made based on the system error (i.e., the difference between the set temperature and the actual temperature), the rate of change of the error, and the accumulation of the error over time. By summing the proportional, integral, and differential terms, the system takes into account the current temperature difference, the accumulated temperature difference over time, and the trend of the temperature difference. An appropriate control signal is generated to adjust the coolant flow rate so that the actual temperature reaches the target temperature as quickly and stably as possible.

[0095] In some embodiments, the process of dynamically adjusting the coolant flow pattern of a target coolant in a liquid-cooled assembly can be achieved using an adaptive PID controller, an intelligent control algorithm that automatically adjusts its proportional (P), integral (I), and differential (D) parameters based on system dynamics or environmental conditions. Building on the foundation of traditional PID controllers, this approach introduces an adaptive mechanism, overcoming the limitations of traditional PID in complex nonlinear or time-varying systems and demonstrating enhanced adaptability and robustness. The adaptive PID controller monitors system status (such as errors, system model changes, and disturbances) in real time and dynamically adjusts PID parameters to maintain optimal system performance.

[0096] Maintain stable control performance despite system parameter changes, external disturbances, or nonlinear conditions. Reduce performance degradation caused by modeling inaccuracies or environmental disturbances. Reduce reliance on engineer experience by automatically optimizing parameters through adaptive algorithms. Optimize proportional and differential parameters based on system response speed requirements and actual dynamic characteristics, ensuring a fast and stable system response to setpoint changes or disturbances.

[0097] In some embodiments, in addition to dynamically adjusting the coolant flow scheme, future working conditions can also be predicted. Specifically, this can include: obtaining multiple real-time temperature values ​​within a preset time period; performing temperature prediction on multiple real-time temperature values ​​through a linear regression model to obtain temperature prediction values; and adjusting the coolant flow scheme of the target coolant in the liquid cooling component based on the temperature prediction values.

[0098] It should be noted that the temperature change of the server may have certain regular changes within a certain period of time. Therefore, multiple real-time temperature values ​​within a preset time period can be collected, and then the future temperature trend can be predicted through a linear regression model, which can be expressed by the following formula:

[0099]

[0100] in, represents the predicted temperature value, For time, is the intercept, is the slope.

[0101] Based on the predicted temperature value, the flow rate and direction of the target coolant can be adjusted in advance. If it is predicted that the temperature of certain areas inside the server will rise, the target coolant flow rate in that area can be increased in advance to prevent overheating and ensure the normal operation of the server.

[0102] In some embodiments, during operation of the liquid cooling assembly, temperature uniformity and thermal response time can also be calculated.

[0103] Temperature uniformity is a key indicator for measuring the consistency of temperature distribution within an enclosed space or system. Its core objective is to reflect the temperature differences between different locations. Temperature uniformity is typically expressed using the maximum temperature difference or standard deviation: the maximum temperature difference is the difference between the highest and lowest temperatures in the space, while the standard deviation is the root mean square of the squared deviations of each point's temperature from the average temperature, reflecting the degree of temperature dispersion.

[0104] Among them, thermal response time is a key parameter that describes the time required for the temperature of a system or material to reach a target value or stable state under the stimulation of temperature change. Thermal response time usually refers to the time required for the system to change from the initial temperature state to the target temperature state (or reach a certain proportion of the steady state, such as 90% or 95%).

[0105] In some embodiments, temperature uniformity and thermal response time can be evaluated based on actual measured data or through simulation tools. For example, during the design phase, CAE software might be used to simulate different operating conditions to predict and optimize temperature uniformity and thermal response time. In actual operation, sensor data might be used for real-time monitoring and adjustment. Both temperature uniformity and thermal response time can be used to evaluate the cooling performance of liquid cooling components on servers.

[0106] In some embodiments, since strict data monitoring is required during the operation of the server and the liquid cooling component to ensure safe operation, the setting of sensors needs to ensure full coverage. Therefore, it can specifically include: determining at least one sensor to be operated based on the area to be monitored of the server and the effective monitoring area of ​​the sensor, and the effective monitoring area of ​​at least one sensor to be operated covers the area to be monitored; through at least one sensor to be operated, the real-time temperature value of the server is collected according to a preset period.

[0107] It's important to note that sensor configuration must ensure data collection from all areas of the server. Sensor placement can be determined based on the sensor's effective monitoring area. Of course, sensors also need to be placed in specific locations. For example, distributed fiber optic sensors can be placed along the server motherboard's main heat-generating areas (such as the CPU, GPU, and power supply module). This means running fiber along the top or side of the server chassis, near the main chipset. Thermocouples can be placed at key locations for precise temperature measurement, such as directly on the bottom of the CPU heat sink, near memory modules, or next to the hard drive.

[0108] The number of sensors can be determined by the following formula:

[0109]

[0110] in, is the number of sensors required, is the total area of ​​the server's internal area to be monitored, and is the area of ​​the effective monitoring area of ​​a single sensor.

[0111] In some embodiments, the sensor may include: a distributed fiber optic sensor and a thermocouple. The fiber optic sensor can measure temperature by analyzing the scattering characteristics of the light signal. Specifically, it can detect the change of Raman scattering or Brillouin scattering with temperature, thereby obtaining the temperature distribution along the length of the optical fiber; the thermocouple works based on the Seebeck effect, that is, the voltage difference generated by the temperature difference when two different metals come into contact. By measuring this voltage difference, it can be directly converted into a corresponding temperature reading. Therefore, the thermocouple can directly give the specific temperature value at its location.

[0112] In some embodiments, the location and number of sensors in the server are determined by the coverage relationship and area ratio between the monitoring area of ​​each sensor and the area that needs to be monitored by the server as a whole. This ensures that there are sensors in each heating area of ​​the server for real-time monitoring, avoiding the situation where some areas are not cooled and dissipated in time, and ensuring timely cooling and normal operation of the server.

[0113] In some embodiments, during the operation of the liquid cooling assembly, the real-time heat removal efficiency can also be evaluated. This heat removal efficiency can help evaluate the performance of the liquid cooling assembly in actual operation. By comparing the heat removal efficiencies under different designs or operating conditions, it can be determined which solution is more effective in removing heat generated by the server. The heat removal efficiency can be expressed by the following formula:

[0114]

[0115] in, is the heat removal efficiency, The heat is emitted from the server. The target coolant exchanges heat with the heating element through a radiator or other intermediate heat exchange device. is the specific heat capacity of the target coolant, is the inlet and outlet temperature difference of the target coolant.

[0116] 213. Adjust the server workload using standardized load testing tools.

[0117] In the embodiment of the present application, the server may process various tasks when it is working. When processing complex tasks, the server load will be large, and the heat generation of the server will be more serious. Therefore, the cooling effect of the liquid cooling component can be tested through different loads. Specifically, a standardized load testing tool can be used to gradually adjust the server's workload, which can include: low load, medium load and high load.

[0118] In some embodiments, the load can be gradually increased in the order of low load, medium load, and high load, and the server can be run continuously for a period of time, for example: first run at low load for 10 minutes, then increase to medium load and continue to run for 10 minutes, and then increase to high load and continue to run for 10 minutes.

[0119] 214. When the server is under target load, collect the current temperature data of the server.

[0120] In an embodiment of the present application, when the server is at a target load, the temperature data of the server under the current load can be recorded. The target load can be any one of low load, medium load and high load. That is, when the server is at a low load, the temperature data corresponding to the low load is recorded; when the server load increases to a medium load, the temperature data corresponding to the medium load is recorded; when the server load increases to a high load, the temperature data corresponding to the high load is recorded.

[0121] It should be noted that the temperature data may be temperature data collected at a certain period by distributed optical fiber sensors and thermocouples.

[0122] 215. Determine the load performance test result based on the temperature data and the theoretical temperature data corresponding to the target load.

[0123] In the embodiment of the present application, when the server is under different loads, the corresponding theoretical temperature data is also different. The theoretical temperature data can be understood as the temperature prediction value obtained by predicting the temperature value using the same linear regression model described in the above steps. Then, the root mean square error (RMSE) method can be used to compare the real-time temperature data corresponding to the current load with the theoretical temperature data to evaluate the cooling performance. Specifically, it can be expressed by the following formula:

[0124]

[0125] in, represents the root mean square error comparison result, Represents the actual temperature data collected at the i-th time point, represents the theoretical temperature data corresponding to the i-th time point, and n represents the total number of sampling points, that is, a total of n temperature data at each moment are collected.

[0126] It should be noted that under different load conditions, the temperature data of the server collected can be used to analyze the temperature at each data collection moment using MATLAB data analysis software to plot a temperature change trend graph over time and load. The load performance test results can be obtained by comparing the trend graph with the root mean square error.

[0127] In some embodiments, by simulating different workload conditions through standardized load testing tools, monitoring the temperature changes inside the server in real time through a sensor network, and recording detailed temperature change curves, the performance of the cooling system under various working conditions can be comprehensively evaluated. The root mean square error (RMSE) method is used to quantify the difference between the actual measurement results and the theoretical model predictions. The evaluation method runs through the entire working process of the liquid cooling component, which helps to verify the design rationality of the cooling system, so that potential problems can be discovered and solved in a timely manner, and the cooling strategy can be optimized.

[0128] In some embodiments, when applying a load to a server, it is not added arbitrarily, and there are certain restrictions, which may include: when the server is at the target load, obtaining the power consumption value corresponding to the ambient temperature and the target load; determining the expected temperature data corresponding to the target load based on the power consumption value corresponding to the ambient temperature and the target load; when it is detected that the expected temperature data does not meet the safe temperature range, adjusting the target load.

[0129] It should be noted that for multiple load conditions, the expected maximum tolerable temperature value can be calculated to verify whether the design of the liquid cooling component can effectively control the temperature inside the server under different load conditions. The expected tolerable temperature data can be determined based on the ambient temperature and power consumption value and can be expressed as:

[0130]

[0131] in, For expected temperature data, is the ambient temperature, is the power consumption corresponding to the target load, is the convective heat transfer coefficient, is the heat dissipation area.

[0132] It should be noted that if the calculated expected temperature data exceeds the preset temperature safety range, it indicates that the liquid cooling component may not be sufficient to cope with the current high load situation and needs to be adjusted.

[0133] 216. Output the target component configuration plan, coolant flow plan and load performance test results.

[0134] In an embodiment of the present application, through the configuration of liquid cooling components, dynamic adjustment of coolant flow, load performance testing, etc. described in the above steps, the target component configuration scheme, coolant flow scheme and load performance test results can be summarized and output as a design report.

[0135] In some embodiments, the target component configuration scheme, coolant flow scheme, and load performance test results under different load conditions are organized into a detailed design report, which not only helps to summarize the design and optimization process of the entire liquid cooling component, but also provides reference materials for subsequent research and applications. By systematically presenting the entire process from preliminary design to final testing, it ensures that each link is well documented, further improving the reliability and practicality of the cooling system.

[0136] In some embodiments, the server cooling method provided herein primarily addresses heat dissipation generated by server operation through liquid cooling. With the development of 5G and future mobile communication technologies, mobile base stations (such as macro and micro base stations) and edge computing nodes face similar heat dissipation challenges. These devices are often deployed in locations with limited space and variable environmental conditions, making traditional air cooling difficult to meet their efficient heat dissipation requirements.

[0137] To this end, existing coolants can be further optimized or screened for the unique environments of mobile base stations (such as high outdoor temperatures and fluctuating humidity) to ensure stability over a wider temperature range. Given the compact internal structure of mobile base stations, the flow channels need to be redesigned to accommodate smaller space constraints. For example, a micro-channel design can be used to improve heat exchange efficiency per unit volume. The flow and direction of the coolant can be dynamically adjusted based on the workload characteristics of the base station (such as peak data traffic hours) to ensure optimal heat dissipation under different operating conditions. Because mobile base stations are often located in harsh environments, additional weather resistance testing is required to ensure the cooling system can operate stably under various climatic conditions.

[0138] Through the description of the above implementation methods, those skilled in the art can clearly understand that the method according to the above embodiment can be implemented by means of software plus the necessary general hardware platform, and of course it can also be implemented by hardware, but in many cases the former is a better implementation method.

[0139] like Figure 3 As shown, an embodiment of the present application further provides a cooling processing device for a server, and the cooling processing device for the server may include:

[0140] An acquisition module 301 is configured to acquire a plurality of candidate coolant samples, each of which has a different concentration;

[0141] A processing module 302 is configured to determine a target coolant based on performance evaluation results of a plurality of candidate coolant samples;

[0142] The processing module 302 is further configured to determine a target component configuration scheme based on numerical simulation results of the target coolant and multiple candidate component models;

[0143] Configuration module 303, configured to configure the liquid cooling components in the server according to the target component configuration plan;

[0144] The processing module 302 is further configured to cool the server using the liquid cooling component and the target cooling liquid.

[0145] In some embodiments, the processing module 302 is specifically configured to perform a performance evaluation on a plurality of candidate coolant samples using a steady-state heat flow method to obtain a thermal conductivity coefficient of each candidate coolant sample;

[0146] The processing module 302 is specifically configured to determine the candidate coolant sample with the largest thermal conductivity as the target coolant.

[0147] In some embodiments, the acquisition module 301 is specifically used to acquire a plurality of pre-stored candidate component models, each candidate component model having different pipeline parameters, the pipeline parameters including at least: pipeline width and pipeline height;

[0148] The processing module 302 is specifically configured to perform numerical simulation on multiple candidate component models using a target coolant to obtain a heat exchange efficiency corresponding to each candidate component model;

[0149] The processing module 302 is specifically configured to determine the candidate component model with the highest heat exchange efficiency as the target component configuration solution.

[0150] In some embodiments, the processing module 302 is specifically configured to determine the length of the coolant pipe according to the target component configuration scheme and the volume of the server to be cooled;

[0151] The processing module 302 is specifically configured to determine the surface area of ​​the radiator based on the heat and temperature data of the server;

[0152] The configuration module 303 is specifically used to configure the liquid cooling components in the server according to the length of the coolant pipe and the surface area of ​​the radiator.

[0153] In some embodiments, the processing module 302 is specifically configured to perform a pressure test on the liquid cooling component using a target coolant and monitor coolant leakage in real time;

[0154] The processing module 302 is specifically configured to cool the server using the liquid cooling assembly and the target coolant if the coolant leakage indication indicates that no leakage has occurred.

[0155] In some embodiments, the processing module 302 is further configured to dynamically adjust a cooling liquid flow scheme of a target cooling liquid in the liquid cooling assembly based on the collected temperature data of the server.

[0156] In some embodiments, the acquisition module 301 is specifically configured to collect the real-time temperature value of the server according to a preset period;

[0157] The processing module 302 is specifically configured to dynamically adjust a coolant flow scheme of a target coolant in a liquid cooling assembly according to a temperature difference between a real-time temperature value and a standard temperature value, so that the temperature difference is within a preset temperature difference range.

[0158] In some embodiments, the acquisition module 301 is further configured to acquire a plurality of real-time temperature values ​​within a preset time period;

[0159] The processing module 302 is further configured to perform temperature prediction on the multiple real-time temperature values ​​through a linear regression model to obtain a temperature prediction value;

[0160] The processing module 302 is further configured to adjust a coolant flow plan of the target coolant in the liquid cooling assembly according to the temperature prediction value.

[0161] In some embodiments, the processing module 302 is specifically configured to determine at least one sensor to be operated based on the server's monitored area and the sensor's effective monitoring area, wherein the effective monitoring area of ​​the at least one sensor to be operated covers the monitored area;

[0162] The acquisition module 301 is specifically configured to collect the real-time temperature value of the server through at least one working sensor according to a preset period.

[0163] In some embodiments, the processing module 302 is further configured to adjust the workload of the server using a standardized load testing tool, where the workload includes: low load, medium load, and high load;

[0164] The acquisition module 301 is further configured to collect the current temperature data of the server when the server is at a target load;

[0165] The processing module 302 is further configured to determine a load performance test result based on the temperature data and theoretical temperature data corresponding to the target load.

[0166] In some embodiments, the acquisition module 301 is further configured to acquire the ambient temperature and the power consumption value corresponding to the target load when the server is at the target load;

[0167] The processing module 302 is further configured to determine expected temperature data corresponding to the target load based on the ambient temperature and the power consumption value corresponding to the target load;

[0168] The processing module 302 is further configured to adjust the target load when it is detected that the expected temperature data does not meet the safe temperature range.

[0169] In some embodiments, the processing module 302 is further configured to output a target component configuration solution, a coolant flow solution, and load performance test results.

[0170] In the embodiments of the present application, the description of the features in the embodiments corresponding to the cooling processing device of the server can refer to the relevant description of the embodiments corresponding to the cooling processing method of the server, and will not be repeated here.

[0171] like Figure 4 As shown, an embodiment of the present application further provides an electronic device, including a memory 401 and a processor 402, wherein the memory 401 stores a computer program, and the processor 402 is configured to run the computer program to execute the steps in any of the above-mentioned server cooling processing method embodiments.

[0172] An embodiment of the present application further provides a computer-readable storage medium, in which a computer program is stored. The computer program is configured to execute the steps of any of the above-mentioned server cooling processing method embodiments when running.

[0173] In an exemplary embodiment, the computer-readable storage medium may include, but is not limited to, various media that can store computer programs, such as a USB flash drive, a read-only memory (ROM), a random access memory (RAM), a mobile hard disk, a magnetic disk, or an optical disk.

[0174] An embodiment of the present application further provides a computer program product, which includes a computer program. When the computer program is executed by a processor, the steps in any of the above-mentioned server cooling processing method embodiments are implemented.

[0175] An embodiment of the present application further provides another computer program product, including a non-volatile computer-readable storage medium, wherein the non-volatile computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of any of the above-mentioned server cooling processing method embodiments are implemented.

[0176] Professionals may further appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the above description has generally described the components and steps of each example according to their functions. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professionals and technicians may use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0177] The above is a detailed introduction to the process monitoring of a storage system provided by this application. Specific examples are used herein to illustrate the principles and implementation methods of this application. The description of the above embodiments is only intended to help understand the method and core ideas of this application. It should be pointed out that, for those skilled in the art, without departing from the principles of this application, several improvements and modifications can be made to this application, and these improvements and modifications also fall within the scope of protection of the claims of this application.

Claims

1. A cooling method for a server, characterized in that: The method comprises: obtaining a plurality of candidate coolant samples, each candidate coolant sample having a different concentration; Performing a performance evaluation on the plurality of candidate coolant samples using a steady-state heat flow method to obtain a thermal conductivity coefficient of each candidate coolant sample, wherein the thermal conductivity coefficient is used to indicate a cooling and heat dissipation effect of the candidate coolant sample; determining the candidate coolant sample with the largest heat conductivity coefficient as the target coolant; Determining a target component configuration scheme based on numerical simulation results of the target coolant and multiple candidate component models, wherein the numerical simulation results are used to represent heat transfer conditions of the candidate component models with respect to the target coolant, each candidate component model having different pipeline parameters, the pipeline parameters including at least pipeline width and pipeline height, and the target component configuration scheme being a configuration scheme corresponding to a component model having the largest heat transfer amount among the multiple candidate component models indicated by the numerical simulation results; Determining the length of the coolant pipeline according to the target component configuration scheme and the volume to be cooled of the server; determining a radiator surface area based on the heat of the server, temperature difference data indicating a difference between the radiator surface temperature and the ambient temperature, and a heat dissipation coefficient indicating a heat exchange capacity between the radiator and the environment; configuring a liquid cooling component within the server based on the length of the coolant pipe and the surface area of ​​the radiator; The server is cooled by the liquid cooling component and the target cooling liquid.

2. The method according to claim 1, characterized in that Determining a target component configuration scheme based on numerical simulation results of the target coolant and multiple candidate component models includes: Acquire the plurality of pre-stored candidate component models, each candidate component model having different pipeline parameters, the pipeline parameters including at least pipeline width and pipeline height; Performing numerical simulation on the plurality of candidate component models using the target coolant to obtain a heat exchange efficiency corresponding to each candidate component model; The candidate component model with the highest heat exchange efficiency is determined as the target component configuration solution.

3. The method according to claim 1, characterized in that The cooling process of the server by using the liquid cooling component and the target cooling liquid includes: Performing a pressure test on the liquid cooling component using the target coolant and monitoring coolant leakage in real time; If the coolant leakage condition indicates that no leakage has occurred, the server is cooled using the liquid cooling assembly and the target coolant.

4. The method according to claim 1, wherein The method further comprises: The cooling liquid flow scheme of the target cooling liquid in the liquid cooling component is dynamically adjusted based on the collected temperature data of the server.

5. The method according to claim 4, characterized in that The method of dynamically adjusting the cooling liquid flow scheme of the target cooling liquid in the liquid cooling component based on the collected temperature data of the server includes: Collecting the real-time temperature value of the server according to a preset period; According to the temperature difference between the real-time temperature value and the standard temperature value, the coolant flow scheme of the target coolant in the liquid cooling component is dynamically adjusted so that the temperature difference is within a preset temperature difference range.

6. The method according to claim 5, characterized in that The method further comprises: Get multiple real-time temperature values ​​within a preset time period; Performing temperature prediction on the multiple real-time temperature values ​​through a linear regression model to obtain a temperature prediction value; A cooling liquid flow scheme of the target cooling liquid in the liquid cooling component is adjusted according to the temperature prediction value.

7. The method according to claim 5, characterized in that The collecting of the real-time temperature value of the server according to a preset period includes: Determining at least one sensor to be operated according to the server's monitored area and the effective monitoring area of ​​the sensor, wherein the effective monitoring area of ​​the at least one sensor to be operated covers the monitored area; The real-time temperature value of the server is collected according to the preset period by the at least one working sensor.

8. The method according to claim 1, characterized in that The method further comprises: Adjusting the workload of the server using a standardized load testing tool, wherein the workload includes: low load, medium load, and high load; When the server is at a target load, collecting current temperature data of the server; A load performance test result is determined based on the temperature data and theoretical temperature data corresponding to the target load.

9. The method according to claim 8, characterized in that The method further comprises: When the server is at the target load, obtaining an ambient temperature and a power consumption value corresponding to the target load; Determining expected temperature data corresponding to the target load based on the ambient temperature and the power consumption value corresponding to the target load; When it is detected that the expected withstand temperature data does not meet the safe temperature range, the target load is adjusted.

10. The method according to any one of claims 1 to 9, characterized in that The method further comprises: Output the target component configuration scheme, coolant flow scheme and load performance test results.

11. A cooling device for a server, characterized in that: The device comprises: An acquisition module is used to acquire a plurality of candidate coolant samples, each candidate coolant sample having a different concentration; a processing module, configured to perform a performance evaluation on the plurality of candidate coolant samples using a steady-state heat flow method to obtain a thermal conductivity coefficient of each candidate coolant sample, wherein the thermal conductivity coefficient is used to indicate a cooling and heat dissipation effect of the candidate coolant sample; The processing module is further configured to determine the candidate coolant sample with the largest heat conductivity coefficient as the target coolant; The processing module is further configured to determine a target component configuration scheme based on numerical simulation results of the target coolant and multiple candidate component models, the numerical simulation results being used to represent heat transfer conditions of the candidate component models with respect to the target coolant, each candidate component model having different pipeline parameters, the pipeline parameters including at least pipeline width and pipeline height, the target component configuration scheme being a configuration scheme corresponding to a component model having the largest heat transfer amount among the multiple candidate component models indicated by the numerical simulation results; The processing module is further configured to determine the length of the coolant pipeline according to the target component configuration scheme and the volume to be cooled of the server; The processing module is further configured to determine a surface area of ​​a radiator based on the heat of the server, temperature difference data indicating a difference between a surface temperature of the radiator and an ambient temperature, and a heat dissipation coefficient indicating a heat exchange capacity between the radiator and the ambient environment; a configuration module, configured to configure a liquid cooling component in the server according to the length of the coolant pipe and the surface area of ​​the radiator; The processing module is further configured to cool the server using the liquid cooling assembly and the target cooling liquid.

12. An electronic device, characterized in that: include: memory for storing computer programs; A processor, configured to implement the steps of the server cooling processing method according to any one of claims 1 to 10 when executing the computer program.

13. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, wherein when the computer program is executed by a processor, the steps of the server cooling processing method according to any one of claims 1 to 10 are implemented.

Images