A graphics card cooling module control method and system based on ambient temperature

Abstract

The application provides a graphics card heat dissipation module control method and system based on ambient temperature, and relates to the technical field of graphics card heat dissipation.The method comprises the following steps: acquiring the performance characteristics and external ambient temperature of a graphics card in real time; selecting important features of the performance characteristics through a Lasso regression model; dynamically modeling the graphics card heat dissipation module according to the important features and the external ambient temperature and in combination with a Gaussian process regression algorithm; determining a graphics card temperature prediction model based on the dynamic modeling result; controlling the graphics card heat dissipation module according to the output of the graphics card temperature prediction model and in combination with the target temperature of the graphics card; and controlling the graphics card heat dissipation module until the working state of the graphics card heat dissipation module reaches a target working state. The application can dynamically adjust the fan rotating speed or power distribution and improve the performance stability of the graphics card.

Description

A method and system for controlling a graphics card cooling module based on ambient temperature. Technical Field

[0001] This invention relates to the field of graphics card heat dissipation technology, and in particular to a method and system for controlling a graphics card heat dissipation module based on ambient temperature. Background Technology

[0002] A graphics card (GPU), or graphics processing unit, is a dedicated chip in a computer used for tasks such as image rendering, video processing, and artificial intelligence calculations. It generates a significant amount of heat during operation. The cooling module is the heat dissipation device on the graphics card, typically including a fan, heatsinks, and heat pipes, used to conduct and dissipate the heat generated by the graphics card, preventing overheating. Ambient temperature-based graphics card cooling module control methods consider both the ambient temperature (such as room temperature and case temperature) to achieve more precise and intelligent heat dissipation control.

[0003] By incorporating ambient temperature as a control factor, the graphics card cooling system can operate more intelligently and efficiently, ensuring the stability of the graphics card's performance while reducing energy consumption and noise. Simultaneously, it can extend hardware lifespan and prevent performance degradation or malfunctions caused by overheating, which has significant engineering application value for high-performance computing and energy conservation.

[0004] However, existing graphics card cooling module control methods mainly rely on the temperature changes of the graphics card itself to control the fan speed or cooling intensity. They usually use fixed thresholds or simple linear control strategies, ignoring factors such as external ambient temperature, load fluctuations, and aging of the cooling system. This can easily lead to problems such as untimely cooling response, frequent fan start-stop, excessive power consumption, or sudden increase in noise, thus limiting the full performance of the graphics card and the long-term stable operation of the system. Summary of the Invention

[0005] In view of the shortcomings of the prior art, the purpose of this invention is to provide a graphics card heat dissipation module control method based on ambient temperature. This method can solve the technical problems of existing graphics card heat dissipation module control methods, which mainly rely on the temperature changes of the graphics card itself to control the fan speed or heat dissipation intensity. They usually adopt fixed thresholds or simple linear control strategies, ignoring factors such as external ambient temperature, load fluctuations and aging of the heat dissipation system. These methods are prone to problems such as untimely heat dissipation response, frequent fan start-stop, excessive energy consumption or sudden increase in noise.

[0006] A first aspect of this invention provides a method for controlling a graphics card cooling module based on ambient temperature, comprising:

[0007] S1: Real-time acquisition of graphics card performance characteristics and external ambient temperature;

[0008] S2: Select important features for performance characteristics using the Lasso regression model;

[0009] S3: Based on key features and external ambient temperature, and combined with Gaussian process regression algorithm, the graphics card cooling module is dynamically modeled;

[0010] S4: Based on the dynamic modeling results, determine the graphics card temperature prediction model;

[0011] S5: Based on the output of the graphics card temperature prediction model and the target temperature of the graphics card, control the graphics card cooling module.

[0012] S6: Control of the graphics card cooling module is completed until the working state of the graphics card cooling module reaches the target working state.

[0013] A second aspect of this invention provides a graphics card cooling module control system based on ambient temperature, comprising: a processor and a memory;

[0014] The memory stores a program or instruction that can run on the processor, and when the program or instruction is executed by the processor, it implements the steps of the ambient temperature-based graphics card cooling module control method of the first aspect.

[0015] A third aspect of the present invention provides a readable storage medium on which a program or instructions are stored, and when the program or instructions are executed by a processor, implement the steps of the graphics card cooling module control method based on ambient temperature as described in the first aspect.

[0016] The beneficial effects of the technical solutions provided by the embodiments of the present invention include at least the following:

[0017] In this embodiment of the invention, the performance characteristics of the graphics card and the external ambient temperature are acquired in real time. Then, a Lasso regression model is used to select important features of the performance characteristics, thereby extracting the factors that have the greatest impact on temperature and improving the efficiency and accuracy of the model. Based on the important features and the external ambient temperature, a Gaussian process regression algorithm is used to dynamically model the graphics card cooling module. Based on the dynamic modeling results, a graphics card temperature prediction model is determined. Finally, based on the output of the graphics card temperature prediction model and the target temperature of the graphics card, the graphics card cooling module is controlled until the working state of the graphics card cooling module reaches the target working state. This completes the control of the graphics card cooling module, achieving the goals of energy saving and carbon reduction, extending the service life of the hardware, reducing unnecessary energy consumption and fan noise, further improving heat dissipation efficiency, and enhancing the performance stability of the graphics card. Attached Figure Description

[0018] The accompanying drawings are for illustrative purposes only and are not intended to limit the invention. Throughout the drawings, the same reference numerals denote the same parts. Obviously, the drawings described below are merely some embodiments of the present invention, and those skilled in the art can obtain other drawings based on these drawings without any creative effort.

[0019] Figure 1 is a flowchart illustrating a graphics card cooling module control method based on ambient temperature according to an embodiment of the present invention.

[0020] Figure 2 is a schematic diagram of a graphics card heat dissipation module control system based on ambient temperature provided in an embodiment of the present invention. Detailed Implementation

[0021] To enable those skilled in the art to better understand the technical solutions in the embodiments of the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. It should be understood that these descriptions are merely exemplary and are not intended to limit the scope of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0022] The following description, in conjunction with the accompanying drawings, details the method for controlling a graphics card cooling module based on ambient temperature provided by the present invention through specific embodiments and application scenarios.

[0023] Referring to Figure 1 in the specification, a flowchart illustrating a graphics card heat dissipation module control method based on ambient temperature is shown in an embodiment of the present invention.

[0024] This invention provides a method for controlling a graphics card cooling module based on ambient temperature, which may include the following steps:

[0025] S1: Real-time acquisition of graphics card performance characteristics and external ambient temperature.

[0026] Among them, performance characteristics include a series of key parameters that affect the operating status of the graphics card, such as core temperature, power consumption, frequency, load, cache hit rate and memory access volume, which are used to reflect the real-time operating status of the GPU. The external ambient temperature is the air temperature of the working environment of the graphics card, such as room temperature or the internal temperature of the computer case, which directly affects the heat dissipation efficiency.

[0027] It's worth noting that by simultaneously collecting data on the graphics card's performance characteristics and the external ambient temperature, the system can comprehensively understand its current operating status and cooling conditions, avoiding judgment biases caused by relying solely on the graphics card's internal temperature. Real-time performance ensures rapid response in temperature control, while multi-dimensional data makes subsequent modeling and control more precise and scientific.

[0028] In one possible implementation, the performance characteristics specifically include: the graphics card's core temperature, power consumption, operating frequency, workload, cache hit rate, and memory access.

[0029] The core temperature of a graphics card refers to the temperature of the most important computing core area inside the card. It is the most direct indicator of the card's current thermal state. Excessive temperature can lead to throttling, performance degradation, and even hardware damage. The power consumption of a graphics card indicates the electrical energy consumed by the card during its current operation. The operating frequency of a graphics card is the frequency at which the graphics core processor runs.

[0030] Graphics card usage load indicates the current utilization level of graphics card resources, usually expressed as a percentage. Cache hit rate refers to the proportion of times the graphics card accesses data when the data is already in the cache and does not need to be accessed from video memory. Memory access refers to the frequency or intensity of the graphics card accessing its video memory (such as GDDR6).

[0031] S2: Use the Lasso regression model to select important features for performance characteristics.

[0032] Lasso regression is a linear regression method with L1 regularization, which sparsifies the input features by automatically selecting the most important features for the target variable, while compressing the coefficients of other less important features to zero. Important features refer to the factors that most significantly affect graphics card temperature changes from numerous performance characteristics, used for subsequent modeling to improve efficiency and prediction accuracy.

[0033] It should be noted that Lasso regression automatically filters graphics card performance characteristics, avoiding redundancy and noise interference caused by human experience selection or full modeling. Lasso can adaptively eliminate irrelevant or weakly correlated features, making the entire control system lighter, more accurate and reliable, and laying a solid data foundation for the implementation of intelligent temperature control strategies.

[0034] In one possible implementation, S2 specifically includes:

[0035] S201: Normalize the performance characteristics.

[0036] Normalization refers to standardizing feature values ​​of different dimensions and ranges to bring them into a uniform range (such as 0 to 1), which facilitates effective comparison and processing by the algorithm.

[0037] S202: Based on the normalized performance characteristics, and with the objective of minimizing the objective function of the Lasso regression model, determine the important features:

[0038]

[0039] in, This represents the estimated coefficients, argmin represents taking the minimum value, and y i X represents the graphics card temperature of the i-th sample, where i = 1, ..., n, and n represents the total number of samples. i Let β represent the feature vector of the i-th sample, β represent the set of regression coefficients, and λ represent the regularization parameter. j Let represent the regression coefficient corresponding to the j-th performance feature, where j = 1, ..., p, and p represents the total number of performance features.

[0040] It should be noted that normalization ensures that all features are compared on a level playing field, guaranteeing the fairness and scientific rigor of the selection process. This process not only improves the computational efficiency of subsequent modeling (such as Gaussian process regression) but also enhances the interpretability and generalization ability of the model.

[0041] S3: Based on key features and external ambient temperature, and combined with Gaussian process regression algorithm, dynamically model the graphics card cooling module.

[0042] Gaussian process regression is a nonparametric regression method based on Bayesian theory. It describes the relationship between functions through probability distributions, enabling the prediction of future output values ​​and the quantification of uncertainty. Dynamic modeling refers to establishing a mathematical model reflecting changes in system state over time, based on real-time input data, to characterize the temperature change trend of a graphics card under different conditions.

[0043] It should be noted that by introducing Gaussian process regression, dynamic modeling of the graphics card temperature change process is achieved. Its greatest advantage is that it has strong nonlinear fitting ability, high prediction accuracy, and the ability to quantify uncertainty. Compared with traditional linear models, GPR can not only predict the temperature at the next moment based on current key features (such as power consumption, frequency, and ambient temperature), but also provide the confidence interval of the prediction, giving the system a dual capability of "prediction + confidence".

[0044] In one possible implementation, the graphics card cooling module includes: a fan, a heatsink, a heat pipe, thermal paste, and a heatsink.

[0045] In one possible implementation, S3 specifically includes:

[0046] S301: Models the temperature change process of the graphics card using a Gaussian process as the basic framework.

[0047] [P1,P2,…,P t ,P t+1 ]~N(0,K)

[0048] Among them, P t P represents the graphics card temperature at time t. t+1 P represents the graphics card temperature at time t+1, i.e., the temperature that needs to be predicted. p Let represent the graphics card temperature at time p, where p = 1, 2, ..., t+1, t+1 represents the total number of times, N represents a multivariate Gaussian distribution, and K represents the covariance matrix.

[0049] S302: Based on the modeling results, the similarity between important features and the external ambient temperature is determined by calculating the kernel function.

[0050] K jk =k(X) j ,X k )

[0051] X j =(A t A t-1 ,P t-1 )

[0052] X k =(A t A t-1 ,P t-1 )

[0053] Among them, K jk Indicates input feature X j and input features X k The correlation between them, k represents the kernel function, A t A represents the important eigenvalue at time t. t-1 P represents the important eigenvalue at time t-1. t-1 This represents the graphics card temperature at time t-1.

[0054] S303: Determine the expected predicted temperature of the graphics card based on similarity:

[0055] E(P t+1 |X,P,X t+1 )=K(X t+1 ,X)K(X,X) -1 P

[0056] P = (P1, P2, ..., P t )

[0057] X = (X1, X2, ..., X t )

[0058] Where E represents the expectation symbol, P t+1 Let X represent the graphics card temperature at time t+1, X represent the important feature matrix, and P represent the known graphics card temperature vector. t+1 X represents the key features at time t+1. t P represents the key features at time t. q X represents the graphics card temperature at time q. q Let q represent the key features at time q, where q = 1, 2, ..., t.

[0059] S304: Dynamically model the graphics card cooling module based on the expected predicted temperature.

[0060] f = E[P] t+1 |X,P,(A t+1 A t ,P t )]

[0061] Where f represents the predicted output graphics card temperature.

[0062] It's worth noting that the dynamic modeling process, combined with time-series characteristics, enables the model to adapt to complex operating conditions such as graphics card load fluctuations and environmental changes, truly achieving intelligent and stable temperature control strategies. This method is particularly suitable for highly dynamic and sensitive heat dissipation scenarios, helping the cooling module to proactively intervene before overheating occurs, improving response speed and control effectiveness.

[0063] S4: Based on the dynamic modeling results, determine the graphics card temperature prediction model.

[0064] The dynamic modeling result refers to the output of a mathematical model established using methods such as Gaussian process regression, reflecting the trend of graphics card temperature changes over time. It typically includes predicted temperature values ​​and characteristic relationships. A graphics card temperature prediction model is a mathematical expression that, based on the regression relationships obtained during the modeling process and inputting current system state parameters, can predict the graphics card temperature at a future point in time.

[0065] It should be noted that simplifying the complex dynamic modeling results of the Gaussian process into a linear regression model with a clear structure and high computational efficiency makes temperature prediction easier to implement and perform real-time calculations in practical deployments. Compared to complex nonlinear algorithms, the linear model, while maintaining a certain level of prediction accuracy, has a faster response speed and lower resource consumption, making it suitable for applications in embedded systems or hardware driver modules.

[0066] In one possible implementation, S4 specifically refers to:

[0067] The dynamic modeling results at each time point are fitted into a linear regression equation to generate a graphics card temperature prediction model.

[0068] The specific expression for the graphics card temperature prediction model is as follows:

[0069] T pred (t+Δt)=β0+β1·T gpu (t)+β2·P gpu (t)+β3·T amb (t)+β4·Fanspeed(t)

[0070] Among them, T pred (t+Δt) represents the graphics card temperature predicted by t+Δt, β0 represents the constant term, β1 represents the regression coefficient of temperature, and T gpu (t) represents the graphics card temperature at time t, β2 represents the regression coefficient of power consumption, and P gpu (t) represents the graphics card power consumption at time t, β3 represents the regression coefficient of the external ambient temperature, and T amb (t) represents the ambient temperature at time t, β4 represents the regression coefficient of fan speed, and Fanspeed(t) represents the fan speed at time t.

[0071] Specifically, β0 represents the graphics card temperature predicted by the model when all input features (such as graphics card temperature, power consumption, load, etc.) are zero.

[0072] In this invention, β0 is 35, β1 is 0.3, β2 is 1.5, β3 is 0.5, and β4 is -0.7.

[0073] It should be noted that explicit regression coefficients also enhance the interpretability of the model, allowing engineers to clearly understand the impact of each variable on temperature, which is beneficial for later optimization and parameter tuning. This effectively connects the modeling and control stages, providing a solid and practical predictive basis for subsequent fan control based on the prediction results, making it a key bridge in the entire intelligent heat dissipation control system.

[0074] S5: Based on the output of the graphics card temperature prediction model and the target temperature of the graphics card, control the graphics card cooling module.

[0075] The target temperature of the graphics card is a mathematical model obtained from the previous modeling step. It can be used to predict the temperature trend of the graphics card in the future and provide a basis for the control strategy.

[0076] It should be noted that those skilled in the art can set the target temperature according to the actual situation, and this invention does not impose any limitations.

[0077] Specifically, the predicted temperature is combined with the preset target temperature to form a closed-loop control strategy, which precisely regulates the graphics card temperature by adjusting the operating status of the cooling module in real time. Unlike the traditional passive mode of "regulating only when the current temperature exceeds the limit", this method has feedforward control capabilities, which can actively prevent overheating before it occurs, resulting in faster response and higher stability.

[0078] In one possible implementation, S5 specifically involves controlling the graphics card cooling module using a PID algorithm.

[0079] Among them, the PID algorithm, which stands for Proportional-Integral-Derivative control, is a commonly used closed-loop feedback control algorithm. It is widely used in industrial automation, temperature control, robot control, flight control systems and other fields, and is also one of the mainstream adjustment methods in graphics card fan control and heat dissipation systems.

[0080] It should be noted that PID control can comprehensively judge and adjust based on the current value (P), cumulative value (I), and trend (D) of the temperature error, possessing extremely high control precision. Compared to simple control methods that adjust only based on the current error, PID can more quickly and accurately pull the temperature back to the target value, reducing the risk of graphics card overheating.

[0081] In one possible implementation, S5 specifically includes:

[0082] S501: Calculate the error between the predicted temperature and the target temperature.

[0083] e(t) = T GPU -T target

[0084] Where e(t) represents the temperature error at time t, T GPU This indicates the current temperature of the graphics card, T. target This indicates the current target temperature of the graphics card.

[0085] S502: Based on the error, the PID controller determines the fan adjustment amount.

[0086]

[0087] Where Δ represents the increment, T fan Indicates fan speed, K p K represents the proportional gain. i Indicates the integral gain, d represents the differential sign, and K represents the integral gain. d This represents the differential gain.

[0088] S503: Controls the fan speed based on the adjustment amount to control the graphics card cooling module.

[0089] Fan Speed(t) = Fan Speed min +u(t)·(Fan Speed max -Fan Speed min )

[0090] Where FanSpeed(t) represents the fan speed at time t, Fan Speed min The value represents the minimum fan speed, u(t) represents the control signal at time t, and Fan Speed. max This indicates the fan's maximum speed.

[0091] It should be noted that PID control can avoid energy waste and fan noise problems caused by excessive cooling, achieving a multi-objective balance between performance, lifespan, and energy saving. By comparing the predicted temperature with the target temperature as the basis for judgment, the control system possesses stronger adaptability and intelligence, making it one of the key elements in achieving intelligent heat dissipation.

[0092] S6: Control of the graphics card cooling module is completed until the working state of the graphics card cooling module reaches the target working state.

[0093] The "operating state" refers to the state of the cooling module during operation, including the overall performance of parameters such as fan speed, cooling intensity, and power consumption. The "target operating state" refers to the optimal cooling operation state that the system aims to achieve, which typically means a low-power, low-noise, and high-efficiency operating mode while ensuring a suitable graphics card temperature.

[0094] It's important to note that by setting the system to "continue until the target operating state is reached," the system won't stop after a single control command. Instead, it continuously self-adjusts and self-verifies, ensuring stable and reliable control results. This feedback-driven, adaptive mechanism significantly improves heat dissipation efficiency and temperature control accuracy, while also extending the lifespan of hardware components such as the graphics card and fans. It embodies a highly intelligent and sustainable control philosophy and is a crucial final step in the entire intelligent cooling system.

[0095] The beneficial effects of the technical solutions provided by the embodiments of the present invention include at least the following:

[0096] In this embodiment of the invention, the performance characteristics of the graphics card and the external ambient temperature are acquired in real time. Then, a Lasso regression model is used to select important features of the performance characteristics, thereby extracting the factors that have the greatest impact on temperature and improving the efficiency and accuracy of the model. Based on the important features and the external ambient temperature, a Gaussian process regression algorithm is used to dynamically model the graphics card cooling module. Based on the dynamic modeling results, a graphics card temperature prediction model is determined. Finally, based on the output of the graphics card temperature prediction model and the target temperature of the graphics card, the graphics card cooling module is controlled until the working state of the graphics card cooling module reaches the target working state. This completes the control of the graphics card cooling module, achieving the goals of energy saving and carbon reduction, extending the service life of the hardware, reducing unnecessary energy consumption and fan noise, further improving heat dissipation efficiency, and enhancing the performance stability of the graphics card.

[0097] Referring to Figure 2 in the specification, a schematic diagram of a graphics card heat dissipation module control system based on ambient temperature is shown in an embodiment of the present invention.

[0098] This invention provides a graphics card heat dissipation module control system 20 based on ambient temperature, including: a processor 201 and a memory 202;

[0099] The memory 202 stores programs or instructions that can run on the processor 201. When the program or instructions are executed by the processor 201, they implement the steps of the above-described graphics card heat dissipation module control method based on ambient temperature and achieve the same technical effect. To avoid repetition, the present invention will not elaborate further.

[0100] It should be understood that the processor 201 in this embodiment of the invention may be a central processing unit (CPU), or it may be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor.

[0101] It should also be understood that the memory 202 in the embodiments of the present invention can be volatile memory or non-volatile memory, or may include both volatile and non-volatile memory. The non-volatile memory can be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), or flash memory. The volatile memory can be random access memory (RAM), which is used as an external cache. By way of example, but not limitation, many forms of random access memory (RAM) are available, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate synchronous DRAM (DDR SDRAM), enhanced synchronous DRAM (ESDRAM), synchronous linked DRAM (SLDRAM), and direct rambus RAM (DR RAM).

[0102] The above embodiments can be implemented, in whole or in part, by software, hardware (such as circuits), firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, as a computer program product. A computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, all or part of the flow or function according to the embodiments of the present invention is generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. Computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., infrared, wireless, microwave, etc.) means. A computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more sets of available media. Available media can be magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media (e.g., DVDs), or semiconductor media. Semiconductor media can be solid-state drives.

[0103] It should be understood that, in various embodiments of the present invention, the order of the above-mentioned process numbers does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0104] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.

[0105] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the devices, apparatuses, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0106] In the embodiments provided by this invention, it should be understood that the disclosed devices, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another device, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0107] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0108] In addition, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0109] If a function is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0110] This invention provides a readable storage medium that stores a program or instructions on the storage medium. When the program or instructions are executed by a processor, they implement the steps of the above-described graphics card cooling module control method based on ambient temperature, and achieve the same technical effect. To avoid repetition, this invention will not repeat the above description.

[0111] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the embodiments of the present invention, and are not intended to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the protection scope of the present invention.

Claims

1. A method for controlling a graphics card cooling module based on ambient temperature, characterized in that, include: S1: Real-time acquisition of graphics card performance characteristics and external ambient temperature; S2: Select important features for the performance characteristics using a Lasso regression model; S3: Based on the important features and the external ambient temperature, dynamically model the graphics card cooling module using a Gaussian process regression algorithm; S3 specifically includes: S301: Model the temperature change process of the graphics card using a Gaussian process as the basic framework: Among them, P t P represents the graphics card temperature at time t. t+1 P represents the graphics card temperature at time t+1, i.e., the temperature that needs to be predicted. p Let represent the graphics card temperature at time p, where p = 1, 2, ..., t+1, t+1 represents the total number of times, N represents a multivariate Gaussian distribution, and K represents the covariance matrix; S302: Based on the modeling results, determine the similarity between the important features and the external ambient temperature by calculating the kernel function. Among them, K jk Indicates input feature X j and input features X k The correlation between them, k represents the kernel function, A t A represents the important eigenvalue at time t. t-1 P represents the important eigenvalue at time t-1. t-1 S303: Based on the similarity, determine the expected predicted temperature of the graphics card. Where E represents the expectation symbol, P t+1 Let X represent the graphics card temperature at time t+1, X represent the important feature matrix, and P represent the known graphics card temperature vector. t+1 X represents the key features at time t+1. t P represents the key features at time t. q X represents the graphics card temperature at time q. q S304: Based on the expected predicted temperature, dynamically model the graphics card cooling module. This represents the key features at time q, where q = 1, 2, ..., t. Where f represents the predicted output graphics card temperature; S4: Based on the dynamic modeling results, determine the graphics card temperature prediction model; S5: According to the output of the graphics card temperature prediction model, combined with the target temperature of the graphics card, control the graphics card cooling module; control the graphics card cooling module through a PID algorithm; S6: until the working state of the graphics card cooling module reaches the target working state, the control of the graphics card cooling module is completed.

2. The graphics card heat dissipation module control method based on ambient temperature according to claim 1, characterized in that, The specific performance characteristics include: graphics card core temperature, graphics card power consumption, graphics card operating frequency, graphics card usage load, cache hit rate, and memory access.

3. The graphics card heat dissipation module control method based on ambient temperature according to claim 1, characterized in that, S2 specifically includes: S201: normalizing the performance characteristics; S202: determining the important features based on the normalized performance characteristics, with the goal of minimizing the objective function of the Lasso regression model.

4. The graphics card heat dissipation module control method based on ambient temperature according to claim 1, characterized in that, The graphics card cooling module includes: a fan, a heatsink, a heat pipe, thermal paste, and a heatsink.

5. The graphics card heat dissipation module control method based on ambient temperature according to claim 1, characterized in that, Specifically, S4 involves fitting the dynamic modeling results at each time point into a linear regression equation to generate a graphics card temperature prediction model.

6. The graphics card heat dissipation module control method based on ambient temperature according to claim 1, characterized in that, S5 specifically includes: S501: calculating the error between the predicted temperature and the target temperature; S502: determining the fan adjustment amount through a PID controller based on the error; S503: controlling the fan speed based on the adjustment amount to control the graphics card cooling module.

7. A graphics card cooling module control system based on ambient temperature, characterized in that, include: Processor and memory; The memory stores programs or instructions that can run on the processor, which, when executed by the processor, implement the steps of the graphics card cooling module control method based on ambient temperature as described in any one of claims 1 to 6.

8. A readable storage medium, characterized in that, The readable storage medium stores a program or instructions that, when executed by a processor, implement the steps of the graphics card cooling module control method based on ambient temperature as described in any one of claims 1 to 6.

Images