Temperature control method, electronic device, and storage medium

Abstract

This application discloses a temperature control method, an electronic device, and a storage medium, relating to the field of electronic device technology. In the method of this application, a predicted temperature of the heating component in a first time period can be obtained based on the current temperature and current power consumption of the heating component. Since the heat dissipation capacity of the heat dissipation device can be improved in advance based on at least one of the current power consumption and predicted temperature of the heating component, and the heat dissipation device can be controlled in real time based on at least one of the target temperature and current temperature of the heating component, the composite control signal obtained based on at least one of the current power consumption and predicted temperature of the heating component, and at least one of the target temperature and current temperature of the heating component, can solve the problem of excessively high heating component temperature caused by temperature control lag in traditional temperature control technology, and can also solve the problem of current temperature error of the heating component.

Description

Technical Field

[0001] This application relates to the field of electronic equipment technology, and in particular to temperature control methods, electronic devices, and storage media. Background Technology

[0002] In traditional temperature control technology, the operating parameters of the heat dissipation device are usually adjusted after the temperature of one or more heat-generating components in an electronic device exceeds a temperature threshold. This is done to enhance the heat dissipation capacity of the heat dissipation device and prevent the heat-generating components from overheating. For example, when the temperature of a heat-generating component exceeds 80 degrees Celsius, the fan speed in the electronic device will be increased to dissipate heat more efficiently.

[0003] Currently, with the development of technologies such as artificial intelligence, the temperature of heat-generating components in many electronic devices (such as servers used for model training) rises relatively quickly during operation. Traditional temperature control technology has a lag, which makes it easy for heat-generating components to overheat. Summary of the Invention

[0004] This application provides a temperature control method, electronic device, computer-readable storage medium, and program product to at least solve the problem of excessively high temperature of heating components caused by temperature control lag in the related art.

[0005] This application provides a temperature control method, including: Collect the current temperature and current power consumption of at least one heat-generating component in the electronic device; Based on the current temperature and the current power consumption, the predicted temperature of the heating component in the first time period is obtained; A composite control signal is obtained based on at least one of the current power consumption and the predicted temperature, and at least one of the target temperature of the heating component and the current temperature; The composite control signal is output to the heat dissipation device of the heat-generating component to control the heat dissipation device to cool down the heat-generating component.

[0006] This application also provides an electronic device, including: At least one heating element; Heat dissipation device; The acquisition circuit includes a temperature acquisition circuit and a power consumption acquisition circuit. The temperature acquisition circuit is used to acquire the current temperature of the at least one heating component, and the power consumption acquisition circuit is used to acquire the current power consumption of the at least one heating component. The main controller is used to receive the current power consumption and current temperature of the at least one heat-generating component sent by the acquisition circuit, and control the heat dissipation device to cool down the heat-generating component based on any of the methods described above.

[0007] This application also provides a computer-readable storage medium storing a computer program, wherein the computer program, when executed by a processor, implements the steps of the above-described temperature control method.

[0008] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the above-described temperature control method.

[0009] In some embodiments of this application, the predicted temperature of a heat-generating component in a first time period can be obtained based on its current temperature and current power consumption. Since the heat dissipation capacity of the heat dissipation device can be improved in advance based on at least one of the current power consumption and predicted temperature of the heat-generating component, and the heat dissipation device can be controlled in real time based on at least one of the target temperature and current temperature of the heat-generating component, the composite control signal obtained based on at least one of the current power consumption and predicted temperature of the heat-generating component, and at least one of the target temperature and current temperature of the heat-generating component, can solve the problem of excessively high temperature of the heat-generating component caused by temperature control lag in traditional temperature control technology, and can also solve the problem of current temperature error of the heat-generating component. Attached Figure Description

[0010] To more clearly illustrate the embodiments of this application, the accompanying drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0011] Figure 1 A schematic flowchart illustrating a temperature control method provided for some embodiments of this application; Figure 2 A schematic diagram of the functional modules of the composite PID control algorithm provided for some embodiments of this application; Figure 3 Schematic diagrams of the structure of electronic devices provided for some embodiments of this application; Figure 4 A schematic diagram of the modules of an electronic device provided for some embodiments of this application. Detailed Implementation

[0012] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the protection scope of this application.

[0013] It should be noted that, in the description of this application, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. The terms "first," "second," etc., in this application are used to distinguish similar objects and are not used to describe a specific order or sequence.

[0014] To enable those skilled in the art to better understand the present application, the present application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0015] In this application, electronic equipment refers to devices with computing capabilities, such as servers, tablets, desktop computers, and switches. Inside an electronic device, one or more heat-generating components and cooling devices can be installed. A heat-generating component is a hardware component that converts electrical energy into heat energy and raises its own or the surrounding environment's temperature during operation, such as a Central Processing Unit (CPU), a Graphics Processing Unit (GPU), a hard drive, and a network card. A cooling device is used to cool the heat-generating components, such as a fan or a coolant system. A fan accelerates airflow, allowing cool air from outside the electronic device to enter and exchange heat with the heat-generating components, thus cooling them. A coolant system uses a liquid (such as water) as a heat exchange medium, and the liquid exchanges heat with the heat-generating components during its flow, thereby cooling them.

[0016] Typically, one or more cooling devices can be installed in different areas of an electronic device. These devices can perform different cooling tasks. For example, a first set of fans located at the air inlet of the electronic device draws cool air into the device. After heat exchange with the heat-generating components, the cool air becomes hot air. A second set of fans located at the air outlet of the electronic device expels the hot air. The first and second sets of fans can accelerate airflow within the electronic device to cool down multiple heat-generating components. For another example, a third set of fans can be installed near the accelerator board. This third set of fans can accelerate airflow around the accelerator board to cool down heat-generating components (such as the graphics processor) on the accelerator board. Yet another example is a fourth set of fans near the motherboard. This fourth set of fans can accelerate airflow around the motherboard to cool down heat-generating components (such as the central processing unit) on the motherboard.

[0017] Currently, in some traditional temperature control technologies, each heat-generating component has its own corresponding temperature threshold. For example, the temperature threshold for a central processing unit (CPU) is 80 degrees Celsius, and for a network card (NIC) it is 85 degrees Celsius. The baseboard management controller (BMC) inside an electronic device can collect the current temperature of each heat-generating component using temperature sensors and other devices. When the current temperature of a heat-generating component exceeds its corresponding temperature threshold, the BMC can control the corresponding heat dissipation device to enhance its cooling capacity and prevent the component from overheating. For example, when the CPU's current temperature exceeds 80 degrees Celsius, the BMC can control the fourth set of fans near the motherboard, the first set of fans at the electronic device's air inlet, and the second set of fans at the electronic device's air outlet to increase their speeds. This accelerates airflow within the electronic device and around the motherboard, resulting in more efficient cooling of the CPU and preventing it from overheating.

[0018] Traditional temperature control technology suffers from lag, making it suitable for heat-generating components with slow temperature rise rates. However, when the temperature of a heat-generating component rises rapidly, it cannot effectively cool the component in a timely manner, leading to overheating. For example, suppose the image processor temperature exceeds 80 degrees Celsius at 11:20:30. At this point, a control signal can be triggered to increase the fan speed. Since generating and transmitting the control signal takes time, the fan might not start running at the higher speed until 11:20:32 (i.e., it has lag). However, between 11:20:30 and 11:20:32, the temperature of the electronic device might have already soared to 100 degrees Celsius, thus causing the device to overheat.

[0019] In view of this, this application provides a temperature control method that can solve the problem of excessively high temperature of heat-generating components caused by temperature control lag. The temperature control method can be applied to a controller in an electronic device. The controller can be a board management controller, a complex programmable device, etc. Specifically, the controller can execute the temperature control method of this application according to an iterative cycle. The duration of each iterative cycle can be 5 seconds, 10 seconds, etc., and this application does not limit this. For example, after the electronic device starts working, the first 1-5 seconds can be used as the first iterative cycle, the second 6-10 seconds as the second iterative cycle, and so on. In each iterative cycle, the controller can execute the temperature control method of this application separately. In this way, the heat dissipation device can be controlled in real time according to the heat dissipation state of the heat-generating component in each iterative cycle, ensuring that the heat dissipation capacity of the heat dissipation device is adapted to the heat dissipation state of the heat-generating component in real time. Based on the above description, and in conjunction with reference to... Figure 1 This is a schematic flowchart of a temperature control method provided in some embodiments of this application. Figure 1 In this process, the temperature control method includes the following steps: Step S101: Collect the current temperature and current power consumption of at least one heat-generating component in the electronic device.

[0020] Specifically, the current temperature refers to the temperature of the heating component within the current iteration cycle. For any heating component, the average temperature of the heating component at multiple time points within the current iteration cycle is calculated, and the result can be used as the current temperature of the heating component. Of course, the temperature of the heating component at any one time point within the current iteration cycle can also be used as the current temperature of the heating component. For example, the temperature of the heating component at the beginning of the current iteration cycle can be used as the current temperature. This application does not impose any restrictions on this.

[0021] Similarly, the current power consumption refers to the power consumption of the heat-generating component within the current iteration cycle. For any heat-generating component, the average power consumption of the component at multiple time points within the current iteration cycle is calculated, and the result can be used as the current power consumption of the heat-generating component. Of course, the power consumption of the heat-generating component at any one time point within the current iteration cycle can also be used as the current power consumption. For example, the power consumption of the heat-generating component at the end of the current iteration cycle can be used as the current power consumption. This application does not impose any restrictions on this.

[0022] Furthermore, in this embodiment, the controller can collect the current temperature of multiple heat-generating components in parallel using devices such as temperature sensors. Specifically, when a heat-generating component comprises multiple sub-components with significant temperature differences between them, the current temperatures of multiple sub-components within the same heat-generating component can be collected in parallel. That is, the same heat-generating component can have multiple current temperatures, and these multiple current temperatures represent the current temperatures of different sub-components within the corresponding heat-generating component. For example, a network interface card (NIC) typically includes a NIC body and an optical module. During NIC operation, the temperature difference between the NIC body and the optical module can be significant. Therefore, when collecting the current temperature of the NIC, the current temperatures of both the NIC body and the optical module can be collected in parallel. Thus, the NIC has two current temperatures: the current temperature of the NIC body and the current temperature of the optical module. It can be understood that when there are significant temperature differences between different areas of a heat-generating component, temperatures can also be collected separately in multiple areas of the heat-generating component to obtain multiple current temperatures for that component.

[0023] In this embodiment, the controller can use a power consumption acquisition circuit to collect the current voltage and current of multiple heat-generating components in parallel. Based on the current voltage and current, the current power consumption of each heat-generating component can be obtained. Specifically, when a heat-generating component includes multiple sub-components and the power consumption differences between different sub-components are significant, the current voltage and current of multiple sub-components within the same heat-generating component can be collected in parallel, and the current power consumption of each sub-component can be obtained based on the current voltage and current. That is, a single heat-generating component can have multiple current power consumptions, and these multiple current power consumptions represent the current power consumption of different sub-components within the corresponding heat-generating component. For example, a central processing unit (CPU) typically includes multiple power supply pins, and the voltage and current of different power supply pins are different. In this case, the current voltage and current of multiple power supply pins of the CPU can be collected in parallel, and the current power consumption of each power supply pin can be obtained based on the current voltage and current of each power supply pin. Thus, the CPU can have multiple current power consumptions, which are the current power consumptions of each power supply pin.

[0024] Based on the above description, after executing step S101, data similar to that shown in Table 1 can be obtained.

[0025] Table 1 Current temperature and power consumption of the heating element

[0026] Taking component 1 in Table 1 as an example, it means that component 1 has two current temperatures A11 and A12 and three current power consumptions B11, B12 and B13.

[0027] The current temperatures of multiple components can be stored in the form of a first vector, and the current power consumption of multiple components can be stored in the form of a second vector. For example, the current temperatures of multiple components can be stored as vector P1 = {current temperature of component 1, current temperature of component 2, ..., current temperature of component n}, and the current power consumption of multiple components can be stored as vector P2 = {current power consumption of component 1, current power consumption of component 2, ..., current power consumption of component n}.

[0028] Step S102: Based on the current temperature and current power consumption, obtain the predicted temperature of the heat-generating component in the first time period.

[0029] In this embodiment, an electronic device model can be pre-built. In the electronic device model, the correlation between the temperature and power consumption of the heat-generating component and the heat dissipation capacity of the heat dissipation device can be constructed, and a temperature prediction function for the heat-generating component can be constructed, as shown in expressions (1) to (4).

[0030] (1) (2) (3) predict_temperature(float T_current,float current_power,thermal_model_t model,float delta_t) { float T_steady = model.T_ambient + current_power *model.R_ja; return T_current+ (T_steady-T_current)*exp(-delta_t / model.tau) (4) In expression (1), The expression (1) represents the temperature rise of the heating component, and current_power represents the current power consumption of the heating component. It is proportional to the current power consumption P.

[0031] In expression (2), The expression (2) represents the cooling of the heat-generating component, where M represents the heat dissipation capacity of the heat dissipation device. It is directly proportional to the heat dissipation capacity M of the heat dissipation device.

[0032] In expression (3), represents the heating of the heating component. With cooling Integrating the difference yields the temperature change of the heating component. For example, in the current iteration cycle, the temperature increase of the heating component... With cooling By integrating the difference, we can obtain the temperature change of the heating component in the current iteration cycle.

[0033] Expression (4) is a temperature prediction function used to obtain the ambient temperature of the environment where the heating component is located and the junction-to-ambient thermal resistance of the heating component. Based on the ambient temperature, current power consumption, and junction-to-ambient thermal resistance, it determines the steady-state temperature of the heating component and predicts the temperature of the heating component in the first time period delta_t (e.g., 5 seconds after the current iteration cycle) based on the difference between the current temperature, the steady-state temperature, and the current temperature. Specifically, in this function, model represents the parameter container of the heating component. Each heating component has its own corresponding parameter container. The parameters in the parameter container can be preset parameters or parameters obtained in real time through sensors or other devices. model.T_ambient represents the ambient temperature of the heating component obtained from the parameter container of the heating component. model.R_ja represents the junction-to-ambient thermal resistance of the heating component obtained from the parameter container of the heating component. P*model.R_ja represents the temperature rise obtained from the current power consumption of the heating component (i.e., the above). T_current represents the current temperature of the heat-generating component. model.tau represents the thermal time constant obtained from the parameter container of the heat-generating component. T_steady represents the steady-state temperature that the heat-generating component can reach, calculated based on the current power consumption of the heat-generating component and the ambient temperature. The steady-state temperature refers to the temperature at which the heat dissipation rate of the heat-generating component is equal to its heat generation rate.

[0034] T_steady - T_current represents the difference between the current temperature and the steady-state temperature of the heating element. It indicates the total temperature change required for the heating element to reach a steady state from its current temperature. For example, assuming the current temperature of the heating element is 55 degrees Celsius, and the calculated steady-state temperature based on the current power consumption and ambient temperature is 60 degrees Celsius, then the total temperature change required for the heating element to reach a steady state is 5 degrees Celsius. A positive total temperature change indicates an increase in temperature, while a negative total temperature change indicates a decrease in temperature. It's understandable that the temperature change from the current temperature (e.g., 55 degrees Celsius) to the steady-state temperature (e.g., 60 degrees Celsius) is a slow process. If the heating element reaches its steady-state temperature in the second time period after the current iteration cycle (e.g., 15 seconds after the current iteration cycle), and this second time period follows the first time period, then in the first time period, the heating element's temperature has not yet reached the steady-state temperature, but is approaching it. For example, suppose we need to predict the temperature of a heating component 10 seconds after the current iteration cycle, but the heating component needs to reach a steady-state temperature 15 seconds after the current iteration cycle. In this case, the temperature of the heating component has not reached a steady-state temperature at 10 seconds after the current iteration cycle, but it is close to the steady-state temperature.

[0035] Therefore, in expression (4), exp(-delta_t / model.tau) represents an exponential decay function with time as the variable, and (T_steady-T_current)*exp(-delta_t / model.tau) represents the amount of temperature change that the heating component has completed by the first time period. For example, suppose the current temperature of the heating component is 55 degrees Celsius, and the heating component needs to reach the steady-state temperature 10 seconds after the current iteration cycle. If the temperature prediction function in expression (4) predicts that the temperature of the heating component will be 57 degrees Celsius 5 seconds after the current iteration cycle (i.e., the first time period), it means that the heating component has completed a 2-degree Celsius temperature increase 5 seconds after the current iteration cycle.

[0036] The sum of the current temperature and the temperature change of the heating element, i.e., T_current + (T_steady - T_current) * exp(-delta_t / model.tau), represents the predicted temperature of the heating element in the first time period.

[0037] Based on the above description, the current temperature and current power consumption of each heat-generating component collected in step S101 can be substituted into the temperature prediction function shown in expression (4) to obtain the temperature of each heat-generating component in the first time period.

[0038] Step S103: Based on at least one of the current power consumption and the predicted temperature, and at least one of the target temperature and the current temperature of the heating component, a composite control signal is obtained.

[0039] Specifically, due to the thermal inertia of the heat-generating component, the temperature rise caused by its current power consumption is delayed. That is, the power consumption of the heat-generating component at the current time point (i.e., the current power consumption) actually causes the component to heat up after that time point. Therefore, by analyzing the current power consumption of the heat-generating component, we can predict its temperature change trend after the current iteration cycle. For example, if the current power consumption of the heat-generating component is relatively high, we can predict that the temperature rise after the current iteration cycle will be relatively large; conversely, if the current power consumption of the heat-generating component is relatively low, we can predict that the temperature rise after the current iteration cycle will be relatively small.

[0040] Based on the predicted temperature change trend, the heat dissipation device of the heating component can be controlled in advance before the temperature of the heating component reaches the temperature threshold. This ensures that the heat dissipation capacity of the heat dissipation device has been improved when or before the temperature of the heating component reaches the temperature threshold, so as to efficiently dissipate heat from the heating component in a timely manner and solve the problem of excessively high temperature of the heating component caused by temperature control lag.

[0041] Similarly, based on the predicted temperature of the heating element in the first period, the heat dissipation device of the heating element can be controlled in advance to solve the problem of excessively high temperature of the heating element caused by temperature control lag.

[0042] Understandably, if it is determined, based on the current power consumption or predicted temperature, that the temperature of the heat-generating component will decrease after the current iteration cycle, the heat dissipation device can be controlled in advance to reduce its heat dissipation capacity and thus reduce the power consumption waste of the heat dissipation device.

[0043] Furthermore, the target temperature of the heating element is a preset temperature. The ideal temperature of the heating element can be less than or equal to the target temperature. Based on traditional temperature control technology, when the current temperature of the heating element is higher than the target temperature, the heat dissipation device can be controlled to increase its heat dissipation capacity and quickly lower the temperature of the heating element. This reduces the error (i.e., deviation) between the current temperature and the target temperature, ensuring that the temperature of the heating element is less than or equal to the target temperature. When the current temperature of the heating element is lower than the target temperature, the heat dissipation capacity of the heat dissipation device can be reduced or the heat dissipation device can be controlled to stop working, thereby reducing the power consumption waste of the heat dissipation device.

[0044] When a target temperature is not available for the heating component, the current temperature of the heating component in the current iteration cycle and the historical temperature in the previous iteration cycle can be obtained, and the difference between the current temperature and the historical temperature can be used as the temperature change rate of the heating component. When the temperature change rate is large, the heat dissipation device can be controlled to increase its heat dissipation capacity; when the temperature change rate is small, the heat dissipation device can be controlled to decrease its heat dissipation capacity.

[0045] Based on the above description, a first control signal can be generated based on at least one of the current power consumption and the predicted temperature. This first control signal is used to pre-control the heat dissipation device of the heat-generating component, thereby solving the problem of excessively high temperature of the heat-generating component caused by temperature control lag. A second control signal can be generated based on at least one of the target temperature and the current temperature of the heat-generating component. This second control signal is used to control the heat dissipation device of the heat-generating component in real time, thereby adjusting the current temperature error or temperature change rate of the heat-generating component. Superimposing the first and second control signals yields a composite control signal.

[0046] It should be noted that the advance control in this application refers to controlling the heat dissipation device in advance. For example, based on at least one of the current power consumption and predicted temperature, it is found that the temperature of the heat-generating component is high in the next iteration cycle. Normally, the heat dissipation device should be controlled to improve its heat dissipation capacity in the next iteration cycle. However, in this application, the heat dissipation device can be controlled to improve its heat dissipation capacity in the current iteration cycle. In this way, the generation and transmission time of the control signal can be saved when entering the next iteration cycle, so that the heat dissipation device can be controlled in a timely manner. Furthermore, the real-time control in this application is the control logic for the heat dissipation device in traditional temperature control technology. For example, based on the temperature error of the heat-generating component in the current iteration cycle, the heat dissipation device is controlled in the current iteration cycle. Although this control has a lag, it can adjust the current temperature error of the heat-generating component.

[0047] Since each heat-generating component has its own corresponding current power consumption, predicted temperature, target temperature, and current temperature, in step S103, a composite control signal can be generated based on the current power consumption and / or predicted temperature, target temperature, and / or current temperature of each heat-generating component. For example, composite control signal a can be obtained based on the current power consumption and / or predicted temperature, target temperature, and / or current temperature of heat-generating component A. Composite control signal b can be obtained based on the current power consumption and / or predicted temperature, target temperature, and / or current temperature of heat-generating component B.

[0048] Step S104: Output a composite control signal to the heat dissipation device of the heat-generating component to control the heat dissipation device to cool down the heat-generating component.

[0049] It is understandable that, since the composite control signal is composed of the first control signal and the second control signal, it is possible to control the heat dissipation device in advance or in real time. In this way, it can solve the problem of excessively high temperature of the heat-generating component caused by temperature control lag, as well as the problem of current temperature deviation of the heat-generating component.

[0050] Specifically, since each heat-generating component has its own corresponding heat dissipation device and composite control signal, the composite control signal corresponding to each heat-generating component can be output to the corresponding heat dissipation device. For example, the composite control signal 'a' corresponding to heat-generating component A is output to the heat dissipation device 'a1' corresponding to heat-generating component A. The composite control signal 'b' corresponding to heat-generating component B is output to the heat dissipation device 'b1' corresponding to heat-generating component A. In this way, the heat dissipation device can be precisely controlled.

[0051] In summary, in the technical solutions of some embodiments of this application, the predicted temperature of the heating component in the first time period can be obtained based on the current temperature and current power consumption of the heating component in the electronic device. Since the heat dissipation capacity of the heat dissipation device can be improved in advance based on at least one of the current power consumption and predicted temperature of the heating component, and the heat dissipation device can be controlled in real time based on at least one of the target temperature and current temperature of the heating component, the composite control signal obtained based on at least one of the current power consumption and predicted temperature of the heating component, and at least one of the target temperature and current temperature of the heating component, can solve the problem of excessively high heating component temperature caused by temperature control lag in traditional temperature control technology, and can also solve the problem of current temperature error of the heating component.

[0052] Furthermore, for any two heat-generating components, if their thermal resistances are different, even if their current power consumption is the same, their predicted temperatures in the first time period should be different, thus requiring different controls to be applied to the heat dissipation devices corresponding to the two components. However, since the current power consumption of the two heat-generating components is the same, the control signals generated for the heat dissipation devices corresponding to the two components based solely on the current power consumption may be identical, making it impossible to apply different controls to different heat dissipation devices.

[0053] Therefore, in some embodiments, obtaining a composite control signal based on at least one of the current power consumption and the predicted temperature of the first time period, and at least one of the target temperature and the current temperature of the heat-generating component in step S103, may include: Based on the target temperature and the current temperature, the first type of heat dissipation control signal is obtained; Based on the current power consumption, the second type of heat dissipation control signal is obtained; Based on the predicted temperature, a third type of heat dissipation control signal is obtained; The first type of heat dissipation control signal, the second type of heat dissipation control signal, and the third type of heat dissipation control signal are superimposed to obtain a composite control signal.

[0054] Specifically, the controller executing the method of this application is pre-set with a composite PID (Proportional-Integral-Derivative) control algorithm. Based on the target temperature, current temperature, current power consumption, and predicted temperature of the heat-generating component, the composite PID control algorithm is run to obtain the aforementioned first type of heat dissipation control signal, second type of heat dissipation control signal, and third type of heat dissipation control signal.

[0055] For ease of understanding, please refer to the following: Figure 2 This is a schematic diagram of the functional modules of the composite PID control algorithm provided in some embodiments of this application. Figure 2In this context, the composite PID control algorithm includes a feedforward control channel, a feedback control channel, and a predictive control channel.

[0056] The system employs a feedforward control channel to map the current power consumption of the heat-generating component to a second-type heat dissipation control signal, and a predictive control channel to map the predicted temperature of the heat-generating component to a third-type heat dissipation control signal. The second-type heat dissipation control signal can be used for initial advance control of the heat dissipation device, while the third-type signal can be used to correct this initial advance control. Specifically, when two heat-generating components have different thermal resistances but the same current power consumption, their second-type heat dissipation control signals can be the same, while their third-type signals can be different. Based on the second-type heat dissipation control signal, the heat dissipation devices corresponding to the two heat-generating components can be subjected to the same initial advance control. For example, the fans of both heat-generating components can be controlled to rotate at a speed of 500 rpm. Based on the third-type heat dissipation control signal, the initial advance control of the heat dissipation device can be corrected. For example, based on the third-type heat dissipation control signal of the first heat-generating component, the fan speed of the first heat-generating component can be adjusted to 490 rpm, and based on the third-type heat dissipation control signal of the second heat-generating component, the fan speed of the second heat-generating component can be adjusted to 505 rpm. This allows for precise control of the heat dissipation device, ensuring that its heat dissipation capacity is matched to the corresponding heat-generating component.

[0057] Furthermore, in some embodiments, the above-mentioned second type of heat dissipation control signal obtained based on the current power consumption may include: Based on the current power consumption, the fourth heat dissipation control signal is obtained; Based on the historical and current power consumption of the heat-generating component, the power consumption change rate of the heat-generating component is determined, and based on the power consumption change rate, the fifth heat dissipation control signal is obtained. At least one of the fourth and fifth heat dissipation control signals shall be used as the second type of heat dissipation control signal.

[0058] Specifically, the current power consumption is directly proportional to the control strength of the fourth heat dissipation control signal on the heat dissipation device. The higher the current power consumption, the stronger the control strength of the fourth heat dissipation control signal on the heat dissipation device, and the greater the change in the heat dissipation capacity of the heat dissipation device. In this way, the heat dissipation components can be cooled down quickly.

[0059] Furthermore, based on the current power consumption of the heat-generating component and the historical power consumption of the previous iteration cycle, the power consumption difference between the current iteration cycle and the previous iteration cycle can be obtained. Based on the power consumption difference and the time difference between the current iteration cycle and the previous iteration cycle, the power consumption change rate can be obtained, as shown in expression (5).

[0060] (5) in, Indicates the rate of change of power consumption. Indicates the current power consumption. Indicates historical power consumption. This indicates the time difference between the current iteration cycle and the previous iteration cycle.

[0061] The fifth heat dissipation control signal, derived from the power consumption change rate, can be used to correct the fourth heat dissipation control signal, thereby reducing temperature oscillations in the heat-generating component. For example, if the heat dissipation device is controlled in advance using the fourth heat dissipation control signal, it can rapidly reduce the temperature of the heat-generating component. However, if the temperature drops too quickly, it may fall to an excessively low level. In this case, the fifth heat dissipation control signal can correct the fourth. For instance, when the power consumption change rate is small, correcting the fourth heat dissipation control signal based on the fifth can slow down the rate of temperature decrease in the heat-generating component, thus preventing it from falling to an excessively low temperature.

[0062] In the above embodiments, a fourth heat dissipation control signal is obtained based on the current power consumption, and a fifth heat dissipation control signal is obtained based on the power consumption change rate, which can enable precise advance control of the heat dissipation device.

[0063] Specifically, in some embodiments, the fourth heat dissipation control signal can be obtained based on the product between the current power consumption and the preset fourth gain coefficient, as shown in expression (6).

[0064] (6) in, 1 indicates the fourth heat dissipation control signal. This represents the fourth gain coefficient.

[0065] Furthermore, the fifth heat dissipation control signal can be obtained based on the product between the power consumption change rate and the preset fifth gain coefficient, as shown in expression (7).

[0066] (7) in, 2 indicates the fifth heat dissipation control signal. This represents the fifth gain coefficient.

[0067] In some embodiments, the third type of heat dissipation control signal obtained based on the predicted temperature includes: The sixth heat dissipation control signal is obtained based on the temperature difference between the target temperature and the predicted temperature; Based on the current temperature and the predicted temperature, the second temperature change rate of the heating component in the first time period is determined, and based on the second temperature change rate, the seventh heat dissipation control signal is obtained. At least one of the sixth and seventh heat dissipation control signals shall be used as the third type of heat dissipation control signal.

[0068] Specifically, the seventh heat dissipation control signal can be used to correct the sixth heat dissipation control signal. The relevant principle is similar to that of the fifth heat dissipation control signal correcting the fourth heat dissipation control signal, which will not be elaborated here.

[0069] In some embodiments, a sixth heat dissipation control signal can be obtained based on the product of the predicted temperature and a preset sixth gain coefficient, and a seventh heat dissipation control signal can be obtained based on the product of the second temperature change rate and a preset seventh gain coefficient. The above process can be shown in expressions (8) to (12).

[0070] (8) (9) (10) (11) (12) in, To predict temperature, expression (8) indicates calling the temperature prediction function shown in expression (4) above. The predicted temperature of the heating element is obtained.

[0071] The target temperature for the heating element. This represents the difference between the predicted temperature and the target temperature. This is the sixth heat dissipation control signal. This is the sixth gain coefficient. The second rate of temperature change, This is the seventh heat dissipation control signal. This is the seventh gain coefficient.

[0072] At this point, the second type of heat dissipation control signal and the third type of heat dissipation control signal can be obtained.

[0073] Continue reading Figure 2 In some embodiments, the feedback control channel is used to perform the following operations: The difference between the target temperature and the current temperature of the heating component is taken as the current temperature difference, and the first heat dissipation control signal is obtained based on the current temperature difference; Based on the historical and current temperatures of the heating element, a first temperature change rate of the heating element is determined, and a second heat dissipation control signal is obtained based on the first temperature change rate. The difference between the historical temperature of the heating component and the target temperature is taken as the historical temperature difference, and the third heat dissipation control signal is obtained based on the historical temperature difference and the integral of the historical temperature difference. At least one of the first heat dissipation control signal, the second heat dissipation control signal, and the third heat dissipation control signal is used as the first type of heat dissipation control signal.

[0074] Specifically, based on the algorithm logic of the traditional PID control algorithm, the above-mentioned first heat dissipation control signal, second heat dissipation control signal, and third heat dissipation control signal can be obtained.

[0075] The first heat dissipation control signal is used to control the heat dissipation device of the heat-generating component to eliminate the temperature error (i.e., the current temperature difference) of the heat-generating component.

[0076] The second heat dissipation control signal is used to correct the first heat dissipation control signal, which can reduce temperature oscillations of the heat-generating component. For example, after controlling the heat dissipation device using the first heat dissipation control signal, the heat dissipation device can quickly reduce the temperature of the heat-generating component. However, if the temperature of the heat-generating component decreases too quickly, it may drop to an excessively low temperature. In this case, the second heat dissipation control signal, based on the first temperature change rate, can correct the first heat dissipation control signal. For example, when the first temperature change rate is small, after correcting the first heat dissipation control signal based on the second heat dissipation control signal, the heat dissipation device can slow down the rate at which the temperature of the heat-generating component decreases, thereby preventing the temperature of the heat-generating component from dropping to an excessively low temperature.

[0077] The third heat dissipation control signal is used to correct the first heat dissipation control signal, which can reduce the steady-state error of the heat-generating component. Steady-state error refers to the situation where, before the heat-generating component reaches its target temperature, its heating rate equals the heat dissipation rate of the heat dissipation device. In this case, the current temperature of the heat-generating component does not change, and the first heat dissipation control signal also remains unchanged. Consequently, the heat dissipation capacity and heat dissipation rate of the heat dissipation device do not change, and the temperature of the heat-generating component remains stable and never reaches the target temperature. In this situation, the integral of the temperature difference of the heat-generating component gradually increases, causing a change in the third heat dissipation control signal. After the third heat dissipation control signal corrects the first heat dissipation control signal, it can change the heat dissipation capacity and heat dissipation rate of the heat dissipation device, thereby eliminating the steady-state error of the heat-generating component.

[0078] Furthermore, in some embodiments, a first heat dissipation control signal can be obtained based on the product of the current temperature difference and a preset first gain coefficient. A second heat dissipation control signal can be obtained based on the product of the first temperature change rate and a preset second gain coefficient. A third heat dissipation control signal can be obtained based on the product of the integral of the temperature difference and a preset third gain coefficient. The above process can be shown as expressions (13) to (18).

[0079] (13) (14) (15) (16) (17) (18) in, Indicates the current temperature difference. Indicates the current temperature. Indicates the target temperature. Indicates the first rate of temperature change. This indicates the historical temperature from the previous iteration cycle. This represents the time difference between the current iteration cycle and the previous iteration cycle. Represents the integral of temperature difference. This indicates the first heat dissipation control signal. Indicates the first gain coefficient. This indicates the second heat dissipation control signal. This represents the second gain coefficient. This indicates the third heat dissipation control signal. This represents the third gain coefficient.

[0080] In the above embodiments, the feedback control channel of the composite PID control algorithm is used to execute the control logic of the traditional PID control algorithm. Based on the traditional PID control algorithm, this application creatively adds a feedforward control channel and a predictive control channel, which can control the heat dissipation device in advance. In this way, the heat dissipation device can be precisely controlled, overcoming the control lag problem in the traditional PID control algorithm.

[0081] In some embodiments, the first type of heat dissipation control signal, the second type of heat dissipation control signal, and the third type of heat dissipation control signal can be superimposed according to expression (19).

[0082] (19) in, It is a superposition function. The composite control signal is obtained by superposition.

[0083] Furthermore, in some embodiments, the temperature difference integral can be obtained based on the following method before the third heat dissipation control signal is received: The historical temperature difference integral is obtained based on the historical temperature difference between at least one historical temperature of the heating component and the target temperature. If the actual operating parameters of the heat dissipation device are outside the preset operating parameter range, the historical temperature difference integral is subtracted from the current temperature difference to obtain the temperature difference integral. If the actual operating parameters of the heat dissipation device are within the preset operating parameter range, the historical temperature difference integral is added to the current temperature difference to obtain the temperature difference integral.

[0084] Specifically, if the actual operating parameters of the cooling device are outside the preset operating parameter range, it means that the actual operating parameters of the cooling device have exceeded its allowable range. For example, the fan speed exceeds its maximum or minimum supported speed. In this case, the historical temperature difference integral can be subtracted from the current temperature difference to make the temperature difference integral value smaller. This can prevent the third cooling control signal from controlling the actual operating parameters of the cooling device to continue to deviate from the operating parameter range, for example, preventing the third cooling control signal from controlling the fan to rotate at a higher speed. In this way, damage to the cooling device can be prevented.

[0085] Conversely, if the actual operating parameters of the heat dissipation device are within the preset operating parameter range, it means that the actual operating parameters of the heat dissipation device have not exceeded its allowable range. For example, the fan speed has not exceeded its maximum or minimum supported speed. In this case, the historical temperature difference integral can be added to the current temperature difference to make the temperature difference integral value larger, and the heat dissipation device can be controlled to continue to increase its heat dissipation capacity through the third heat dissipation control signal.

[0086] For example, the specific code implementation of the above process is as follows: if (output>output_max) { output = output_max; if (error>0) integral_sum -= error * delta_t; } else if (output <output_min) { output = output_min; if (error<0) integral_sum -= error * delta_t; } Wherein, output_max represents the maximum allowable amplitude of the composite control signal when the actual operating parameters of the heat dissipation device do not exceed the operating parameter range, and output_min represents the minimum allowable amplitude of the composite control signal when the actual operating parameters of the heat dissipation device do not exceed the operating parameter range.

[0087] The above code means: after the composite control signal is output to the heat dissipation device, it determines whether the actual operating parameters of the heat dissipation device exceed the preset operating parameter range. If the actual operating parameters exceed the operating parameter range, the composite control signal is adjusted, and the adjusted composite control signal is output to the heat dissipation device to ensure that the actual operating parameters of the heat dissipation device are within the operating parameter range, i.e., output = output_max and output = output_min. Simultaneously, the historical temperature difference integral is subtracted from the current temperature difference, i.e., integral_sum -= error * delta_t. This prevents damage to the heat dissipation device.

[0088] In the above embodiments, the calculation method of the temperature difference integral is determined based on the actual operating parameters of the heat dissipation device, which can effectively protect the heat dissipation device, avoid problems such as damage to the heat dissipation device, and improve the reliability of the solution.

[0089] In some embodiments, when the integral of the temperature difference is large, even if the temperature of the heating element has dropped to or below the target temperature, the third heat dissipation control signal obtained based on the integral of the temperature difference will still control the heat dissipation device to operate with a high heat dissipation capacity. For example, even when the temperature of the heating element has dropped to or below the target temperature, the third heat dissipation control signal will still control the fan to rotate at a high speed, which will result in wasted power consumption of the heat dissipation device.

[0090] Therefore, the third heat dissipation control signal, obtained based on the historical temperature difference and the integral of the historical temperature difference, includes: When the integral of temperature difference is outside the preset integration range, the integral of temperature difference is corrected so that the corrected integral of temperature difference is within the integration range. The third heat dissipation control signal is obtained based on the corrected temperature difference integral.

[0091] Specifically, the preset integration interval is used to characterize the reasonable range of the temperature difference integral. If the temperature difference integral is outside the preset integration interval, it means that the temperature difference integral is too large, and the temperature difference integral can be corrected so that the corrected temperature difference integral is within the integration interval.

[0092] For example, the specific code implementation of the above process is as follows: if (integral_sum>integral_max) { integral_sum = integral_max; } else if (integral_sum <integral_min) { integral_sum = integral_min; } Wherein, integral_max represents the maximum value of the integration interval, integral_min represents the minimum value of the integration interval, and integral_sum represents the integral of temperature difference.

[0093] In this way, the problem of the third heat dissipation control signal still controlling the heat dissipation device to operate with a high heat dissipation capacity can be avoided when the temperature of the heat-generating component has dropped to or below the target temperature.

[0094] In some embodiments, the gain coefficients described above can be adjusted based on at least one of the power consumption distribution characteristics of multiple heat-generating components in the electronic device and the operating characteristics of the heat dissipation device. The power consumption distribution characteristics characterize the power consumption distribution or rate of change among the heat-generating components; for example, the image processor consumes 80 watts, and the central processing unit consumes 30 watts. The operating characteristics of the heat dissipation device refer to the actual operating parameter range of the heat dissipation device.

[0095] Specifically, based on power consumption distribution characteristics, it is possible to predict future temperature ranges or temperature change trends in electronic devices and dynamically adjust at least some gain coefficients. For example, based on power consumption distribution characteristics, if the image processor's power consumption is found to be relatively high, it means the electronic device is operating under high load, and the future temperature will be relatively high and the temperature change rate will be relatively fast. In this case, the first, second, fourth, fifth, sixth, and seventh gain coefficients can be increased. This allows for a rapid response when the temperature of the heat-generating component rises, controlling the heat dissipation component to cool the heat-generating component in a timely manner. At the same time, the third gain coefficient can be decreased to prevent temperature oscillations in the heat-generating component during temperature control adjustment when the first and second gain coefficients are too large.

[0096] Based on the operating characteristics of the heat dissipation device, the adjustable range of its operating parameters can be determined, and at least some gain coefficients can be dynamically adjusted based on this range. For example, when the actual fan speed is much lower than its maximum allowable speed, it indicates that the actual fan speed has considerable room for increase. In this case, the first, second, fourth, and sixth gain coefficients can be increased. This allows for a rapid increase in fan speed and a rapid decrease in the temperature of the heat-generating components. Conversely, when the actual fan speed is close to its maximum allowable speed, it indicates that the actual fan speed has limited room for increase. In this case, the first, second, third, fourth, fifth, sixth, and seventh gain coefficients can be decreased. This slows down the rate of increase in fan speed, thereby reducing the risk of fan damage.

[0097] In the above embodiments, by dynamically adjusting at least part of the gain coefficient based on at least one of the power consumption distribution characteristics and the operating characteristics of the heat dissipation device, the cooling efficiency of the heat-generating component can be improved, while reducing the risk of damage to the heat-generating component.

[0098] Based on the above description, in some embodiments, adjusting at least a portion of the gain coefficient based on the operating characteristics of the heat dissipation device includes: If the difference between the target temperature and the current temperature is greater than the first temperature difference threshold, and the actual operating parameters of the heat dissipation device are less than the first operating parameter threshold, then at least part of the gain coefficient is increased. If the difference between the target temperature and the current temperature is less than the second temperature difference threshold, and the actual operating parameters of the heat dissipation device are greater than the second operating parameter threshold, then at least part of the gain coefficient is reduced. Wherein, the first temperature difference threshold is less than or equal to the second temperature difference threshold, and the first operating parameter threshold is less than or equal to the second operating parameter threshold.

[0099] Specifically, when the difference between the target temperature and the current temperature is greater than the first temperature difference threshold, and the actual operating parameters of the heat dissipation device are less than the first operating parameter threshold, it indicates that the temperature error of the heat-generating component is large (i.e., the temperature adjustment requirement is large), and the actual operating parameters of the heat dissipation device still have a large adjustment space. At this time, by increasing at least part of the gain coefficient, the adjustment rate of the actual operating parameters of the heat dissipation device can be increased, thereby quickly reducing the temperature error of the heat-generating component.

[0100] Conversely, when the difference between the target temperature and the current temperature is less than the second temperature difference threshold, and the actual operating parameters of the heat dissipation device are greater than the second operating parameter threshold, it indicates that the temperature error of the heat-generating component is small (i.e., the temperature adjustment requirement is small), and the adjustment space of the actual operating parameters of the heat dissipation device is small. In this case, by reducing at least part of the gain coefficient, the adjustment rate of the actual operating parameters of the heat dissipation device can be reduced, thereby reducing the risk of damage to the heat dissipation device.

[0101] For example, the specific code implementation of the above process is as follows: if (error>0&&output <output_max * 0.8f) { pid->Kp_temp *= (1.0f + learning_rate); } else if (fabsf(error)<2.0f&&output_max * 0.9f) { pid->Kp_temp *= (1.0f - learning_rate); } The meanings of the parameters error, output, and output_max can be found in the relevant descriptions above, and will not be repeated here. learning_rate represents the single adjustment range of the gain coefficient.

[0102] In some embodiments, adjusting at least a portion of the gain coefficient based on the power consumption distribution characteristics of the heat-generating component includes: Based on the power consumption distribution characteristics of heat-generating components, the load category of electronic devices is determined; Adjust at least some of the gain coefficients based on the load category.

[0103] Specifically, load categories can include, but are not limited to, regular loads, CPU compute-intensive loads, memory compute-intensive loads, GPU compute-intensive loads, and I / O intensive loads. For example, if the power consumption of the GPU is determined to be high based on the power consumption distribution characteristics of heat-generating components, the load category can be classified as a GPU compute-intensive load.

[0104] Multiple gain coefficient groups can be pre-set for different load categories. Each load category corresponds one-to-one with a gain coefficient group. Each gain coefficient group includes values ​​for a first, second, third, fourth, fifth, sixth, and seventh gain coefficient. The gain coefficients in different groups can be different. For example, in gain coefficient group A, the first gain coefficient = 0.8, the second gain coefficient = 0.6, the third gain coefficient = 0.9, the fourth gain coefficient = 0.7, the fifth gain coefficient = 0.75, the sixth gain coefficient = 0.62, and the seventh gain coefficient = 0.85. In gain coefficient group B, the first gain coefficient = 0.6, the second gain coefficient = 0.6, the third gain coefficient = 0.83, the fourth gain coefficient = 0.52, the fifth gain coefficient = 0.23, the sixth gain coefficient = 0.56, and the seventh gain coefficient = 0.35.

[0105] After determining the load category of an electronic device based on the power consumption distribution characteristics of the heat-generating components, the gain coefficient can be adjusted based on the gain coefficient in the gain coefficient group corresponding to the load category.

[0106] For example, the specific code implementation of the above process is as follows: new_workload = classify_workload(current_power, power_derivative); if (new_workload != adaptive->current_workload) { adaptive->current_workload = new_workload; log_workload_transition(adaptive->current_workload); } Wherein, new_workload = classify_workload(current_power, power_derivative): represents the calculation of a new load class new_workload based on the current power consumption current_power and the power consumption change rate power_derivative of each heat-generating component.

[0107] new_workload != adaptive->current_workload): This indicates that the current workload class current_workload is different from the new workload class new_workload.

[0108] adaptive->current_workload = new_workload: This means updating the current workload category current_workload to the new workload category new_workload.

[0109] log_workload_transition(adaptive->current_workload): indicates that the workload type switching log is recorded.

[0110] This application also provides an electronic device adapted to the above-described temperature control method. (See also...) Figure 3 The diagram below is a schematic diagram of the structure of an electronic device provided in some embodiments of this application. Figure 3 The electronic device includes at least one heat-generating component, a heat dissipation device, a data acquisition circuit, and a main controller. The data acquisition circuit includes a temperature acquisition circuit and a power consumption acquisition circuit. The temperature acquisition circuit acquires the current temperature of at least one heat-generating component, and the power consumption acquisition circuit acquires the current power consumption of at least one heat-generating component. The main controller receives the current power consumption and current temperature of at least one heat-generating component from the acquisition circuit and, based on the aforementioned temperature control method, controls the heat dissipation device to cool the heat-generating component.

[0111] Specifically, such as Figure 3 As shown, the main controller in the electronic device can be a baseboard management controller, and the electronic device includes a motherboard module, an acceleration module, a power supply module, a hard disk module, and an expansion module. Different modules can be set on different circuit boards, or multiple modules can be set on the same circuit board; this application does not impose any restrictions on this.

[0112] Each module may include one or more heat-generating components. Specifically, heat-generating components in the motherboard module may include, but are not limited to, a central processing unit (CPU) and dual in-line memory modules (DIMMs). Heat-generating components in the acceleration module may include, but are not limited to, a graphics processor (GPU), a PCIe (Peripheral Component Interconnect express) switch, and complex programmable logic devices (CPLs). Heat-generating components in the power supply module may include, but are not limited to, a power supply unit (PSU) and power supply integrated circuits. Heat-generating components in the hard drive module may include, but are not limited to, a PCIe switch and a hard drive. Heat-generating components in the expansion module may include, but are not limited to, adapter cards, network interface cards (NICs), and RAID (Redundant Array of Independent Disks) cards. It should be noted that, in addition to the modules mentioned above, electronic devices may also include other modules (not shown). Heat-generating components in other modules may include, but are not limited to, integrated circuit modules.

[0113] Each module may include a temperature acquisition circuit and a power consumption acquisition circuit, and at least the module includes a module controller. For any given module, if the module does not include a module controller, the temperature acquisition circuit and the power consumption acquisition circuit of the module are respectively connected to the main controller; if the module includes a module controller, the temperature acquisition circuit and the power consumption acquisition circuit of the module are connected to the main controller through the module controller.

[0114] For any module's temperature acquisition circuit and heating element, the temperature acquisition circuit includes multiple temperature acquisition channels. At least some temperature acquisition channels correspond one-to-one with heating elements, with each channel acquiring the current temperature of its corresponding heating element. Alternatively, when multiple heating elements are distributed across at least two areas of the electronic device, at least some temperature acquisition channels correspond one-to-one with each area, with each channel acquiring the current temperature of each heating element within its corresponding area. Or, at least some temperature acquisition channels correspond one-to-one with different sub-heating elements within the same heating element, with each channel acquiring the current temperature of its corresponding sub-heating element. For example, in a network interface card (NIC) comprising a main body and an optical module, one temperature acquisition channel can be connected to the main body to acquire the temperature of the main body, and another temperature acquisition channel can be connected to the optical module to acquire the temperature of the optical module. By setting multiple temperature acquisition channels, temperature acquisition efficiency can be improved.

[0115] Similarly, for any module's power consumption acquisition circuit and heat-generating components, the power consumption acquisition circuit includes multiple power consumption acquisition channels. When the module includes multiple power supply lines, at least some power consumption acquisition channels correspond one-to-one with the power supply lines, and each power consumption acquisition channel is used to acquire the current power consumption of at least one heat-generating component in the corresponding power supply line. Alternatively, at least some power consumption acquisition channels correspond one-to-one with different power supply pins of the same heat-generating component, and each power consumption acquisition channel is used to acquire the current power consumption of the corresponding power supply pin. Or, at least some power consumption acquisition channels correspond one-to-one with the heat-generating components, and each power consumption acquisition channel is used to acquire the current power consumption of the corresponding heat-generating component. By setting multiple power consumption acquisition channels, power consumption acquisition efficiency can be improved.

[0116] Specifically, further, in some embodiments, the electronic device includes a power supply integrated circuit, which serves as a power consumption acquisition circuit. The power supply integrated circuit includes multiple power supply ports, each corresponding to a power consumption acquisition channel. Specifically, each power supply port of the power supply integrated circuit can be considered a power supply line; acquiring the power consumption of each power supply port is equivalent to acquiring the power consumption of the corresponding power supply line.

[0117] In some embodiments, the power consumption acquisition circuit includes multiple power consumption sampling circuits, each corresponding to a power consumption acquisition channel. Specifically, the power consumption sampling circuit may include electronic components such as a sampling resistor. The resistance value of the sampling resistor is known, and the sampling resistor can be connected in series with the heating component, so that the sampling resistor and the heating component have the same current. By acquiring the voltage across the sampling resistor and based on its resistance value, the current of the sampling resistor can be calculated, and thus the current of the heating component can be obtained. Then, based on the voltage division relationship between the sampling resistor and the heating component, the voltage of the heating component can be calculated. Based on the voltage and current of the heating component, the power consumption of the heating component can be calculated.

[0118] Thus, the temperature of multiple heating components can be collected in parallel through the temperature acquisition circuit, and the power consumption of multiple heating components can be collected in parallel through the power consumption acquisition circuit.

[0119] Furthermore, for any module, if the module includes a module controller, the module controller can acquire the temperature and power consumption data collected by the module's temperature acquisition circuit and power consumption acquisition circuit, preprocess the temperature and power consumption data (e.g., remove abnormal data), and send the preprocessed temperature and power consumption data to the main controller. If the module does not include a module controller, the module's temperature acquisition circuit and power consumption acquisition circuit directly report the acquired temperature and power consumption data to the main controller, which then performs data preprocessing.

[0120] Furthermore, for any module, if it contains heat-generating components capable of collecting its own temperature and power consumption data, these components can be directly connected to the main controller without being connected to the power consumption and temperature acquisition circuits. They then report their temperature and power consumption data to the main controller. For example, the central processing unit (CPU) may have internal temperature and power consumption sensors. Based on these sensors, the CPU can collect its own temperature and power consumption data and report it to the main controller. In this case, the CPU can be directly connected to the main controller without being connected to the motherboard module's temperature and power consumption acquisition circuits.

[0121] After receiving temperature and power consumption data from each heat-generating component, the main controller can generate composite control signals for at least some of the heat dissipation devices based on the aforementioned temperature control method, and output these composite control signals to the corresponding heat dissipation devices to control them to cool the heat-generating components. Specifically, for example... Figure 3 As shown, the electronic device may include a thermal controller and multiple heat dissipation devices. The thermal controller is connected to a main controller and includes multiple thermal control channels. Different thermal control channels are connected to different heat dissipation devices. The main controller can send composite control signals for multiple heat dissipation devices to the thermal controller, which then controls the multiple heat dissipation devices in parallel through the multiple thermal control channels, thereby improving control efficiency.

[0122] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods according to the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method.

[0123] See also Figure 4 Embodiments of this application also provide an electronic device, including a memory 10 and a processor 20, wherein the memory 10 stores a computer program and the processor 20 is configured to run the computer program to perform the steps in any of the above-described communication method embodiments.

[0124] Embodiments of this application also provide a computer-readable storage medium storing a computer program, wherein the computer program is configured to execute the steps in any of the above-described communication method embodiments when it is run.

[0125] In one exemplary embodiment, the aforementioned computer-readable storage medium may include, but is not limited to, various media capable of storing computer programs, such as a USB flash drive, read-only memory (ROM), random access memory (RAM), portable hard disk, magnetic disk, or optical disk.

[0126] Embodiments of this application also provide a computer program product, which includes a computer program that, when executed by a processor, implements the steps in any of the above-described communication method embodiments.

[0127] Embodiments of this application also provide another computer program product, including a non-volatile computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps in any of the above-described communication method embodiments.

[0128] Those skilled in the art will further 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, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. 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 implementation should not be considered beyond the scope of this application.

[0129] The above provides a detailed description of a temperature control method, a temperature control system, and a storage system provided in this application. Specific examples have been used to illustrate the principles and implementation methods of this application. The descriptions of the embodiments above are merely for the purpose of helping to understand the method and its core ideas. It should be noted that those skilled in the art can make various improvements and modifications to this application without departing from its principles, and these improvements and modifications also fall within the protection scope of the claims of this application.

Claims

1. A temperature control method, characterized in that, include: Collect the current temperature and current power consumption of at least one heat-generating component in the electronic device; Based on the current temperature and the current power consumption, the predicted temperature of the heating component in the first time period is obtained; A composite control signal is obtained based on at least one of the current power consumption and the predicted temperature, and at least one of the target temperature of the heating component and the current temperature; The composite control signal is output to the heat dissipation device of the heat-generating component to control the heat dissipation device to cool down the heat-generating component.

2. The method according to claim 1, characterized in that, The composite control signal is obtained based on at least one of the current power consumption and the predicted temperature of the first time period, and at least one of the target temperature of the heating component and the current temperature, including: Based on the target temperature and the current temperature, a first type of heat dissipation control signal is obtained; Based on the current power consumption, a second type of heat dissipation control signal is obtained; Based on the predicted temperature, a third type of heat dissipation control signal is obtained; By superimposing the first type of heat dissipation control signal, the second type of heat dissipation control signal, and the third type of heat dissipation control signal, a composite control signal is obtained.

3. The method according to claim 2, characterized in that, The process of obtaining a first type of heat dissipation control signal based on the target temperature and the current temperature includes: The difference between the target temperature and the current temperature is taken as the current temperature difference, and a first heat dissipation control signal is obtained based on the current temperature difference; Based on the historical temperature and the current temperature of the heating component, a first temperature change rate of the heating component is determined, and based on the first temperature change rate, a second heat dissipation control signal is obtained. The difference between the historical temperature of the heating component and the target temperature is taken as the historical temperature difference, and the third heat dissipation control signal is obtained based on the temperature difference integral of the historical temperature difference and the current temperature difference. At least one of the first heat dissipation control signal, the second heat dissipation control signal, and the third heat dissipation control signal shall be used as the first type of heat dissipation control signal.

4. The method according to claim 3, characterized in that, Before receiving the third heat dissipation control signal, the method further includes: Based on the historical temperature difference between at least one historical temperature of the heating component and the target temperature, the historical temperature difference integral is obtained; If the actual operating parameters of the heat dissipation device are outside the preset operating parameter range, the historical temperature difference integral is subtracted from the current temperature difference to obtain the temperature difference integral. If the actual operating parameters of the heat dissipation device are within the preset operating parameter range, the historical temperature difference integral is added to the current temperature difference to obtain the temperature difference integral.

5. The method according to claim 3, characterized in that, The third heat dissipation control signal is obtained by integrating the temperature difference based on the historical temperature difference and the current temperature difference, including: When the integral of the temperature difference is outside the preset integration range, the integral of the temperature difference is corrected so that the corrected integral of the temperature difference is within the integration range. The third heat dissipation control signal is obtained based on the corrected integral of the temperature difference.

6. The method according to claim 3, characterized in that, Obtaining the first heat dissipation control signal, the second heat dissipation control signal, and the third heat dissipation control signal includes: The first heat dissipation control signal is obtained based on the product of the current temperature difference and the preset first gain coefficient; And / or, the second heat dissipation control signal is obtained based on the product between the first temperature change rate and the preset second gain coefficient; And / or, the third heat dissipation control signal is obtained based on the product between the integral of the temperature difference and the preset third gain coefficient.

7. The method according to claim 3, characterized in that, The second type of heat dissipation control signal obtained based on the current power consumption includes: Based on the current power consumption, a fourth heat dissipation control signal is obtained; Based on the historical power consumption and the current power consumption of the heat-generating component, the power consumption change rate of the heat-generating component is determined, and based on the power consumption change rate, a fifth heat dissipation control signal is obtained. At least one of the fourth heat dissipation control signal and the fifth heat dissipation control signal shall be used as the second type of heat dissipation control signal.

8. The method according to claim 7, characterized in that, Obtaining the fourth heat dissipation control signal and the fifth heat dissipation control signal includes: The fourth heat dissipation control signal is obtained based on the product between the current power consumption and the preset fourth gain coefficient; And / or, the fifth heat dissipation control signal is obtained based on the product between the power consumption change rate and the preset fifth gain coefficient.

9. The method according to claim 2, characterized in that, The third type of heat dissipation control signal obtained based on the predicted temperature includes: A sixth heat dissipation control signal is obtained based on the temperature difference between the target temperature and the predicted temperature; Based on the current temperature and the predicted temperature, a second temperature change rate of the heating component during the first time period is determined, and a seventh heat dissipation control signal is obtained based on the second temperature change rate. At least one of the sixth heat dissipation control signal and the seventh heat dissipation control signal shall be used as the third type of heat dissipation control signal.

10. The method according to claim 9, characterized in that, Obtaining the sixth heat dissipation control signal and the seventh heat dissipation control signal includes: The sixth heat dissipation control signal is obtained based on the product between the predicted temperature and the preset sixth gain coefficient. And / or, the seventh heat dissipation control signal is obtained based on the product between the second temperature change rate and the preset seventh gain coefficient.

11. The method according to any one of claims 6, 8, and 10, characterized in that, The method further includes: Based on at least one of the power consumption distribution characteristics of multiple heat-generating components in the electronic device and the operating characteristics of the heat dissipation device, at least a portion of the gain coefficient is adjusted.

12. The method according to claim 11, characterized in that, Based on the operating characteristics of the heat dissipation device, at least part of the gain coefficient is adjusted, including: If the difference between the target temperature and the current temperature is greater than a first temperature difference threshold, and the actual operating parameters of the heat dissipation device are less than a first operating parameter threshold, then at least part of the gain coefficient is increased. If the difference between the target temperature and the current temperature is less than the second temperature difference threshold, and the actual operating parameters of the heat dissipation device are greater than the second operating parameter threshold, then at least part of the gain coefficient is reduced. Wherein, the first temperature difference threshold is less than or equal to the second temperature difference threshold, and the first operating parameter threshold is less than or equal to the second operating parameter threshold.

13. The method according to claim 11, characterized in that, Based on the power consumption distribution characteristics of the heat-generating component, at least some gain coefficients are adjusted, including: Based on the power consumption distribution characteristics of the heat-generating components, the load category of the electronic device is determined; Based on the load category, adjust at least part of the gain coefficient.

14. The method according to claim 1 or 2, characterized in that, Before outputting the composite control signal to the heat dissipation device of the heat-generating component, the method includes: After the composite control signal is output to the heat dissipation device, determine whether the actual operating parameters of the heat dissipation device exceed the preset operating parameter range; If the actual operating parameters exceed the operating parameter range, the composite control signal is adjusted, and the adjusted composite control signal is output to the heat dissipation device so that the actual operating parameters of the heat dissipation device are within the operating parameter range.

15. The method according to claim 1 or 2, characterized in that, The step of predicting the temperature of the heat-generating component in the first time period based on the current temperature and the current power consumption includes: Obtain the ambient temperature of the environment where the heating component is located and the junction-to-environment thermal resistance of the heating component; The steady-state temperature of the heat-generating component is determined based on the ambient temperature, the current power consumption, and the junction-to-ambient thermal resistance. Based on the current temperature, the steady-state temperature, and the difference between the current temperature, the temperature of the heating component during the first time period is predicted.

16. An electronic device, characterized in that, The electronic device includes: At least one heating element; Heat dissipation device; The acquisition circuit includes a temperature acquisition circuit and a power consumption acquisition circuit. The temperature acquisition circuit is used to acquire the current temperature of the at least one heating component, and the power consumption acquisition circuit is used to acquire the current power consumption of the at least one heating component. A main controller is configured to receive the current power consumption and current temperature of the at least one heat-generating component sent by the acquisition circuit, and control the heat dissipation device to cool down the heat-generating component based on the method described in any one of claims 1 to 15.

17. The electronic device according to claim 16, characterized in that, The temperature acquisition circuit includes multiple temperature acquisition channels; At least some of the temperature acquisition channels correspond one-to-one with the heating components, and each temperature acquisition channel is used to acquire the current temperature of the corresponding heating component; And / or, when multiple heat-generating components are distributed in at least two regions of the electronic device, at least some of the temperature acquisition channels correspond one-to-one with the regions, and each of the temperature acquisition channels is used to acquire the current temperature of each heat-generating component in the corresponding region; And / or, at least some of the temperature acquisition channels correspond one-to-one with different sub-heating components in the same heating component, and each temperature acquisition channel is used to acquire the current temperature of the corresponding sub-heating component.

18. The electronic device according to claim 16, characterized in that, The power consumption acquisition circuit includes multiple power consumption acquisition channels; When the electronic device includes multiple power supply lines, at least some of the power consumption acquisition channels correspond one-to-one with the power supply lines, and each power consumption acquisition channel is used to acquire the current power consumption of at least one heat-generating component in the corresponding power supply line; And / or, at least some of the power consumption acquisition channels correspond one-to-one with different power supply pins of the same heat-generating component, and each power consumption acquisition channel is used to acquire the current power consumption of the corresponding power supply pin; And / or, at least some of the power consumption acquisition channels correspond one-to-one with the heat-generating components, and each power consumption acquisition channel is used to acquire the current power consumption of the corresponding heat-generating component.

19. The electronic device according to claim 18, characterized in that, The electronic device includes a power supply integrated circuit, which serves as the power consumption acquisition circuit. The power supply integrated circuit includes multiple power supply ports, and each power supply port corresponds to a power consumption acquisition channel. And / or, the power consumption acquisition circuit includes multiple power consumption sampling circuits, each corresponding to a power consumption acquisition channel.

20. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, wherein the computer program, when executed by a processor, implements the temperature control method as described in any one of claims 1 to 15.

Images