Heat dissipation control method and device of equipment, all-in-one machine equipment, equipment and medium

Abstract

The invention relates to the technical field of equipment heat dissipation, and discloses a heat dissipation control method and device of equipment, all-in-one machine equipment, equipment and a medium. Firstly, for a target hardware module in at least one hardware module, an evaluation temperature and a working frequency corresponding to the target hardware module are obtained, and the target hardware module corresponds to slot position information and type information representing the position of the target hardware module in equipment. Then, based on the type information of the target hardware module, determining a target mapping relation corresponding to the type information and representing the relation among the evaluation temperature, the working frequency and the control parameters corresponding to the target hardware module; then, based on the target mapping relation, according to the evaluation temperature and the working frequency, target control parameters are determined; and finally, based on the slot position information of the target hardware module, determining a heat dissipation component corresponding to the target hardware module, and based on the target control parameter, controlling the heat dissipation component to dissipate heat of the target hardware module.

Description

Technical Field

[0001] The present invention relates to the technical field of equipment heat dissipation, and in particular to a heat dissipation control method, device, all-in-one equipment, equipment and medium for equipment. Background Art

[0002] With the rapid development of technology, the performance and computing power of smart all-in-one products are getting higher and higher, and higher requirements are also placed on their heat dissipation design.

[0003] In the related art, a fan speed table corresponding to different temperatures is used to control the fan. However, the method using the fan speed table is not universal. Therefore, a new heat dissipation control method for equipment needs to be proposed. Summary of the invention

[0004] The embodiments of this specification are intended to solve at least one of the technical problems in the related art to a certain extent. To this end, the embodiments of this specification propose a heat dissipation control method, device, all-in-one device, equipment and medium for equipment.

[0005] The embodiment of this specification provides a method for controlling heat dissipation of a device, wherein the device includes at least one hardware module and a heat dissipation component corresponding to each hardware module, and the method includes:

[0006] For a target hardware module in at least one hardware module, obtaining an evaluation temperature and an operating frequency corresponding to the target hardware module; wherein the target hardware module corresponds to slot information and type information, and the slot information represents a position of the target hardware module in the device;

[0007] Based on the type information of the target hardware module, determine a target mapping relationship corresponding to the type information; wherein the target mapping relationship represents a relationship between the evaluation temperature, the operating frequency and the control parameter corresponding to the target hardware module;

[0008] Based on the target mapping relationship, determining a target control parameter according to the evaluation temperature and the operating frequency;

[0009] Based on the slot information of the target hardware module, a heat dissipation component corresponding to the target hardware module is determined, and based on the target control parameter, the heat dissipation component is controlled to dissipate heat for the target hardware module.

[0010] In one embodiment, the target hardware module includes a core chip, and obtaining the evaluation temperature corresponding to the target hardware module includes:

[0011] Acquire a temperature parameter list corresponding to the target hardware module; wherein the temperature parameter list includes the core temperature of the core chip and at least one external temperature of the target hardware module;

[0012] Acquire a temperature weight list, wherein the temperature weight list is associated with a temperature parameter list corresponding to the target hardware module;

[0013] The temperature parameter list is processed using the temperature weight list to obtain the evaluation temperature.

[0014] In one embodiment, the target mapping relationship includes a temperature control prediction model; and determining the target control parameter based on the target mapping relationship according to the evaluation temperature and the operating frequency includes:

[0015] The temperature control prediction model is used to predict the evaluation temperature and the operating frequency to obtain target control parameters.

[0016] In one implementation, the target mapping relationship is obtained by training in the following manner:

[0017] Based on the initial mapping relationship, determining a first prediction control parameter according to the evaluation sample temperature and the sample frequency;

[0018] Determining a first evaluation parameter based on the first predicted control parameter and the expected control parameter;

[0019] The target mapping relationship is determined based on a comparison result between the first evaluation parameter and a preset threshold parameter.

[0020] In one embodiment, the determining the target mapping relationship based on a comparison result between the first evaluation parameter and a preset threshold parameter includes:

[0021] When the first evaluation parameter is less than the preset threshold parameter, the initial mapping relationship is determined to be the target mapping relationship.

[0022] In one embodiment, the determining the target mapping relationship based on a comparison result between the first evaluation parameter and a preset threshold parameter includes:

[0023] When the first evaluation parameter is greater than or equal to the preset threshold parameter, updating the target parameter of the initial mapping relationship based on the learning efficiency, and determining the updated mapping relationship; wherein the target parameter is used to determine the mapping relationship;

[0024] Based on the updated mapping relationship, determining a second prediction control parameter according to the evaluation sample temperature and the sample frequency;

[0025] determining a second evaluation parameter based on the second predicted control parameter and the expected control parameter;

[0026] The target mapping relationship is determined based on a comparison result between the second evaluation parameter and a preset threshold parameter.

[0027] In one embodiment, the initial mapping relationship includes an input layer, a hidden layer, and an output layer, the input layer includes a first node and a second node, the hidden layer includes a third node and a fourth node, and the target parameter of the initial mapping relationship includes a first weight, a second weight, a third weight, and a fourth weight;

[0028] The first weight is the weight between the first node and the third node;

[0029] The second weight is the weight between the first node and the fourth node;

[0030] The third weight is the weight between the second node and the third node;

[0031] The fourth weight is a weight between the second node and the fourth node.

[0032] In one embodiment, the determining the first prediction control parameter based on the initial mapping relationship and according to the evaluation sample temperature and the sample frequency includes:

[0033] Taking the sample temperature as the value of the first node and taking the sample frequency as the value of the second node;

[0034] Based on the first weight, the second weight, the third weight, and the fourth weight, the values ​​of the first node and the second node are processed to obtain the value of the third node and the value of the fourth node;

[0035] Based on the value of the third node and the value of the fourth node, the first prediction control parameter is obtained, and the first prediction control parameter is output through the output layer.

[0036] In one embodiment, the target parameter of the initial mapping relationship further includes a first bias parameter corresponding to the third node and a second bias parameter corresponding to the fourth node; and the processing of the values ​​of the first node and the second node based on the first weight, the second weight, the third weight, and the fourth weight to obtain the value of the third node and the value of the fourth node includes:

[0037] Based on the first weight, the third weight and the first bias parameter, the values ​​of the first node and the second node are processed to obtain the value of the third node;

[0038] Based on the second weight, the fourth weight and the second bias parameter, the values ​​of the first node and the second node are processed to obtain the value of the fourth node.

[0039] In one embodiment, the expected control parameter is determined by:

[0040] The expected control parameters are determined by performing simulation based on the sample temperature list, the type information of the sample module, and the sample frequency.

[0041] The present specification provides an all-in-one device, the device comprising:

[0042] at least one hardware module;

[0043] at least one heat dissipation component, the at least one heat dissipation component corresponding to the at least one hardware module in a one-to-one manner;

[0044] A heat dissipation control module is used to implement the method according to any of the above embodiments.

[0045] The embodiment of this specification provides a heat dissipation control device for a device, wherein the device includes at least one hardware module and a heat dissipation component corresponding to each hardware module, and the device includes:

[0046] A data acquisition module, configured to acquire, for a target hardware module in at least one hardware module, an evaluation temperature and an operating frequency corresponding to the target hardware module; wherein the target hardware module corresponds to slot information and type information, and the slot information represents a position of the target hardware module in the device;

[0047] A mapping relationship determination module, configured to determine a target mapping relationship corresponding to the type information based on the type information of the target hardware module; wherein the target mapping relationship represents a relationship between the evaluation temperature, the operating frequency and a control parameter corresponding to the target hardware module;

[0048] A control parameter determination module, configured to determine a target control parameter based on the target mapping relationship, according to the evaluation temperature and the operating frequency;

[0049] The heat dissipation component control module is used to determine the heat dissipation component corresponding to the target hardware module based on the slot information of the target hardware module, and control the heat dissipation component to dissipate heat for the target hardware module based on the target control parameter.

[0050] An embodiment of the present specification provides a computer device, a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the steps of the method described in any of the above embodiments are implemented.

[0051] The embodiments of this specification provide a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the steps of the method described in any of the above embodiments are implemented.

[0052] An embodiment of the present specification provides a computer program product, wherein the computer program product includes instructions, and when the instructions are executed by a processor of a computer device, the computer device is enabled to perform the steps of the method described in any one of the above embodiments.

[0053] In the above-mentioned implementation manner of the specification, the device includes at least one hardware module and a heat dissipation component corresponding to each hardware module. First, for the target hardware module in at least one hardware module, the evaluation temperature and operating frequency corresponding to the target hardware module are obtained, and the target hardware module corresponds to the slot information and type information representing the position of the target hardware module in the device. Next, based on the type information of the target hardware module, the target mapping relationship corresponding to the type information representing the relationship between the evaluation temperature, the operating frequency and the control parameter corresponding to the target hardware module is determined. Then, based on the target mapping relationship, the target control parameter is determined according to the evaluation temperature and the operating frequency. Finally, based on the slot information of the target hardware module, the heat dissipation component corresponding to the target hardware module is determined, and the heat dissipation component is controlled based on the target control parameter to dissipate heat for the target hardware module. By evaluating the multi-data fusion method of temperature and operating frequency, the target control parameter for controlling the heat dissipation component of the target hardware module is more accurately determined, thereby improving the independent control accuracy of the heat dissipation component and reducing noise and energy consumption. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Figure 1a A schematic diagram of the structure of an all-in-one device provided for the implementation of this specification;

[0055] Figure 1b A flow chart of a heat dissipation control method for a device provided in an embodiment of this specification;

[0056] Figure 2 A schematic diagram of a heat dissipation control method for a device provided in an embodiment of this specification;

[0057] Figure 3 A schematic diagram of a process for obtaining an evaluation temperature provided in an embodiment of this specification;

[0058] Figure 4 A schematic diagram of a process for obtaining a target mapping relationship provided in an implementation manner of this specification;

[0059] Figure 5 A schematic diagram of a process for determining a target mapping relationship provided in an implementation manner of this specification;

[0060] Figure 6 A schematic diagram of a flow chart for outputting a first prediction control parameter provided in an embodiment of this specification;

[0061] Figure 7 A schematic diagram of a flow chart of obtaining a value of a third node and a value of a fourth node provided in an embodiment of this specification;

[0062] Figure 8 A schematic diagram of a heat dissipation control device for a device provided in an embodiment of this specification;

[0063] Fig. 9 A schematic diagram of the internal structure of a computer device provided for an embodiment of this specification. DETAILED DESCRIPTION

[0064] Embodiments of the present invention are described in detail below, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present invention, and should not be construed as limiting the present invention.

[0065] With the rapid development of technology, the performance and computing power of smart all-in-one products are getting higher and higher, and higher requirements are also put forward for their heat dissipation design. Forced air cooling technology is often used for heat dissipation of smart all-in-one products. Fan control is the core of forced air cooling technology and the key to ensuring the normal operation of smart all-in-one products.

[0066] In the related art, the fans are controlled as a whole, but the related art cannot control the fans individually, and when all fans are running at full power, the noise and power consumption are large.

[0067] The fan speed table corresponding to different temperatures is used to control the fan. However, the method of using the fan speed table is not universal.

[0068] Based on the above analysis, the embodiment of this specification provides a heat dissipation control method for a device. The device includes at least one hardware module and a heat dissipation component corresponding to each hardware module. First, for the target hardware module in at least one hardware module, the evaluation temperature and operating frequency corresponding to the target hardware module are obtained, and the target hardware module corresponds to the slot information and type information representing the position of the target hardware module in the device. Then, based on the type information of the target hardware module, the target mapping relationship corresponding to the type information representing the relationship between the evaluation temperature, operating frequency and control parameters corresponding to the target hardware module is determined. Then, based on the target mapping relationship, the target control parameters are determined according to the evaluation temperature and operating frequency. Finally, based on the slot information of the target hardware module, the heat dissipation component corresponding to the target hardware module is determined, and the heat dissipation component is controlled based on the target control parameters to dissipate heat for the target hardware module. By evaluating the multi-data fusion method of temperature and operating frequency, the target control parameters for controlling the heat dissipation component of the target hardware module are more accurately determined, thereby improving the independent control accuracy of the heat dissipation component and reducing noise and energy consumption.

[0069] The heat dissipation control method of the device provided in the implementation mode of this specification can be applied to an all-in-one device with a 19-inch 6U rack structure. Figure 1a The all-in-one device consists of a power module, a main switching module, a switching module, an intelligent management module, a computing module, a storage module, an AI module, an IO board, a backplane module, and a fan module. It should be noted that except for the power module, the intelligent management module, and the fan module, which use fixed slots, the backplane provides mixed slots for other modules, and the required hardware modules can be connected as required. The fan module is connected to the backplane below other hardware modules. The fan module consists of multiple independent fans, and two fans in a vertical row form a fan group. Each fan group is responsible for the heat dissipation of two slot hardware modules.

[0070] Exemplarily, a target hardware module is determined in an all-in-one device with a 19-inch 6U rack-type structure. Taking the target hardware module as an example, the target hardware module includes a core chip. The target hardware module is provided with multiple high-precision temperature sensors. Generally, a high-precision temperature sensor is placed at the air inlet of the target hardware module, responsible for monitoring the air inlet temperature as the external temperature, a high-precision temperature sensor is placed at the air outlet of the target hardware module, responsible for monitoring the air outlet temperature as the external temperature, and a high-precision temperature sensor is placed near the core chip of the target hardware module, responsible for monitoring the temperature near the core chip of the target hardware module as the external temperature. The core chip collects the core temperature of the core chip and multiple external temperatures to obtain a temperature parameter list corresponding to the target hardware module.

[0071] The temperature rise of the target hardware module is related to the operating frequency of the core chip, and the operating frequency of the target hardware module is collected through the core chip.

[0072] Since different slots can be mixed, it is necessary to determine the slot information and type information of the target hardware module. The type information of the target hardware module is saved through EEPROM (Electrically Erasable Programmable Read-Only Memory), and the slot information of the target hardware module is collected through the IO port. Different slots correspond to different IDs.

[0073] Specifically, see Figure 1b ,The temperature parameter list of the target hardware module, the operating frequency of the target hardware module, ,and the slot information and type information of the target hardware module are transmitted to the ,intelligent management module through the RS485 bus.

[0074] Based on the type information of the target hardware module, the intelligent management module determines the target mapping relationship corresponding to the type information. The intelligent management module obtains the temperature weight list associated with the temperature parameter list corresponding to the target hardware module, and uses the temperature weight list to process the temperature parameter list to obtain the evaluation temperature. Then, based on the target mapping relationship, the target control parameters are determined according to the evaluation temperature and the operating frequency. Next, based on the slot information of the target hardware module, the fan group corresponding to the target hardware module is determined. Finally, the intelligent management module sends the target control parameters to the fan group corresponding to the target hardware module through the RS485 bus, thereby controlling the fan group to dissipate heat for the target hardware module.

[0075] This specification provides a method for controlling heat dissipation of a device. Figure 2 The device includes at least one hardware module and a heat dissipation component corresponding to each hardware module. The heat dissipation control method of the device may include the following steps:

[0076] S210. For a target hardware module in at least one hardware module, obtain an evaluation temperature and an operating frequency corresponding to the target hardware module.

[0077] The target hardware module corresponds to slot information and type information, and the slot information represents the position of the target hardware module in the device.

[0078] Specifically, since the device includes at least one hardware module, a hardware module can be selected from multiple hardware modules as a target hardware module. The target hardware module corresponds to slot information and type information. After determining the target hardware module, the evaluation temperature and operating frequency corresponding to the target hardware module are determined using the collected data information corresponding to the target hardware module.

[0079] S220. Based on the type information of the target hardware module, determine a target mapping relationship corresponding to the type information.

[0080] The target mapping relationship represents the relationship between the evaluation temperature, operating frequency and control parameters corresponding to the target hardware module.

[0081] Specifically, due to the different power consumption of different hardware modules, their mapping relationships are also different. Each type of hardware module corresponds to a mapping relationship that matches it. Therefore, after determining the type information of the target hardware module, the type information of the target hardware module is used to determine the corresponding mapping relationship as the target mapping relationship. In some embodiments, the type information of the hardware module and the corresponding mapping relationship are stored to form a mapping relationship table. The type information of the target hardware module is used to search in the mapping relationship table, and the mapping relationship corresponding to the type information of the target hardware module is determined and used as the target mapping relationship.

[0082] S230 , determining target control parameters based on the target mapping relationship, according to the evaluation temperature and the operating frequency.

[0083] S240: Determine a heat dissipation component corresponding to the target hardware module based on the slot information of the target hardware module, and control the heat dissipation component to dissipate heat for the target hardware module based on the target control parameter.

[0084] Specifically, the evaluation temperature and the operating frequency are used as inputs of the target mapping relationship, and based on the target mapping relationship, the evaluation temperature and the operating frequency are calculated to obtain the corresponding target control parameters. Slots at different positions correspond to different heat dissipation components. Therefore, the slot information of the target hardware module is used to determine the heat dissipation component corresponding to the slot information of the target hardware module. The target control parameters are used to control the heat dissipation component corresponding to the target hardware module, thereby achieving heat dissipation of the target hardware module.

[0085] For example, when the heat dissipation component is a fan, since the fan is generally controlled by a PWM (Pulse Width Modulation) waveform, the duty cycle of the PWM waveform is used as a control parameter.

[0086] It should be noted that the temperature response has a lag, and when the operating frequency increases, the temperature does not rise immediately. In the embodiments of this specification, the control parameters for controlling the heat dissipation component are determined by using the relationship between the evaluation temperature, operating frequency and control parameters corresponding to the hardware module. Therefore, based on the above relationship, when the operating frequency is monitored to increase, the fan can intervene in advance to improve the efficiency of heat dissipation.

[0087] In the above implementation, the device includes at least one hardware module and a heat dissipation component corresponding to each hardware module. First, for the target hardware module in at least one hardware module, the evaluation temperature and operating frequency corresponding to the target hardware module are obtained, and the target hardware module corresponds to the slot information and type information representing the position of the target hardware module in the device. Next, based on the type information of the target hardware module, the target mapping relationship corresponding to the type information representing the relationship between the evaluation temperature, the operating frequency and the control parameter corresponding to the target hardware module is determined. Then, based on the target mapping relationship, the target control parameter is determined according to the evaluation temperature and the operating frequency. Finally, based on the slot information of the target hardware module, the heat dissipation component corresponding to the target hardware module is determined, and the heat dissipation component is controlled based on the target control parameter to dissipate heat for the target hardware module. By means of a multi-data fusion method of evaluating the temperature and the operating frequency, the target control parameter for controlling the heat dissipation component of the target hardware module is more accurately determined, thereby improving the independent control accuracy of the heat dissipation component and reducing noise and energy consumption.

[0088] In some embodiments, see Figure 3 The target hardware module includes a core chip, and obtaining the evaluation temperature corresponding to the target hardware module may include the following steps:

[0089] S310. Obtain a temperature parameter list corresponding to the target hardware module.

[0090] S320: Obtain a temperature weight list.

[0091] S330 , using the temperature weight list, processing the temperature parameter list to obtain an evaluation temperature.

[0092] The temperature parameter list includes the core temperature of the core chip and at least one external temperature of the target hardware module. The temperature weight list is associated with the temperature parameter list corresponding to the target hardware module.

[0093] Specifically, the temperature sensors located around the target hardware module are used to determine the temperature outside the target hardware module to obtain at least one external temperature. The core temperature of the target hardware module is collected through the core chip of the target hardware module. The temperature parameter list corresponding to the target hardware module is formed by using at least one external temperature and the core temperature. Since temperature sensors at different positions have different weights, simulation is performed based on the slot information of the target hardware module to determine the temperature weight list associated with the temperature parameter list of the target hardware module. The elements between the temperature weight list and the temperature parameter list correspond one to one, and the temperature weight list is used to calculate the temperature parameter list to obtain the evaluation temperature.

[0094] For example, the temperature parameter list T = [t1t2t3t4 ...], the temperature weight list The temperature weight list is multiplied by the temperature parameter list to obtain the evaluation temperature TP=T*W.

[0095] In some implementations, thermal design simulation technology may be used to perform simulations to determine a temperature weight list, and then configuration is performed based on the temperature weight list on the management interface of the intelligent control all-in-one machine.

[0096] In the above implementation, a temperature parameter list corresponding to the target hardware module is obtained, a temperature weight list is obtained, and the temperature parameter list is processed using the temperature weight list to obtain an evaluation temperature, and the input of the target mapping relationship is determined to determine the control parameters for controlling the heat dissipation component.

[0097] In some embodiments, the target mapping relationship includes a temperature control prediction model. Based on the target mapping relationship, determining the target control parameter according to the evaluation temperature and the operating frequency may include: using the temperature control prediction model to predict the evaluation temperature and the operating frequency to obtain the target control parameter.

[0098] Specifically, the evaluation temperature and operating frequency are used as inputs of the temperature control prediction model, and the temperature control prediction model is used to perform data processing (such as fitting processing) on ​​the evaluation temperature and operating frequency to predict the control parameters and obtain the target control parameters.

[0099] In the above implementation, the temperature control prediction model is used to predict the evaluation temperature and the operating frequency to obtain the target control parameters, thereby providing data support for controlling the heat dissipation component to dissipate heat for the target hardware module.

[0100] In some embodiments, see Figure 4 , the target mapping relationship is obtained by training in the following way:

[0101] S410 . Determine a first prediction control parameter based on an initial mapping relationship and according to evaluation sample temperature and sample frequency.

[0102] S420. Determine a first evaluation parameter based on the first predicted control parameter and the expected control parameter.

[0103] S430: Determine a target mapping relationship based on a comparison result between the first evaluation parameter and a preset threshold parameter.

[0104] The preset threshold parameter may be a standard for evaluating the performance of the mapping relationship, and the preset threshold parameter may be determined according to actual conditions and requirements. The first evaluation parameter may be a degree of difference between the initial mapping relationship and the expected target.

[0105] Specifically, it is necessary to determine whether the initial mapping relationship achieves the expected goal. Therefore, the evaluation sample temperature and sample frequency are used as the input of the initial mapping relationship, and the evaluation sample temperature and sample frequency are calculated based on the initial mapping relationship to obtain the first prediction control parameter. The deviation between the first prediction control parameter and the expected control parameter is evaluated to obtain the first evaluation parameter. The size between the first evaluation parameter and the preset threshold parameter is compared to determine the comparison result between the first evaluation parameter and the preset threshold parameter. According to the comparison result, the target mapping relationship is determined.

[0106] Exemplarily, the first evaluation parameter is determined by the following formula:

[0107]

[0108] Among them, LOSS is the first evaluation parameter, D is the first predictive control parameter, is the expected control parameter.

[0109] In the above implementation, based on the initial mapping relationship, the first prediction control parameter is determined according to the evaluation sample temperature and sample frequency, based on the first prediction control parameter and the expected control parameter, the first evaluation parameter is determined, and based on the comparison result between the first evaluation parameter and the preset threshold parameter, the target mapping relationship is determined to provide more accurate prediction results for practical applications.

[0110] In some implementations, determining the target mapping relationship based on a comparison result between the first evaluation parameter and a preset threshold parameter may include: when the first evaluation parameter is less than the preset threshold parameter, determining the initial mapping relationship as the target mapping relationship.

[0111] Specifically, if the comparison result between the first evaluation parameter and the preset threshold parameter indicates that the first evaluation parameter is less than the preset threshold parameter, indicating that the initial mapping relationship achieves the expected goal, the initial mapping relationship is used as the target mapping relationship.

[0112] In the above implementation, when the first evaluation parameter is less than the preset threshold parameter, the initial mapping relationship is determined to be the target mapping relationship, so as to provide a more accurate prediction result for practical application.

[0113] In some embodiments, see Figure 5 , based on the comparison result between the first evaluation parameter and the preset threshold parameter, determining the target mapping relationship may include the following steps:

[0114] S510: When the first evaluation parameter is greater than or equal to a preset threshold parameter, update the target parameter of the initial mapping relationship based on the learning efficiency, and determine the updated mapping relationship.

[0115] Among them, the target parameter is used to determine the mapping relationship.

[0116] Specifically, if the comparison result between the first evaluation parameter and the preset threshold parameter indicates that the first evaluation parameter is greater than or equal to the preset threshold parameter, it indicates that the initial mapping relationship does not achieve the expected target. Therefore, when the first evaluation parameter is greater than or equal to the preset threshold parameter, it is necessary to update the target parameter of the initial mapping relationship using the learning efficiency. Then, the target parameter of the initial mapping relationship is adjusted using the updated target parameter to obtain an updated mapping relationship.

[0117] In some embodiments, the updated target parameter is determined by a gradient descent method, and the updated target parameter is determined by the following formula:

[0118]

[0119]

[0120]

[0121]

[0122] Where D is the first predictive control parameter, is the expected control parameter, η is the learning efficiency, TP is the value of the first node in the input layer, i is the ith node included in the input layer, j is the jth node included in the hidden layer, Gj is the value of the jth node in the hidden layer, Zj is the weight between the jth node in the hidden layer and the output layer, ΔWij is the updated weight between the ith node in the input layer and the jth node in the hidden layer, Δβj is the updated bias parameter corresponding to the jth node in the hidden layer, ΔZj is the updated weight between the jth node in the hidden layer and the output layer, and Δθ is the updated bias parameter corresponding to the output layer. For example, ΔW13 is the updated weight between the first node in the input layer and the third node in the hidden layer, that is, the updated first weight, ΔW14 is the updated weight between the first node in the input layer and the fourth node in the hidden layer, that is, the updated second weight, ΔW23 is the updated weight between the second node in the input layer and the third node in the hidden layer, that is, the updated third weight, and ΔW24 is the updated weight between the second node in the input layer and the fourth node in the hidden layer, that is, the updated fourth weight.

[0123] ΔZ3 is the updated weight between the third node in the hidden layer and the output layer, that is, the updated first output weight, and ΔZ4 is the updated weight between the fourth node in the hidden layer and the output layer, that is, the updated second output weight.

[0124] Δβ3 is the updated bias parameter corresponding to the third node in the hidden layer, i.e., the updated first bias parameter, and Δβ4 is the updated bias parameter corresponding to the fourth node in the hidden layer, i.e., the updated second bias parameter. S520: Based on the updated mapping relationship, the second prediction control parameter is determined according to the evaluation sample temperature and the sample frequency.

[0125] S530. Determine a second evaluation parameter based on the second predicted control parameter and the expected control parameter.

[0126] S540: Determine a target mapping relationship based on a comparison result between the second evaluation parameter and a preset threshold parameter.

[0127] Specifically, after updating the mapping relationship, it is necessary to determine whether the updated mapping relationship has achieved the expected goal. Therefore, the evaluation sample temperature and sample frequency are used as the input of the updated mapping relationship, and the evaluation sample temperature and sample frequency are calculated based on the updated mapping relationship to obtain the second prediction control parameter. The deviation is evaluated according to the second prediction control parameter and the expected control parameter to obtain the second evaluation parameter. The size between the second evaluation parameter and the preset threshold parameter is compared to obtain a comparison result. If the comparison result between the second evaluation parameter and the preset threshold parameter indicates that the second evaluation parameter is less than the preset threshold parameter, it means that the updated mapping relationship has achieved the expected goal, and the updated mapping relationship is used as the target mapping relationship. If the comparison result between the second evaluation parameter and the preset threshold parameter indicates that the second evaluation parameter is greater than or equal to the preset threshold parameter, it means that the updated mapping relationship has not achieved the expected goal, and it is necessary to use the learning efficiency to update the target parameter again until the evaluation parameter is less than the preset threshold parameter.

[0128] In the above implementation, when the evaluation parameter is greater than or equal to the preset threshold parameter, the target parameter is updated based on the learning efficiency, and the updated mapping relationship is determined. Based on the updated mapping relationship, the second prediction control parameter is determined according to the evaluation sample temperature and the sample frequency. Based on the second prediction control parameter and the expected control parameter, the second evaluation parameter is determined. Based on the comparison result between the second evaluation parameter and the preset threshold parameter, the target mapping relationship is determined to provide more accurate prediction results for practical applications.

[0129] In some implementations, the initial mapping relationship includes an input layer, a hidden layer, and an output layer, the input layer includes a first node and a second node, the hidden layer includes a third node and a fourth node, and the target parameter of the initial mapping relationship includes a first weight, a second weight, a third weight, and a fourth weight:

[0130] The first weight is the weight between the first node and the third node; the second weight is the weight between the first node and the fourth node; the third weight is the weight between the second node and the third node; and the fourth weight is the weight between the second node and the fourth node.

[0131] Specifically, the input layer is the first layer of the neural network, which is responsible for receiving input data. The input layer includes the first node and the second node, representing different features. The hidden layer is located between the input layer and the output layer, and includes the third node and the fourth node. The weights between nodes are learned parameters used to adjust the connection strength between nodes.

[0132] The first weight is the weight of connecting the first node of the input layer and the third node of the hidden layer, which determines the influence of the first node on the third node. The second weight is the weight of connecting the first node of the input layer and the fourth node of the hidden layer, which determines the influence of the first node on the fourth node. The third weight is the weight of connecting the second node of the input layer and the third node of the hidden layer, which determines the influence of the second node on the third node. The fourth weight is the weight of connecting the second node of the input layer and the fourth node of the hidden layer, which determines the influence of the second node on the fourth node. Among them, the initial mapping relationship can be a BP neural network (Backpropagation Neural Network) including an input layer, a hidden layer and an output layer. It should be noted that in this example, the hidden layer takes two nodes as an example, and the hidden layer can also include more nodes, which is not limited in this application.

[0133] In the above implementation, the initial mapping relationship includes an input layer, a hidden layer and an output layer, the input layer includes a first node and a second node, the hidden layer includes a third node and a fourth node, and the target parameters of the initial mapping relationship include a first weight, a second weight, a third weight and a fourth weight, which are used to determine the target mapping relationship.

[0134] In some embodiments, see Figure 6 , based on the initial mapping relationship, determining the first prediction control parameter according to the evaluation sample temperature and sample frequency may include the following steps:

[0135] S610: Use the sample temperature as the value of the first node, and use the sample frequency as the value of the second node.

[0136] S620 . Based on the first weight, the second weight, the third weight, and the fourth weight, the values ​​of the first node and the second node are processed to obtain the value of the third node and the value of the fourth node.

[0137] S630. Obtain a first prediction control parameter based on the value of the third node and the value of the fourth node, and output the first prediction control parameter through an output layer.

[0138] Specifically, the sample temperature and the sample frequency are used as input data, so the sample temperature is used as the value of the first node, and the sample frequency is used as the value of the second node. The weights between different nodes are different, so according to the first weight, the second weight, the third weight and the fourth weight, the values ​​of the first node and the second node are processed to obtain the values ​​of the third node and the fourth node. The third node and the output layer correspond to the first output weight, the fourth node and the output layer correspond to the second output weight, and the output layer corresponds to the output bias parameter. Therefore, the first output weight, the second output weight and the output bias parameter are used to process the values ​​of the third node and the fourth node to obtain the first prediction control parameter. Finally, the output layer outputs the first prediction control parameter.

[0139] Exemplarily, the first output weight is Z3, the second output weight is Z4, the output bias parameter is θ, the value of the third node is G3, and the value of the fourth node is G4. The first predictive control parameter D is determined based on the following formula:

[0140] D = sigmoid (G3*Z3+G4*Z4-θ)

[0141] In the above implementation, the sample temperature is used as the value of the first node, and the sample frequency is used as the value of the second node. Based on the first weight, the second weight, the third weight and the fourth weight, the values ​​of the first node and the second node are processed to obtain the value of the third node and the value of the fourth node. Based on the value of the third node and the value of the fourth node, the first prediction control parameter is obtained, and the first prediction control parameter is output through the output layer to provide a data basis for determining the target mapping relationship.

[0142] In some embodiments, see Figure 7 , the target parameter of the initial mapping relationship also includes a first bias parameter corresponding to the third node and a second bias parameter corresponding to the fourth node. Based on the first weight, the second weight, the third weight and the fourth weight, the values ​​of the first node and the second node are processed to obtain the value of the third node and the value of the fourth node, which may include the following steps:

[0143] S710 . Based on the first weight, the third weight, and the first bias parameter, the values ​​of the first node and the second node are processed to obtain the value of the third node.

[0144] S720: Based on the second weight, the fourth weight, and the second bias parameter, the values ​​of the first node and the second node are processed to obtain the value of the fourth node.

[0145] Specifically, the first weight is the weight between the first node and the third node, the third weight is the weight between the second node and the third node, and the first bias parameter is a translation or offset introduced between the first node, the second node and the third node. Therefore, the value of the first node and the second node is processed using the first weight, the third weight and the first bias parameter to obtain the value of the third node.

[0146] The second weight is the weight between the first node and the fourth node, the fourth weight is the weight between the second node and the fourth node, and the second bias parameter is a translation or offset introduced between the first node, the second node and the fourth node. Therefore, the value of the first node and the second node is processed using the second weight, the fourth weight and the second bias parameter to obtain the value of the fourth node.

[0147] In some implementations, when the control parameter is the duty cycle, since the value of the duty cycle is between 0 and 1, which conforms to the characteristics of the sigmoid function, the sigmoid function is selected as the activation function of each node in the hidden layer.

[0148] Exemplarily, the first weight is W13, the third weight is W23, the first bias parameter is β3, the value of the first node is TP, the value of the second node is P, and the value G3 of the third node is determined based on the following formula:

[0149] ∑3=W13*TP+W23*P-β3

[0150] G3=sigmoid(∑3)

[0151] The second weight is W14, the fourth weight is W24, the second bias parameter is β4, the value of the first node is TP, the value of the second node is P, and the value of the fourth node G4 is determined based on the following formula:

[0152] ∑4=W14*TP+W24*P-β4

[0153] G4=sigmoid(∑4)

[0154] In the above implementation, based on the first weight, the third weight and the first bias parameter, the values ​​of the first node and the second node are processed to obtain the value of the third node, and based on the second weight, the fourth weight and the second bias parameter, the values ​​of the first node and the second node are processed to obtain the value of the fourth node, thereby providing a data basis for the subsequent determination of the predictive control parameters.

[0155] In some implementations, the expected control parameters are determined by: performing simulation based on a sample temperature list, type information of a sample module, and a sample frequency to determine the expected control parameters.

[0156] Specifically, the core temperature of the core chip corresponding to the sample module, the external temperature corresponding to the sample module, and the sample frequency are collected. The core temperature of the core chip corresponding to the sample module and the external temperature corresponding to the sample module constitute a sample temperature list. The type information of the sample module is determined through an interface or the like. The actual situation is simulated using the sample temperature list, the type information of the sample module, and the sample frequency. According to the simulation results, the control parameters that can achieve the heat dissipation effect of the heat dissipation component are determined, and are used as the expected control parameters. Exemplarily, thermal design simulation technology can be used to perform simulation based on the sample temperature list, the type information of the sample module, and the sample frequency to determine the expected control parameters.

[0157] In the above implementation, simulation is performed based on the sample temperature list, the type information of the sample module, and the sample frequency to determine the expected control parameters, thereby providing a data basis for determining the target mapping relationship.

[0158] This specification embodiment provides a heat dissipation control device 800 for a device, see Figure 8 The device includes at least one hardware module and a heat dissipation component corresponding to each hardware module. The heat dissipation control device 800 of the device includes: a data acquisition module 810, a mapping relationship determination module 820, a control parameter determination module 830, and a heat dissipation component control module 840.

[0159] The data acquisition module 810 is used to acquire, for a target hardware module in at least one hardware module, an evaluation temperature and an operating frequency corresponding to the target hardware module; wherein the target hardware module corresponds to slot information and type information, and the slot information represents a position of the target hardware module in the device;

[0160] A mapping relationship determination module 820 is used to determine a target mapping relationship corresponding to the type information based on the type information of the target hardware module; wherein the target mapping relationship represents the relationship between the evaluation temperature, the operating frequency and the control parameter corresponding to the target hardware module;

[0161] A control parameter determination module 830, configured to determine a target control parameter based on the target mapping relationship, according to the evaluation temperature and the operating frequency;

[0162] The heat dissipation component control module 840 is used to determine the heat dissipation component corresponding to the target hardware module based on the slot information of the target hardware module, and control the heat dissipation component to dissipate heat for the target hardware module based on the target control parameter.

[0163] For a detailed description of the heat dissipation control device of the device, please refer to the description of the heat dissipation control method of the device above, which will not be repeated here.

[0164] The embodiment of this specification provides an all-in-one device, the device comprising:

[0165] at least one hardware module;

[0166] at least one heat dissipation component, the at least one heat dissipation component corresponding to the at least one hardware module in a one-to-one manner;

[0167] A heat dissipation control module is used to implement the method according to any one of the above embodiments.

[0168] In some embodiments, a computer device is provided, which may be a terminal, and its internal structure diagram may be as shown in FIG. Fig. 9 As shown. The computer device includes a processor, a memory, a communication interface, a display screen and an input device connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be achieved through WIFI, an operator network, NFC (near field communication) or other technologies. When the computer program is executed by the processor, a heat dissipation control method of a device is implemented. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covered on the display screen, or a button, trackball or touchpad set on the computer device housing, or an external keyboard, touchpad or mouse, etc.

[0169] Those skilled in the art will understand that Fig. 9 The structure shown in the figure is merely a block diagram of a portion of the structure related to the solution disclosed in this specification, and does not constitute a limitation on the computer device to which the solution disclosed in this specification is applied. Specifically, the computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0170] An embodiment of the present specification provides a computer device, including a memory and a processor, wherein a computer program is stored in the memory, and when the processor executes the computer program, the method steps in the above embodiment are implemented.

[0171] An embodiment of the present specification provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the steps of the method in any of the above embodiments are implemented.

[0172] The embodiments of this specification provide a computer program product, which includes instructions. When the instructions are executed by a processor of a computer device, the computer device can perform the steps of any method of the above embodiments.

[0173] It should be noted that the logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be specifically implemented in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in combination with these instruction execution systems, devices or apparatuses. For the purposes of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in combination with these instruction execution systems, devices or apparatuses. More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection portion with one or more wirings (electronic device), a portable computer disk box (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium and then editing, interpreting or otherwise processing in a suitable manner if necessary, and then stored in a computer memory.

Claims

1. A method for controlling heat dissipation of a device, characterized in that: The device includes at least one hardware module and a heat dissipation component corresponding to each hardware module, and the method includes: For a target hardware module in at least one hardware module, obtaining an evaluation temperature and an operating frequency corresponding to the target hardware module; wherein the target hardware module corresponds to slot information and type information, and the slot information represents a position of the target hardware module in the device; Based on the type information of the target hardware module, determine a target mapping relationship corresponding to the type information; wherein the target mapping relationship represents a relationship between the evaluation temperature, the operating frequency and the control parameter corresponding to the target hardware module; Based on the target mapping relationship, determining a target control parameter according to the evaluation temperature and the operating frequency; Based on the slot information of the target hardware module, a heat dissipation component corresponding to the target hardware module is determined, and based on the target control parameter, the heat dissipation component is controlled to dissipate heat for the target hardware module.

2. The method according to claim 1, characterized in that The target hardware module includes a core chip, and obtaining an evaluation temperature corresponding to the target hardware module includes: Acquire a temperature parameter list corresponding to the target hardware module; wherein the temperature parameter list includes the core temperature of the core chip and at least one external temperature of the target hardware module; Acquire a temperature weight list, wherein the temperature weight list is associated with a temperature parameter list corresponding to the target hardware module; The temperature parameter list is processed using the temperature weight list to obtain the evaluation temperature.

3. The method according to claim 1, characterized in that The target mapping relationship includes a temperature control prediction model; the target control parameter is determined based on the target mapping relationship according to the evaluation temperature and the operating frequency, including: The temperature control prediction model is used to predict the evaluation temperature and the operating frequency to obtain target control parameters.

4. The method according to claim 1, characterized in that: The target mapping relationship is obtained by training in the following way: Based on the initial mapping relationship, determining a first prediction control parameter according to the evaluation sample temperature and the sample frequency; Determining a first evaluation parameter based on the first predicted control parameter and the expected control parameter; The target mapping relationship is determined based on a comparison result between the first evaluation parameter and a preset threshold parameter.

5. The method according to claim 4, characterized in that The determining the target mapping relationship based on a comparison result between the first evaluation parameter and a preset threshold parameter includes: When the first evaluation parameter is less than the preset threshold parameter, the initial mapping relationship is determined to be the target mapping relationship.

6. The method according to claim 4, characterized in that The determining the target mapping relationship based on a comparison result between the first evaluation parameter and a preset threshold parameter includes: When the first evaluation parameter is greater than or equal to the preset threshold parameter, updating the target parameter of the initial mapping relationship based on the learning efficiency, and determining the updated mapping relationship; wherein the target parameter is used to determine the mapping relationship; Based on the updated mapping relationship, determining a second prediction control parameter according to the evaluation sample temperature and the sample frequency; determining a second evaluation parameter based on the second predicted control parameter and the expected control parameter; The target mapping relationship is determined based on a comparison result between the second evaluation parameter and a preset threshold parameter.

7. The method according to claim 4, characterized in that The initial mapping relationship includes an input layer, a hidden layer and an output layer, the input layer includes a first node and a second node, the hidden layer includes a third node and a fourth node, and the target parameters of the initial mapping relationship include a first weight, a second weight, a third weight and a fourth weight; The first weight is the weight between the first node and the third node; The second weight is the weight between the first node and the fourth node; The third weight is the weight between the second node and the third node; The fourth weight is a weight between the second node and the fourth node.

8. The method according to claim 7, characterized in that The determining of the first prediction control parameter based on the initial mapping relationship and according to the evaluation sample temperature and the sample frequency includes: Taking the sample temperature as the value of the first node and taking the sample frequency as the value of the second node; Based on the first weight, the second weight, the third weight, and the fourth weight, the values ​​of the first node and the second node are processed to obtain the value of the third node and the value of the fourth node; Based on the value of the third node and the value of the fourth node, the first prediction control parameter is obtained, and the first prediction control parameter is output through the output layer.

9. The method according to claim 8, characterized in that The target parameter of the initial mapping relationship further includes a first bias parameter corresponding to the third node and a second bias parameter corresponding to the fourth node; and the processing of the values ​​of the first node and the second node based on the first weight, the second weight, the third weight, and the fourth weight to obtain the value of the third node and the value of the fourth node includes: Based on the first weight, the third weight and the first bias parameter, the values ​​of the first node and the second node are processed to obtain the value of the third node; Based on the second weight, the fourth weight and the second bias parameter, the values ​​of the first node and the second node are processed to obtain the value of the fourth node.

10. The method according to claim 4, characterized in that The expected control parameters are determined by: The expected control parameters are determined by performing simulation based on the sample temperature list, the type information of the sample module, and the sample frequency.

11. An all-in-one device, characterized in that: The device comprises: at least one hardware module; at least one heat dissipation component, the at least one heat dissipation component corresponding to the at least one hardware module in a one-to-one manner; A heat dissipation control module, used to implement the method according to any one of claims 1-10.

12. A heat dissipation control device for equipment, characterized in that: The device comprises at least one hardware module and a heat dissipation component corresponding to each hardware module, and the apparatus comprises: A data acquisition module, configured to acquire, for a target hardware module in at least one hardware module, an evaluation temperature and an operating frequency corresponding to the target hardware module; wherein the target hardware module corresponds to slot information and type information, and the slot information represents a position of the target hardware module in the device; A mapping relationship determination module, configured to determine a target mapping relationship corresponding to the type information based on the type information of the target hardware module; wherein the target mapping relationship represents a relationship between the evaluation temperature, the operating frequency and a control parameter corresponding to the target hardware module; A control parameter determination module, configured to determine a target control parameter based on the target mapping relationship, according to the evaluation temperature and the operating frequency; The heat dissipation component control module is used to determine the heat dissipation component corresponding to the target hardware module based on the slot information of the target hardware module, and control the heat dissipation component to dissipate heat for the target hardware module based on the target control parameter.

13. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 10 are implemented.

14. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 10 are implemented.

Images