Heat dissipation control method, device, equipment, storage medium and program product

Abstract

The present application discloses a heat dissipation control method, device, equipment, storage medium and program product, which relate to the field of server heat dissipation technology. By obtaining the current data of the heat dissipation components, the thermal status of the server can be monitored in real time. Then, these data are input into a pre-established heat dissipation model. The model is based on the server's thermal balance equation and state space equation and can accurately predict the temperature adjustment amount. This enables the system to actively adjust the heat dissipation strategy instead of passively responding to temperature changes, reducing dependence on experimental data and avoiding complex PID parameter adjustments. By dynamically controlling according to the temperature adjustment amount, the system can quickly respond to temperature changes, optimize fan speed, reduce energy consumption, and improve the operational stability and reliability of the server, thereby solving the heat dissipation control technical problems and achieving the heat dissipation control technical effect.

Description

Technical Field

[0001] The present application relates to the field of server heat dissipation technology, and in particular to heat dissipation control methods, devices, equipment, storage media, and program products. Background Art

[0002] With the rapid development of information technology, the improvement of server performance has led to an increase in the heat generated by internal server components, posing new challenges to the server's heat dissipation control.

[0003] In related technologies, the Proportional-Integral-Derivative (PID) control method can be used for heat dissipation control. However, this method has caused overshoot and oscillation, affecting the stable operation of the server. Summary of the Invention

[0004] The present application provides a heat dissipation control method, apparatus, device, storage medium, and program product, which improve the efficiency and accuracy of heat dissipation control, avoid overshoot oscillation, and improve the stability of the server.

[0005] This application provides a heat dissipation control method, including:

[0006] Acquire current data of heat dissipation parameters of a heat dissipation component of the server to be processed;

[0007] Input the current data into the heat dissipation model corresponding to the server to be processed to obtain the corresponding temperature adjustment amount; the heat dissipation model is determined in advance based on the heat balance equation and state space equation of the server to be processed; the heat balance equation is determined based on the heat dissipation system architecture of the processing server;

[0008] The heat dissipation of the server to be processed is controlled according to the temperature adjustment amount.

[0009] The present application also provides a heat dissipation control device, comprising:

[0010] An acquisition module, used to acquire current data of heat dissipation parameters of a heat dissipation component of a server to be processed;

[0011] An input module is used to input current data into the heat dissipation model corresponding to the server to be processed to obtain the corresponding temperature adjustment amount; the heat dissipation model is determined in advance based on the heat balance equation and state space equation of the server to be processed; the heat balance equation is determined based on the heat dissipation system architecture of the processing server;

[0012] The control module is used to control the heat dissipation of the server to be processed according to the temperature adjustment amount.

[0013] 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 heat dissipation control methods when executing the computer program.

[0014] 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 heat dissipation control methods are implemented.

[0015] The present application also provides a computer program product, including a computer program, which implements the steps of any of the above-mentioned heat dissipation control methods when executed by a processor.

[0016] By using this application, the current data of the heat dissipation components is obtained, and the thermal status of the server can be monitored in real time. This data is then input into a pre-established heat dissipation model, which is based on the server's thermal balance equation and state-space equation and can accurately predict the temperature adjustment amount. This enables the system to actively adjust the heat dissipation strategy rather than passively responding to temperature changes, reducing dependence on experimental data and avoiding complex PID parameter adjustments. By dynamically controlling according to the temperature adjustment amount, the system can quickly respond to temperature changes, optimize fan speed, reduce energy consumption, and improve the operational stability and reliability of the server. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] 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.

[0018] Figure 1 A flow chart of a heat dissipation control method in related art;

[0019] Figure 2 A schematic diagram of an application scenario of the heat dissipation control method provided in an embodiment of the present application;

[0020] Figure 3 Schematic diagram of the heat dissipation control method provided in this embodiment of the application Figure 1 ;

[0021] Figure 4 Schematic diagram of the heat dissipation control method provided in this embodiment of the application Figure 2 ;

[0022] Figure 5 A schematic diagram of the structure of a heat dissipation control device provided in an embodiment of the present application;

[0023] Figure 6This is a schematic diagram of the structure of the electronic device provided in this application. DETAILED DESCRIPTION

[0024] 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.

[0025] 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.

[0026] With the rapid development of information technology, servers, as core devices for data storage and processing, are crucial for ensuring the continuity and quality of various online services. However, as server performance continues to improve, the heat generated by internal components (such as the Central Processing Unit (CPU), Graphics Processing Unit (GPU), and memory) has also increased dramatically, posing a severe challenge to the server's cooling system. In the field of server cooling technology, air cooling is widely used. This method regulates the temperature of internal components by controlling fan speed, thereby preventing performance degradation or damage caused by overheating.

[0027] In related technologies, due to the inability to establish an accurate server heat dissipation model, only Proportional-Integral-Derivative (PID) is usually used as a heat dissipation control method. For example, Figure 1 As shown, temperature control can be performed based on the temperature control curve and PID. In the specific implementation process, the server periodically collects the temperature of each component. If the temperature is lower than the preset temperature value, the fan speed is controlled by the temperature control curve. If the temperature is greater than the preset temperature, the fan speed is controlled by PID.

[0028] However, the temperature control curve used in the above method requires extensive test data to fit and establish, and for servers with different configurations, the temperature control curve often requires re-data collection and establishment. The PID control employed adjusts the control input based on the deviation between the current system state and the target state. Adjusting PID parameters is complex and often requires readjustment for different server models. PID control also has limited accuracy and is prone to overshoot and oscillation, impacting the stable operation of the server.

[0029] To address the above technical issues, the inventors of this application have discovered that it is possible to construct a heat dissipation model that accurately reflects the server's heat dissipation process, thereby improving server heat dissipation efficiency, reducing energy consumption, and ensuring stable server operation. Specifically, a heat balance equation for the server's heat dissipation system can be constructed based on the heat dissipation system architecture of an air-cooled server. Furthermore, a linearized state-space equation for the server's heat dissipation system can be constructed based on the heat balance equation, and a heat dissipation model can be established based on this heat dissipation model. Heat dissipation control can then be performed based on this heat dissipation model. Based on this, an embodiment of this application provides a heat dissipation control method.

[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] In conjunction with the specific application environment architecture or specific hardware architecture on which the execution of the heat dissipation control method depends, the specific application environment architecture or specific hardware architecture is described here. Figure 2 , Figure 2 Schematic diagram of the application scenario of the heat dissipation control method provided in the embodiment of the present application. Figure 2 As shown, the air-cooled server includes a GPU, memory, CPU, and fan group.

[0032] During implementation, the heat-generating components (e.g., GPU, memory, CPU) and heat-dissipating components (e.g., fan assembly) in the cooling system architecture of an air-cooled server can be determined. Based on this cooling system architecture, a heat balance equation for the air-cooled server can be constructed. The heat balance equation is then linearized to obtain a linearized state-space equation. A heat dissipation model for the air-cooled server can then be constructed based on this state-space equation. In practical applications of heat dissipation control, current data on the heat dissipation parameters of the heat dissipation components of the server being processed can be obtained. This current data can be input into the corresponding heat dissipation model for the server being processed to obtain a corresponding temperature adjustment value. Heat dissipation control can then be performed on the server being processed based on the temperature adjustment value. The heat dissipation control method provided in this embodiment monitors the server's thermal status in real time by acquiring current data from the heat dissipation components. This data is then input into a pre-established heat dissipation model, which accurately predicts the temperature adjustment value based on the server's heat balance equation and state-space equation. This enables the system to proactively adjust its heat dissipation strategy rather than passively responding to temperature changes, reducing reliance on experimental data and avoiding complex PID parameter adjustments. By dynamically controlling based on the temperature adjustment value, the system can quickly respond to temperature changes, optimize fan speed, reduce energy consumption, and improve server operational stability and reliability.

[0033] Figure 3 Schematic diagram of the heat dissipation control method provided in this embodiment of the application Figure 1 ,like Figure 3 As shown, an embodiment of the present application provides a heat dissipation control method, which is described in detail as follows:

[0034] 301. Obtain current data of heat dissipation parameters of a heat dissipation component of a server to be processed.

[0035] The execution subject of this embodiment can be a terminal device or a server. Figure 1 Air-cooled server shown.

[0036] In this embodiment, the server to be processed may be an air-cooled server. The heat dissipation component may be a fan group. The heat dissipation parameter may be a fan speed. The current data may be a current value of the fan speed of the fan group of the air-cooled server.

[0037] 302. Input the current data into the heat dissipation model corresponding to the server to be processed to obtain the corresponding temperature adjustment amount; the heat dissipation model is determined in advance based on the heat balance equation and state space equation of the server to be processed; the heat balance equation is determined based on the heat dissipation system architecture of the processing server.

[0038] Specifically, the heat-generating components (such as GPU, memory, CPU) and heat-dissipating components (such as fan group) in the cooling system architecture of the air-cooled server can be first determined, and then the heat balance equation of the air-cooled server can be constructed based on the cooling system architecture. The heat balance equation can be linearized to obtain a linearized state-space equation, and then the cooling model of the air-cooled server can be constructed based on the state-space equation. After the cooling model is constructed, taking the server to be processed as an air-cooled server as an example, the current data of cooling parameters such as fan speed obtained in real time can be input into the cooling model, and then the cooling model outputs the temperature adjustment amount corresponding to the current fan speed. The temperature adjustment amount can be the difference between the actual temperature of the server to be processed and the preset target temperature (steady-state temperature).

[0039] 303. Perform heat dissipation control on the server to be processed according to the temperature adjustment amount.

[0040] Specifically, cooling control is performed on the server being processed based on the obtained temperature adjustment value. This process is a key step in ensuring that the server operates within the optimal temperature range. Specifically, the temperature adjustment value reflects the difference between the current server temperature and the preset target temperature. By analyzing this difference, the specific cooling measures required can be determined. First, the system evaluates the magnitude and direction of the temperature adjustment value. If the temperature adjustment value is positive, it means the current temperature is above the target and increased cooling is required. In this case, the system can increase the fan speed to enhance cooling and quickly reduce the server temperature. Conversely, if the temperature adjustment value is negative, it means the current temperature is below the target and excessive cooling may not be necessary. The system can appropriately reduce the fan speed to save energy and reduce fan wear. Furthermore, the cooling control process can incorporate other parameters, such as ambient temperature and server load, for more refined adjustments. This dynamic adjustment mechanism not only improves cooling efficiency but also effectively extends the service life of servers and fans, ensuring system stability and reliability. Through real-time monitoring and adjustment, the system can quickly respond to temperature changes and keep servers within a safe operating temperature range.

[0041] As can be seen from the above description, the heat dissipation control method provided in the embodiment of the present application can monitor the thermal status of the server in real time by obtaining the current data of the heat dissipation components. Then, these data are input into a pre-established heat dissipation model, which is based on the server's thermal balance equation and state space equation and can accurately predict the amount of heat dissipation change. This enables the system to actively adjust the heat dissipation strategy instead of passively responding to temperature changes, reducing dependence on experimental data and avoiding complex PID parameter adjustments. By performing dynamic control based on the amount of heat dissipation change, the system can quickly respond to temperature changes, optimize fan speed, reduce energy consumption, and at the same time improve the operational stability and reliability of the server.

[0042] Figure 4 Schematic diagram of the heat dissipation control method provided in this embodiment of the application Figure 2 ,like Figure 4 As shown, based on the above embodiment, for example Figure 3 Based on the illustrated embodiment, the embodiment of the present application describes in detail the process of constructing the heat dissipation model as follows:

[0043] 401. Based on the heat dissipation system architecture of the processing server, a heat balance equation of the server to be processed is constructed; the heat dissipation system architecture includes a heat dissipation component and multiple heat generating components.

[0044] Specifically, constructing a heat balance equation for the server under investigation is the foundation of the entire heat dissipation control method. This step involves a detailed analysis of the server's cooling system architecture to accurately describe the heat generation and dissipation during server operation. First, all heat-generating and heat-dissipating components in the server must be identified. These components are then decoupled. Heat-generating components typically include the central processing unit (CPU), graphics processing unit (GPU), and memory modules, and generate significant amounts of heat during operation. Heat dissipating components, such as fan modules, are used to remove heat from the server. After identifying the heat-generating and heat-dissipating components, a heat balance equation can be established based on their characteristics and interrelationships. This equation describes the input, output, and storage of heat within the server. Specifically, this equation takes into account factors such as the heat generation rate of each heat-generating component, the heat transfer efficiency of the heat dissipating components, and the overall thermal capacity of the system.

[0045] In some embodiments, constructing a heat balance equation for a server to be processed based on the server's heat dissipation system architecture may include: obtaining the server's heat dissipation system architecture; determining, for each heat-generating component, the heat generation power of the heat-generating component and the heat dissipation power of the corresponding heat-generating component; determining, based on a heat calculation formula, a first relationship corresponding to the heat-generating component based on the heat generation power and the heat dissipation power; and determining the heat balance equation for the server to be processed based on the first relationship corresponding to each of the multiple heat-generating components. The heat dissipation control method provided in this embodiment accurately constructs the server's heat balance equation by obtaining the server's heat dissipation system architecture and determining, for each heat-generating component, the heat generation power and the corresponding heat dissipation power. Using the heat calculation formula, the system establishes a first relationship for each heat-generating component. These relationships comprehensively reflect the heat generation and dissipation within the server. Ultimately, by integrating the first relationships for multiple heat-generating components, a complete heat balance equation is formed. This method accurately describes the server's thermal dynamic behavior under different operating conditions, making the heat dissipation model more accurate and reliable. Through this precise heat balance analysis, the system can optimize the heat dissipation strategy in real time, improve heat dissipation efficiency, reduce energy consumption, and ensure stable operation of the server within a safe temperature range. This not only extends the life of the equipment but also improves overall system reliability and performance.

[0046] Specifically, first, a comprehensive understanding of the cooling system architecture of the server to be processed is required, including identifying all heat-generating components (such as the CPU, GPU, and memory modules) and heat-dissipating components (such as fans and heat sinks), and determining their physical layout and interactions. Next, for each heat-generating component, the heat-generating power is calculated, either through measurement or estimation based on device specifications. Simultaneously, the heat dissipation power of the heat dissipating components associated with each heat-generating component is evaluated, which involves analyzing the efficiency of the heat dissipating components and their performance under different conditions. Using heat calculation formulas and / or heat power formulas, a first equation is established for each heat-generating component, describing the balance between heat-generating power and heat dissipation power, reflecting the dynamic balance between heat generation and heat dissipation under steady-state conditions. The first equations for all heat-generating components are integrated to form a thermal balance equation for the entire server. This equation comprehensively considers the heat exchange processes of all components and provides a global perspective on the server's dynamic thermal behavior. The thermal balance equation can not only be used to monitor and predict the server's thermal status in real time, but also serve as the basis for optimizing cooling strategies, enabling the system to dynamically adjust cooling measures under varying operating conditions to ensure the server operates within the optimal temperature range. This precise thermal management not only improves heat dissipation efficiency and reduces energy consumption, but also significantly enhances server stability and reliability.

[0047] In some embodiments, determining the heat generation power of a heat-generating component may include determining the heat generation power of the heat-generating component based on the operating power and heat generation coefficient of the heat-generating component. The heat dissipation control method provided in this embodiment can achieve more precise thermal management by determining the heat generation power of a heat-generating component based on its operating power and heat generation coefficient. The operating power represents the actual power consumption of the heat-generating component under a specific load, while the heat generation coefficient reflects the efficiency of converting electrical energy into thermal energy. By combining these two parameters, the system can accurately calculate the actual heat generation power of each heat-generating component, thereby providing reliable data support for the heat balance equation. This precise calculation method reduces reliance on empirical estimates and improves the accuracy and efficiency of heat dissipation design. The system can more effectively optimize heat dissipation strategies to ensure that the optimal temperature is maintained under different operating conditions. This not only reduces energy consumption, but also extends the service life of the equipment and improves the stability and reliability of the server. Through precise thermal management, the server can operate efficiently while reducing the risk of overheating and improving overall performance.

[0048] Specifically, when heat-generating components such as the CPU, GPU, and memory are powered on, they generate heat. That is, electrical energy is converted into thermal energy. This is also the source of heat for the server. Therefore, the calculation formula for the heat generation power of the server is as follows:

[0049] P 产热 =B×P (1)

[0050] Among them, P 产热 is the heating power generated by the heat generating component, B is the heat generating coefficient of the heat generating component, and P is the operating power of the heat generating component.

[0051] In some embodiments, determining the heat dissipation power of a heat dissipating component relative to a heat-generating component may include determining the heat dissipation power of the heat dissipating component relative to the heat-generating component based on the heat transfer coefficient and heat dissipation parameters of the heat dissipating component, the temperature of the heat-generating component, and the ambient temperature. The heat dissipation control method provided in this embodiment achieves more efficient thermal management by determining the heat dissipation power based on the heat transfer coefficient and heat dissipation parameters of the heat dissipating component, the temperature of the heat-generating component, and the ambient temperature. The heat transfer coefficient and heat dissipation parameters together describe the performance characteristics of the heat dissipating component, while the difference between the temperature of the heat-generating component and the ambient temperature determines the driving force of heat transfer. By integrating these factors, the system can accurately calculate the actual heat dissipation power of the heat dissipating component. This method has the beneficial effect of providing precise quantification of the heat dissipation process, enabling the cooling system to dynamically adjust to varying heat loads and environmental changes. The system can more effectively optimize fan speeds and other cooling measures to ensure that the server operates within the optimal temperature range. This not only improves cooling efficiency and reduces energy consumption, but also enhances server stability and reliability, extending the service life of the equipment.

[0052] Specifically, taking the fan as the heat dissipation component of the system as an example, the heat dissipation principle is the law of convection heat transfer. Based on this law, the heat dissipation power of the fan of the server to be processed can be calculated as follows:

[0053] P 散热 =H×A×ΔT (2)

[0054] Among them, P 散热 is the heat dissipation power of the fan, H is the convective heat transfer coefficient, which expresses the heat flow per unit area and unit temperature difference, A is the surface area of ​​the solid (heat-generating component) in contact with the fluid, and ΔT is the temperature difference between the solid (heat-generating component) surface and the fluid.

[0055] Among them, the convective heat transfer coefficient H has a positive correlation with the fan speed U, so it can be approximated as:

[0056] H=D×U (3)

[0057] Where D is the fan speed heat transfer proportional coefficient, which is determined by the fan's physical properties and air duct design, and U is the fan speed.

[0058] In this embodiment, the heat transfer coefficient K may be the product of the fan speed heat transfer proportional coefficient D and the surface area A of the solid in contact with the fluid, and the heat dissipation parameter may be the fan speed U.

[0059] In some embodiments, determining a first relationship corresponding to a heat-generating component based on heat generation power and heat dissipation power based on a heat calculation formula may include determining, based on the heat calculation formula, a first relationship between the heat dissipation parameter of the heat dissipation component, the heat generation power of the heat dissipation component, the temperature difference of the heat dissipation component within a first time period, the heat capacity of the heat dissipation component, the temperature of the heat dissipation component at the current moment, the ambient temperature at the current moment, and the first time period. The heat dissipation control method provided in this embodiment enables accurate modeling of thermal dynamic behavior by determining the first relationship between factors such as the heat dissipation parameter of the heat dissipation component, the heat generation power of the heat dissipation component, the temperature difference, the heat capacity, and the ambient temperature based on the heat calculation formula. By considering these key parameters, the system can accurately describe the heat changes of the heat-generating component within a specific time period. This method provides a comprehensive thermal management framework that enables the cooling system to dynamically adapt to different operating conditions and environmental changes. The system can more effectively optimize cooling strategies to ensure that the server operates within the optimal temperature range. This not only improves cooling efficiency and reduces energy consumption, but also enhances server stability and reliability, extending the service life of the equipment. Through precise thermal management, the server can operate efficiently while reducing the risk of overheating and improving overall performance.

[0060] Based on the heat calculation formula Q=M×C×ΔT, the heat balance equation of the server heat dissipation control system in time t can be obtained as follows:

[0061] M×C×(T K+1 –T K )=(P 产热 -P 散热 )×t (4)

[0062] Where M is the mass of the heat generating component, C is the specific heat capacity of the heat generating component, T K+1 is the temperature of the heat-generating component at the next moment, T K is the temperature of the heat generating component at the current moment, and t is the first duration.

[0063] Taking the heat-generating component CPU as an example, the above formula can be written as:

[0064] C M1 ×ΔT1=(P 产热1 -P 散热1 ) × t (5)

[0065] P 散热1 =K1×U×(T1–T0) (6)

[0066] Among them, C M1 is the product of the CPU's mass M1 and specific heat capacity C1, which is determined by the physical properties of the CPU and is a fixed parameter; ΔT1 is the temperature difference of the CPU within time t (the first duration); P 产热1 is the heat generation power of the CPU; P 散热1 is the heat dissipation power of the fan for the CPU; K1 is the heat transfer coefficient, which is the product of the heat transfer proportional coefficient D1 of the fan speed to the CPU and the surface area A1 of the CPU in contact with the fluid, and is a fixed parameter; U is the fan speed; T1 is the CPU temperature at time t, and T0 is the ambient temperature at time t.

[0067] By analogy, we can derive the thermal balance equations for other heat-generating components in the server system, such as memory and GPU.

[0068] Taking the CPU and memory in the server as an example, based on expressions (5) and (6), the thermal balance formula of the CPU and memory can be obtained, that is, the first relationship corresponding to the heat-generating components CPU and memory is as follows:

[0069] C M1 ×ΔT1=(P 产热1 -K1×U×(T1–T0))×t(CPU) (7)

[0070] C M2 ×ΔT2=(P 产热2-K2×U×(T2–T0))×t(memory) (8)

[0071] Among them, C M1 The product of the CPU's mass M1 and specific heat capacity C1, C M2 is the product of the mass M2 and specific heat capacity C2 of the memory, which is determined by the physical properties of the CPU and memory and is a fixed parameter; ΔT1 is the temperature difference of the CPU within time t, and ΔT2 is the temperature difference of the memory within time t; P 产热1 is the heat generation power of the CPU, P 产热2 is the heat generation power of the memory; P 散热1 P is the cooling power of the fan for the CPU, 散热2 is the heat dissipation power of the fan for the memory; K1 is the heat transfer coefficient corresponding to the CPU, which is the product of the fan's speed heat transfer ratio coefficient D1 for the CPU and the surface area A1 of the CPU in contact with the fluid; K2 is the heat transfer coefficient corresponding to the memory, which is the product of the fan's speed heat transfer ratio coefficient D2 for the memory and the surface area A2 of the memory in contact with the fluid, and is a fixed parameter; U is the fan speed; T1 is the CPU temperature at time t, T2 is the memory temperature at time t, and T0 is the ambient temperature at time t; t is the first time duration.

[0072] 402. Linearize the heat balance equation to obtain a state space equation of the server to be processed.

[0073] Specifically, the heat balance equation is typically nonlinear, describing the heat exchange and dynamic balance between various components within a server. However, nonlinear equations are complex to analyze and calculate, and direct application to real-time control systems can result in excessive computational overhead. Linearization aims to simplify complex nonlinear equations into linear equations, making them easier to analyze and control. In one implementation, numerical methods can be used to solve the heat balance equation's response to various inputs and states. Fitting techniques (such as the least squares method) then convert the numerical solution into a linear state-space model. In another implementation, linearization is typically performed near a certain operating point of the system, allowing the behavior of the nonlinear system to be approximated by linear equations. This method transforms the heat balance equation into a state-space equation, which represents the system's dynamic characteristics in matrix form, including state variables, inputs, outputs, and system matrices. State-space equations enable the system to utilize various methods from modern control theory to design and optimize thermal control strategies, thereby improving the control system's response speed and accuracy. This approach enables the system to dynamically adapt to varying operating conditions and environmental changes, ensuring efficient server operation within the optimal temperature range. Linearization and state-space modeling lay the foundation for intelligent and automated thermal control.

[0074] In some embodiments, linearizing the heat balance equation to obtain the state-space equation of the server to be processed may include linearizing the heat balance equation based on a steady-state operating point to obtain the state-space equation of the server to be processed. The heat dissipation control method provided in this embodiment significantly simplifies complex thermal management problems by linearizing the heat balance equation based on a steady-state operating point to obtain the state-space equation of the server to be processed. By selecting an appropriate steady-state operating point, the system can linearly approximate the nonlinear thermal dynamic behavior near that point, thereby converting the complex heat balance equation into a linear state-space model that is easy to analyze and control. This method has the beneficial effect of reducing computational complexity and enabling real-time control and optimization. The linearized state-space equation provides a clear framework that facilitates the application of modern control theory for system design and optimization. The system can more effectively adjust the heat dissipation strategy to ensure that the server maintains the optimal temperature under different load conditions. This not only improves heat dissipation efficiency and reduces energy consumption, but also enhances the stability and reliability of the server.

[0075] In some embodiments, linearizing the heat balance equation based on a steady-state operating point to obtain a state-space equation for the server to be processed may include: determining, for each heat-generating component, a second relationship between the heat-generating component's heat-generating power, the steady-state parameter value, and the steady-state temperature, based on the heat balance equation corresponding to the heat-generating component, when the heat-generating component is at a steady-state temperature and the heat dissipation parameter of the heat dissipating component is at a steady-state parameter value; determining, based on the second relationship corresponding to each of the multiple heat-generating components, the state-space equation for the server to be processed; the state-space equation including a third relationship between the first-order derivative of the temperature regulation variable, a system matrix, a state vector, an input matrix, and an input vector; the state vector including the temperature difference within a first time period corresponding to each of the multiple heat-generating components, and the input vector including the heat dissipation parameter of the heat dissipating component. The heat dissipation control method provided in this embodiment accurately constructs the state-space equation for the server by determining, for each heat-generating component, the relationship between the heat-generating power, the steady-state parameter, and the temperature, based on the heat balance equation under steady-state conditions. Through this linearization process, the system can simplify complex thermal dynamic behavior into a linear model, where the state-space equation includes the first-order derivative of the temperature regulation variable, the system matrix, the state vector, and the input vector. The state vector reflects the temperature changes of each heat-generating component, while the input vector contains the parameters of the heat-dissipating components. The beneficial effect of this method is that it provides a precise and easily analyzable framework that enables real-time temperature control and optimization. The system can dynamically adjust cooling strategies to ensure that the server maintains the optimal temperature under various load conditions. This not only improves cooling efficiency and reduces energy consumption, but also enhances server stability and reliability.

[0076] Specifically, if the time t is short enough, then (ΔT1 / t)=T1~(T1 ~is the first derivative of variable T1), so based on expressions (7) and (8), we can get:

[0077] T1 ~ =(P 产热1 -K1×U×(T1–T0)) / C M1 (CPU) (9)

[0078] T2 ~ =(P 产热2 -K2×U×(T2–T0)) / C M2 (Memory) (10)

[0079] Where T1 is the CPU temperature, T2 is the memory temperature, and they are the system state variables; U is the fan speed and is the system input variable. The meaning of each parameter can be found in the introduction to formulas (7) and (8). Since formulas (9) and (10) contain the product of U and T1 and T2, the system model is nonlinear and needs to be linearized to facilitate subsequent model-based analysis and control.

[0080] Furthermore, the model can be linearized using the steady-state operating point method. If the heat dissipation control target of the server to be processed is to stabilize the temperature of the CPU and memory at the steady-state temperature T 1W and T 2W , then the final control effect is that the system is in (T 1W ,T 2W ,U W ) This working point achieves thermal equilibrium steady state, U W is the steady-state speed of the fan. Because at the steady-state operating point, the temperature of the CPU and memory no longer changes, that is, T1 ~ =T2 ~ =0, then we have the following formula, which is the second relationship:

[0081] P 产热1 =K1×U W × (T 1W –T0)(CPU) (11)

[0082] P 产热2 =K2×U W × (T 2W –T0)(Memory) (12)

[0083] Among them, P 产热1 is the heat generation power of the CPU, P 产热2is the heat generation power of the memory; K1 is the heat transfer coefficient corresponding to the CPU, which is the product of the heat transfer ratio coefficient D1 of the fan to the CPU speed and the surface area A1 of the CPU in contact with the fluid; K2 is the heat transfer coefficient corresponding to the memory, which is the product of the heat transfer ratio coefficient D2 of the fan to the memory speed and the surface area A2 of the memory in contact with the fluid, and is a fixed parameter; U W is the steady-state speed of the fan, i.e. the steady-state heat dissipation parameter; T 1W is the steady-state temperature of the CPU, T 2W is the steady-state temperature of the CPU, and T0 is the ambient temperature.

[0084] Furthermore, based on the steady-state operating point, let ΔT1=T1-T 1W , ΔT2=T2–T 2W , ΔU=U–U W , substituting into equation (9) we can get:

[0085] (ΔT1 + T 1W ) ~ =(P 产热1 -K1×(ΔU+U W )×(ΔT1+T 1W –T0)) / C M1 (CPU) (13)

[0086] Among them, P 产热1 is the heat generation power of the CPU, ΔT1 is the actual CPU temperature T1 and the CPU control target temperature (steady-state temperature) T 1W The difference between the actual speed U and the steady-state speed U of the fan is W The difference between the two values; K1 is the heat transfer coefficient corresponding to the CPU, which is the product of the heat transfer ratio coefficient D1 of the fan to the CPU speed and the surface area A1 of the CPU in contact with the fluid; C M1 It is the product of the CPU's mass M1 and specific heat capacity C1, and T0 is the ambient temperature.

[0087] Since P in expression (11) 产热1 =K1×U W × (T 1W –T0), and ignoring the high-order terms ΔU×ΔT1, we can obtain:

[0088] ΔT1 ~ =(-K1×U W / C M1 )×ΔT1+(-K1(T 1W –T0) / C M1 ) × ΔU (CPU) (14)

[0089] Among them, T 1W is a constant, (ΔT1 + T1W ) ~ Can be simplified to ΔT1 ~ ; ΔT1 is the actual CPU temperature T1 and the CPU control target temperature (steady-state temperature) T 1W The difference between the actual speed U and the steady-state speed U of the fan is W The difference between the two values; K1 is the heat transfer coefficient corresponding to the CPU, which is the product of the heat transfer ratio coefficient D1 of the fan to the CPU speed and the surface area A1 of the CPU in contact with the fluid; C M1 It is the product of the CPU's mass M1 and specific heat capacity C1, and T0 is the ambient temperature.

[0090] Similarly, for memory:

[0091] ΔT2~ =(-K 2× U W / C M2 )×ΔT2+(-K2(T 2W –T0) / C M2 )×ΔU(memory) (15)

[0092] Among them, T 2W is a constant, (ΔT2+T 2W ) ~ Can be simplified to ΔT2 ~ ; ΔT2 is the actual memory temperature T2 and the memory control target temperature (steady-state temperature) T 2W The difference between the actual speed U and the steady-state speed U of the fan is W K2 is the heat transfer coefficient corresponding to the memory, which is the product of the heat transfer ratio coefficient D2 of the fan speed to the memory and the surface area A2 of the memory in contact with the fluid; C M2 It is the product of the mass M2 and specific heat capacity C2 of the memory, and T0 is the ambient temperature.

[0093] Furthermore, the state space equation of the heat dissipation control system of the server to be processed can be constructed. The state space equation of the system is a mathematical model that describes the behavior of the dynamic system and consists of two parts: the state equation and the output equation.

[0094] Based on the above formulas (14) and (15) and matrix operations, the state equation of the system can be obtained:

[0095] ΔT ~ =A×ΔT+B×U (16)

[0096] Among them, ΔT=[ΔT1,ΔT2]T is the state vector;

[0097] U=[U]: input vector;

[0098] A=[(-K1×UW / C M1 ,0),(0,-K2×U W / C M2 )]T: is the system matrix;

[0099] B=[(-K1(T 1W –T0) / C M1 ),(-K2(T 2W –T0) / C M2 )]T: is the input matrix.

[0100] The meaning of each parameter can be found in the introduction of formulas (14) and (15).

[0101] Based on the above formulas (14) and (15) and matrix operations, the output equation of this system can be obtained:

[0102] Y = C × ΔT (17)

[0103] Among them, Y=[ΔT1,ΔT2]T: is the output vector;

[0104] ΔT=[ΔT1,ΔT2]T: is the state vector;

[0105] C=[(1,0),(0,1)]T: is the output matrix; the meaning of each parameter can be found in formulas (14) and (15).

[0106] 403. Determine a heat dissipation model corresponding to the server to be processed based on the state-space equation.

[0107] In some embodiments, determining the cooling model corresponding to the server to be processed based on the state-space equation may include: determining the heat transfer coefficients and heat capacities corresponding to multiple heat-generating components, as well as the steady-state cooling parameters of the heat dissipating components, through experimental calibration; determining the system matrix and input matrix in the state-space equation based on the heat transfer coefficients and heat capacities corresponding to the multiple heat-generating components, as well as the steady-state cooling parameters of the heat dissipating components; and determining the cooling model of the server to be processed based on the system matrix and input matrix based on the state-space equation. The cooling control method provided in this embodiment determines the heat transfer coefficients and heat capacities of each heat-generating component, as well as the steady-state cooling parameters of the heat dissipating components, through experimental calibration, enabling the system to accurately construct a cooling model for the server. Using this experimental data, the system can accurately define the system matrix and input matrix in the state-space equation, thereby precisely describing the thermal dynamic behavior. This method has the beneficial effect of providing a reliable data foundation for the cooling model, enabling the model to truly reflect the thermal response of the server under different operating conditions. Through this precise modeling, the system can optimize cooling strategies and dynamically adjust cooling measures to ensure that the server maintains an optimal temperature under various load conditions. This not only improves cooling efficiency and reduces energy consumption, but also enhances server stability and reliability, extending the service life of the equipment.

[0108] Specifically, after obtaining the state equation (16) for the server's heat dissipation control system (using the CPU and memory as an example), we only need to obtain matrices A and B to establish the relationship between ΔT and U, and then construct the heat dissipation model of the server's heat dissipation control system. For matrices A and B (using the CPU as an example), we only need to obtain K1, UW, and CM1.

[0109] In some embodiments, determining the heat transfer coefficients corresponding to multiple heat-generating components through experimental calibration may include: placing each heat-generating component in a constant temperature environment, operating it at a constant power, and setting a heat dissipation parameter to a constant value. After the server to be processed reaches thermal equilibrium, obtaining the heat generation power of the heat-generating component, the current temperature, the ambient temperature, and the actual value of the heat dissipation parameter of the heat dissipation component; and determining the heat transfer coefficient corresponding to the heat-generating component based on the heat generation power, the current temperature, the ambient temperature, and the actual value. The heat dissipation control method provided in this embodiment can accurately determine the heat transfer coefficient corresponding to each heat-generating component by placing the heat-generating component in a constant temperature environment, operating it at a constant power, and obtaining key parameters after reaching thermal equilibrium. Through this experimental calibration method, the system can calculate the heat transfer coefficient under known environmental conditions using the heat generation power, current temperature, ambient temperature, and actual values ​​of the heat dissipation parameter. The beneficial effect of this method is that it provides accurate heat transfer characteristic parameters for the heat dissipation model, allowing the model to more accurately reflect actual thermal dynamic behavior. The system can optimize the heat dissipation strategy based on the accurate heat transfer coefficient to ensure that the server maintains the optimal temperature under various load conditions. This not only improves heat dissipation efficiency and reduces energy consumption, but also enhances the stability and reliability of the server.

[0110] Specifically, for the parameter K1, if the heat dissipation control system is in a steady state, that is, a thermal equilibrium state, the temperature of the heat generating component will no longer change, and then based on expression (7), C M1 ×Δ T1 =(P 产热1 -K1×U×(T1–T0))×t=0, that is:

[0111] K1=P 产热1 / (U×(T1–T0)) (17).

[0112] The meaning of each parameter can be found in the introduction of formula (7).

[0113] In practice, to achieve thermal equilibrium in a server, place the server in a constant temperature environment, operate it at a constant power level, and set the fan speed to a constant. After a period of normal operation, observe that the temperatures of various components no longer fluctuate, indicating that the server has reached thermal equilibrium. In this state, solve for K1.

[0114] Among them, P 产热1= B1 × P1, where B1 is the CPU's heat generation coefficient, which can be obtained from the component's data sheet; P1 is the CPU's operating power, which can be obtained from the server's BMC; U is the actual fan speed, which can be obtained from the server's BMC; T1 and T0 are the actual CPU temperature and ambient temperature, respectively, which can be obtained from the server's BMC. Then, using formula (17), we can calculate K1 for the CPU. Similarly, we can calculate K2 for the memory and the parameter K for other components.

[0115] In some embodiments, determining the steady-state heat dissipation parameters of a heat dissipation component through experimental calibration may include: obtaining the heat generation power and ambient temperature of each heat-generating component; determining the steady-state heat dissipation parameters of the heat dissipation component corresponding to the heat-generating component based on the heat generation power and ambient temperature based on a second relationship; and determining the steady-state heat dissipation parameters of the heat dissipation component based on the steady-state heat dissipation parameters corresponding to each heat-generating component. The heat dissipation method provided in this embodiment, by obtaining the heat generation power and ambient temperature of each heat-generating component and determining the steady-state heat dissipation parameters corresponding to each heat-generating component based on the second relationship, can accurately evaluate the performance of the heat dissipation component under different heat source conditions. By integrating the steady-state heat dissipation parameters corresponding to each heat-generating component to determine the steady-state heat dissipation parameters of the entire heat dissipation component, the system can more comprehensively reflect the heat dissipation capacity under actual operating conditions. This method has the beneficial effect of providing a more accurate and adaptable heat dissipation model, enabling the system to optimize heat dissipation strategies under various load and environmental conditions. This not only improves heat dissipation efficiency and reduces energy consumption, but also enhances the stability and reliability of the server, ensuring efficient operation in a changing operating environment.

[0116] For example, the average or median of the steady-state heat dissipation parameters corresponding to each heat-generating component can be determined and used as the steady-state heat dissipation parameter for the heat dissipation component. The heat dissipation method provided in this embodiment calculates the average or median of the steady-state heat dissipation parameters corresponding to each heat-generating component and uses this as the steady-state heat dissipation parameter for the entire heat dissipation component, thereby balancing heat dissipation performance under different heat source conditions. This method advantageously provides a robust and adaptable heat dissipation model, improving the heat dissipation efficiency and stability of the system.

[0117] In some embodiments, determining the steady-state heat dissipation parameter of the heat dissipation component based on the steady-state heat dissipation parameter corresponding to each heat-generating component may include: determining the maximum value of the steady-state heat dissipation parameters of the heat dissipation component corresponding to each heat-generating component as the steady-state heat dissipation parameter of the heat dissipation component. The heat dissipation control method provided in this embodiment can accurately identify the performance of the heat dissipation component under different conditions by obtaining the heat generation power and ambient temperature of each heat-generating component and determining the steady-state heat dissipation parameter corresponding to the heat dissipation component based on the second relationship (11). By taking the maximum value of the steady-state heat dissipation parameters of the heat dissipation component corresponding to each heat-generating component as the final steady-state heat dissipation parameter, the system ensures the heat dissipation capacity under the most demanding conditions. The beneficial effect of this method is that it provides a conservative and reliable parameter selection for the heat dissipation model, ensuring that the heat dissipation control system of the server can effectively respond to heat changes under various load conditions. This not only improves the heat dissipation efficiency and reduces energy consumption, but also enhances the stability and reliability of the server, ensuring that it can still operate safely under high load conditions.

[0118] Specifically, for solving the parameter U W , can be obtained from formula (11):

[0119] U W =P 产热1 / (K1×(T 1W –T0)) (18).

[0120] The meaning of each parameter can be found in the introduction of formula (11). 产热1 Both T and T0 can be obtained, T 1W The CPU's target temperature is set by yourself, and the CPU's U W , similarly, we can get the U of memory W and other components of U W . Obviously, the U required by each component W are not equal, but there is only one parameter U in the model W In order to ensure that the temperature of all controlled components is within the safe target range, the maximum U W As a parameter in the model for redundant design.

[0121] In some embodiments, determining the corresponding heat capacities of multiple heat-generating components through experimental calibration may include: for each heat-generating component, when the heat dissipation parameter of the heat dissipation component is 0 and the heat dissipation power is 0, obtaining the temperature difference, heat generation power, and second time period of the heat-generating component; and determining the corresponding heat capacity of the heat-generating component based on the temperature difference, heat generation power, and second time period of the heat-generating component. The heat dissipation method provided in this embodiment accurately calculates the heat capacity of each heat-generating component by obtaining the temperature difference and heat generation power of each heat-generating component over a specific time period when the heat dissipation parameter and heat dissipation power of the heat dissipation component are both 0. Through this experimental calibration method, the system can accurately assess the heat storage capacity of heat-generating components in an isolated thermal environment. This method has the beneficial effect of providing key thermal characteristic parameters to the thermal management model, enabling the model to more accurately simulate and predict thermal dynamic behavior. Based on accurate heat capacity data, the system can optimize the heat dissipation strategy to ensure that the server maintains the optimal temperature under various load conditions. This not only improves heat dissipation efficiency and reduces energy consumption, but also enhances the stability and reliability of the server and extends the service life of the equipment.

[0122] Specifically, for solving the parameter C M1 If the fan speed is set to 0, the heat dissipation power of the system is 0, and then based on formula (5), we can get C M1 ×ΔT1=P 产热1 ×t, that is:

[0123] C M1 =P 产热1 ×t / ΔT1 (19).

[0124] Among them, t is the second duration; P 产热1 is the heat generation power of the CPU; C M1 It is the product of the CPU's mass M1 and specific heat capacity C1; ΔT1 is the temperature difference of the CPU within time t.

[0125] Therefore, in order to solve C M1 , you need to set the server to run at constant power and set the fan speed to 0, run for a period of time, for example 120s, record the CPU temperature at 0s and 120s, and you can get the CPU temperature rise ΔT1 during this period, and P 产热1 and t can be obtained, and the CPU parameter C can be obtained. M1 , similarly, we can get C for memory M2 and other components of C M .

[0126] In summary, the parameters K and U are obtained respectively W and C M After that, the matrices A and B can be constructed, thus completing the establishment of the heat dissipation model.

[0127] 404. Obtain current data of heat dissipation parameters of a heat dissipation component of the server to be processed.

[0128] 405. Input the current data into the heat dissipation model corresponding to the server to be processed to obtain the corresponding temperature adjustment amount; the heat dissipation model is determined in advance based on the heat balance equation and state space equation of the server to be processed; the heat balance equation is determined based on the heat dissipation system architecture of the processing server.

[0129] 406. Perform heat dissipation control on the server to be processed according to the temperature adjustment amount.

[0130] In this embodiment, steps 404 to 406 are similar to steps 201 to 203 in the above embodiment and are not described again here.

[0131] From the above description, it can be seen that the heat dissipation control method provided in this embodiment is

[0132] 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.

[0133] Figure 5 This is a schematic diagram of the structure of the heat dissipation control device provided in the embodiment of the present application. Figure 5 As shown, an embodiment of the present application further provides a heat dissipation control device 50 , comprising: an acquisition module 501 , an input module 502 and a control module 503 .

[0134] An acquisition module 501 is used to acquire current data of heat dissipation parameters of a heat dissipation component of a server to be processed;

[0135] Input module 502 is used to input current data into the heat dissipation model corresponding to the server to be processed to obtain the corresponding temperature adjustment amount; the heat dissipation model is pre-determined based on the heat balance equation and state space equation of the server to be processed; the heat balance equation is determined based on the heat dissipation system architecture of the processing server;

[0136] The control module 503 is used to control the heat dissipation of the server to be processed according to the temperature adjustment amount.

[0137] The heat dissipation control device provided in the embodiment of the present application can monitor the thermal status of the server in real time by acquiring the current data of the heat dissipation components. Then, these data are input into a pre-established heat dissipation model, which is based on the server's thermal balance equation and state space equation and can accurately predict the temperature adjustment amount. This enables the system to actively adjust the heat dissipation strategy instead of passively responding to temperature changes, reducing dependence on experimental data and avoiding complex PID parameter adjustments. By performing dynamic control based on the temperature adjustment amount, the system can quickly respond to temperature changes, optimize fan speed, reduce energy consumption, and at the same time improve the operational stability and reliability of the server.

[0138] In some embodiments, the heat dissipation control device 50 also includes a model building module 500, which is used to: construct a thermal balance equation for the server to be processed based on the heat dissipation system architecture of the processing server; the heat dissipation system architecture includes heat dissipation components and multiple heat-generating components; linearize the heat balance equation to obtain the state space equation of the server to be processed; and determine the heat dissipation model corresponding to the server to be processed based on the state space equation.

[0139] In some embodiments, the model building module 500 is specifically used to: obtain the cooling system architecture of the server to be processed; determine the heat generation power of the heat generation component and the heat dissipation power of the heat dissipation component corresponding to the heat generation component for each heat generation component; based on the heat calculation formula, determine the first relationship corresponding to the heat generation component according to the heat generation power and the heat dissipation power; determine the thermal balance equation of the server to be processed according to the first relationship corresponding to multiple heat generation components.

[0140] In some embodiments, the model building module 500 is specifically used to determine the heat generation power of the heat generation component according to the operating power and heat generation coefficient of the heat generation component.

[0141] In some embodiments, the model building module 500 is specifically used to determine the heat dissipation power of the heat dissipation component corresponding to the heat generating component based on the heat transfer coefficient and heat dissipation parameters of the heat dissipation component, the temperature of the heat generating component, and the ambient temperature.

[0142] In some embodiments, the model building module 500 is specifically used to: determine the heat dissipation parameters of the heat dissipation component, the heat generation power of the heat generation component, the temperature difference of the heat generation component within the first time period, the heat capacity of the heat generation component, the temperature of the heat generation component at the current moment, the ambient temperature at the current moment and the first relationship between the first time period based on the heat calculation formula.

[0143] In some embodiments, the model building module 500 is specifically used to: linearize the heat balance equation based on the steady-state operating point to obtain the state space equation of the server to be processed.

[0144] In some embodiments, the model building module 500 is specifically used to: for each heat-generating component, determine, according to the heat balance equation corresponding to the heat-generating component, the second relationship between the heat-generating power of the heat-generating component and the steady-state parameter value and the steady-state temperature when the heat-generating component is at a steady-state temperature and the heat dissipation parameter of the heat dissipation component is a steady-state parameter value; determine the state space equation of the server to be processed according to the second relationship corresponding to multiple heat-generating components; the state space equation includes the first-order derivative of the temperature adjustment amount, the system matrix, the state vector, the input matrix and the third relationship between the input vector; the state vector includes the temperature difference within the first time period corresponding to multiple heat-generating components, and the input vector includes the heat dissipation parameters of the heat dissipation component.

[0145] In some embodiments, the model building module 500 is specifically used to: determine the heat transfer coefficients and heat capacities corresponding to multiple heat-generating components, as well as the steady-state heat dissipation parameters of the heat dissipation components through experimental calibration; determine the system matrix and input matrix in the state-space equation based on the heat transfer coefficients and heat capacities corresponding to multiple heat-generating components, as well as the steady-state heat dissipation parameters of the heat dissipation components; based on the state-space equation, determine the heat dissipation model of the server to be processed according to the system matrix and the input matrix.

[0146] In some embodiments, the model building module 500 is specifically used to: for each heat-generating component, place the heat-generating component in a constant temperature environment, operate at constant power, and set the heat dissipation parameters to constant values; after the server to be processed reaches a thermal equilibrium state, obtain the heat-generating power of the heat-generating component and the current temperature, ambient temperature, and the actual value of the heat dissipation parameter of the heat-generating component; and determine the heat transfer coefficient corresponding to the heat-generating component based on the heat-generating power, current temperature, ambient temperature, and actual value.

[0147] In some embodiments, the model building module 500 is specifically used to: obtain the heat generation power and ambient temperature of each heat generating component, and based on the second relationship, determine the steady-state heat dissipation parameters of the heat dissipation component corresponding to the heat generating component according to the heat generation power and the ambient temperature; and determine the maximum value of the steady-state heat dissipation parameters of the heat dissipation component corresponding to each heat generating component as the steady-state heat dissipation parameter of the heat dissipation component.

[0148] In some embodiments, the model building module 500 is specifically used to: for each heat-generating component, when the heat dissipation parameter of the heat dissipation component is 0 and the heat dissipation power is 0, obtain the temperature difference, heat generation power and second time length of the heat-generating component within the second time length, and determine the corresponding heat capacity of the heat-generating component based on the temperature difference, heat generation power and second time length of the heat-generating component within the second time length.

[0149] For the description of the features in the embodiment corresponding to the heat dissipation control device, reference can be made to the relevant description of the embodiment corresponding to the heat dissipation control method, and no further details will be given here.

[0150] Figure 6 This is a schematic diagram of the structure of the electronic device provided in this application. Figure 6 As shown, the electronic device 60 provided in this embodiment includes: at least one processor 601 and a memory 602. Optionally, the electronic device 60 further includes a communication component 603. The processor 601, the memory 602 and the communication component 603 are connected via a bus.

[0151] In a specific implementation process, at least one processor 601 executes the computer-executable instructions stored in the memory 602, so that the at least one processor 601 executes the above-mentioned heat dissipation control method embodiment.

[0152] The specific implementation process of the processor 601 can be found in the above method embodiment. Its implementation principle and technical effects are similar and will not be repeated here in this embodiment.

[0153] In the above embodiments, it should be understood that the processor may be a central processing unit (CPU), other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), etc. A general-purpose processor may be a microprocessor or any conventional processor. The steps of the method disclosed in the application may be directly executed by a hardware processor or by a combination of hardware and software modules within the processor.

[0154] The memory may include random access memory (RAM) and may also include non-volatile memory (NVM), such as at least one disk storage.

[0155] A bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus. Buses can be categorized as address buses, data buses, and control buses. For ease of illustration, the buses in the drawings of this application are not limited to just one bus or just one type of bus.

[0156] 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 heat dissipation control method embodiments when running.

[0157] 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.

[0158] 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 heat dissipation control method embodiments are implemented.

[0159] 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 in any of the above-mentioned heat dissipation control method embodiments are implemented.

[0160] 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.

[0161] The above is a detailed introduction to a heat dissipation control method provided by the present application. Specific examples are used herein to illustrate the principles and implementation methods of the present application. The description of the above embodiments is only used to help understand the method and core ideas of the present application. It should be pointed out that for ordinary technicians in this technical field, without departing from the principles of the present application, several improvements and modifications can be made to the present application, and these improvements and modifications also fall within the scope of protection of the claims of the present application.

Claims

1. A heat dissipation control method, characterized in that: include: Acquire current data of heat dissipation parameters of a heat dissipation component of the server to be processed; The heat dissipation system architecture of the server to be processed includes a heat dissipation component and multiple heat generating components, and the heat dissipation component is a fan group; Inputting the current data into a heat dissipation model corresponding to the server to be processed to obtain a corresponding temperature adjustment amount; the heat dissipation model is pre-determined based on a heat balance equation and a state space equation of the server to be processed; Performing heat dissipation control on the server to be processed according to the temperature adjustment amount; The method further includes: constructing a heat balance equation for the server to be processed based on the heat dissipation system architecture of the server to be processed; Based on the steady-state operating point, linearizing the heat balance equation to obtain the state space equation of the server to be processed; Determining a heat dissipation model corresponding to the server to be processed according to the state-space equation; Wherein, for each of the heat-generating components, the heat balance equation is determined based on the heat calculation formula, which includes the heat dissipation parameters of the heat-generating component, the heat-generating power of the heat-generating component, the temperature difference of the heat-generating component within the first time period, the heat capacity of the heat-generating component, the temperature of the heat-generating component at the current moment, the ambient temperature at the current moment, and the first relationship between the heat-generating component and the ambient temperature at the current moment. The heat balance equation is: M×C×(T K+1 –T K )=(P 产热 -P 散热 )×t; where M is the mass of the heat generating component, C is the specific heat capacity of the heat generating component, and T K+1 For heat generating components in the next Temperature at the moment, T K is the temperature of the heat generating component at the current moment, t is the first duration, P 产热 is the heat generation power generated by the heat generating components, P 散热 The heat dissipated by the fan; The state space equation is: ΔT ~ =A×ΔT+B×U; Wherein, ΔT=[ΔT1, ΔT2]T is the state vector, the state vector is the first-order derivative of the temperature adjustment amount, ΔT1 is the actual CPU temperature T1 and the CPU control steady-state temperature T 1W ΔT2 is the difference between the actual memory temperature T2 and the memory control steady-state temperature T 2W The difference between U=[U]: is an input vector; the input vector includes the heat dissipation parameters of the heat dissipation component; A=[(-K1×U W / C M1 ,0),(0,-K2×U W / C M2 )]T: is the system matrix, U W is the steady-state speed of the fan; B=[(-K1(T 1W –T0) / C M1 ),(-K2(T 2W –T0) / C M2 )]T: is the input matrix; Where K1 is the heat transfer coefficient corresponding to the heat generating component CPU, which is the product of the heat transfer ratio coefficient D1 of the fan speed to the CPU and the surface area A1 of the CPU in contact with the fluid; T 1W is the steady-state temperature of the CPU, T0 is the ambient temperature, C M1 is the product of the CPU's mass M1 and specific heat capacity C1; K2 is the heat transfer coefficient corresponding to the memory, which is the product of the fan's heat transfer proportional coefficient D2 to the memory's speed and the surface area A2 of the memory in contact with the fluid; T 2W is the steady-state temperature constant of the memory; C M2 It is the product of the mass M2 and specific heat capacity C2 of the memory, and T0 is the ambient temperature.

2. The heat dissipation control method according to claim 1, wherein: The heat balance equation of the server to be processed is constructed based on the heat dissipation system architecture of the processing server, including: Obtaining a heat dissipation system architecture of the server to be processed; For each heat-generating component, determining the heat-generating power of the heat-generating component and the heat-dissipating power of the heat-dissipating component corresponding to the heat-generating component; determining a first relationship corresponding to the heat-generating component based on the heat-generating power and the heat-dissipating power based on a heat calculation formula; The heat balance equation of the server to be processed is determined according to the first relationship expressions respectively corresponding to the plurality of heat-generating components.

3. The heat dissipation control method according to claim 2, wherein: The determining the heat generating power of the heat generating component includes: The heat generating power of the heat generating component is determined according to the operating power and the heat generating coefficient of the heat generating component.

4. The heat dissipation control method according to claim 2, wherein: The determining the heat dissipation power of the heat dissipation component corresponding to the heat generating component includes: The heat dissipation power of the heat dissipation component corresponding to the heat generating component is determined according to the heat transfer coefficient and heat dissipation parameters of the heat dissipation component, the temperature of the heat generating component, and the ambient temperature.

5. The heat dissipation control method according to claim 1, wherein: The step of linearizing the heat balance equation based on the steady-state operating point to obtain the state space equation of the server to be processed includes: For each heat-generating component, determining, based on a heat balance equation corresponding to the heat-generating component, a second relationship between the heat-generating power of the heat-generating component, the steady-state parameter value, and the steady-state temperature when the heat-generating component is at a steady-state temperature and the heat dissipation parameter of the heat dissipation component is a steady-state parameter value; According to the second relationship corresponding to the multiple heat-generating components, the state space equation of the server to be processed is determined; the state space equation includes the first-order derivative of the temperature adjustment amount, the system matrix, the state vector, the input matrix and the third relationship between the input vector; the state vector includes the temperature difference within the first time length corresponding to the multiple heat-generating components.

6. The heat dissipation control method according to claim 1, wherein: Determining the heat dissipation model corresponding to the server to be processed according to the state-space equation includes: Determining the heat transfer coefficients and heat capacities corresponding to the plurality of heat-generating components, and the steady-state heat dissipation parameters of the heat dissipation components, through experimental calibration; Determining a system matrix and an input matrix in the state-space equation according to the heat transfer coefficients and heat capacities corresponding to the plurality of heat-generating components, and the steady-state heat dissipation parameters of the heat dissipation components; Based on the state-space equation, a heat dissipation model of the server to be processed is determined according to the system matrix and the input matrix.

7. The heat dissipation control method according to claim 6, characterized in that: The heat transfer coefficients corresponding to the plurality of heat generating components are determined by experimental calibration, including: For each heat-generating component, the heat-generating component is placed in a constant temperature environment, operated at constant power, and the heat dissipation parameter is set to a constant value. After the server to be processed reaches a thermal equilibrium state, the heat-generating power of the heat-generating component and the current temperature, the ambient temperature, and the actual value of the heat dissipation parameter of the heat dissipation component are obtained. The heat transfer coefficient corresponding to the heat-generating component is determined based on the heat-generating power, the current temperature, the ambient temperature and the actual value.

8. The heat dissipation control method according to claim 6, wherein: The step of determining the steady-state heat dissipation parameters of the heat dissipation component through test calibration includes: For each heat-generating component, obtaining the heat-generating power and the ambient temperature of the heat-generating component, and determining the steady-state heat dissipation parameter of the heat-dissipating component corresponding to the heat-generating component based on the heat-generating power and the ambient temperature based on a second relationship; The steady-state heat dissipation parameters of the heat dissipation component are determined according to the steady-state heat dissipation parameters corresponding to each heat-generating component.

9. The heat dissipation control method according to claim 8, characterized in that: The test calibration is to determine the heat capacities corresponding to the plurality of heat generating components, including: For each heat-generating component, when the heat dissipation parameter of the heat dissipation component is 0 and the heat dissipation power is 0, the temperature difference, heat generation power, and second time duration of the heat-generating component are obtained, and the heat capacity corresponding to the heat-generating component is determined based on the temperature difference, heat generation power, and second time duration of the heat-generating component in the second time duration.

10. An electronic device, characterized in that: include: memory for storing computer programs; A processor, configured to implement the steps of the heat dissipation control method according to any one of claims 1 to 9 when executing the computer program.

11. 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 heat dissipation control method according to any one of claims 1 to 9 are implemented.

12. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the heat dissipation control method according to any one of claims 1 to 9 are implemented.

Images