Bidirectional convection enhanced cooling fan control method and control system

Abstract

The invention relates to the technical field of intelligent fan control, and provides a bidirectional convection enhanced cooling fan control method and system, and the method comprises the steps: controlling a bidirectional convection enhanced cooling fan to execute various operation modes according to each group of cooling working condition feature data, and recording corresponding cooling effect data; the heat dissipation working condition characteristics comprise equipment temperature, equipment load, environment temperature and environment humidity; the heat dissipation effect data comprises the equipment temperature reduction amplitude and the fan energy consumption in the preset time; determining an optimal operation mode corresponding to each group of heat dissipation working condition characteristic data based on the heat dissipation effect data; constructing a decision tree model based on the heat dissipation working condition feature data and the corresponding optimal operation mode; determining an optimal operation mode of the equipment under the current working condition by utilizing the decision tree model; and controlling the bidirectional convection enhanced cooling fan to operate according to the optimal operation mode under the current working condition. The matching degree of the fan operation control mode and the heat dissipation requirement is improved, and the energy consumption and the heat dissipation effect are balanced.

Description

Technical Field

[0001] This application relates to the technical field of intelligent control of fans, and particularly to a control method and control system for a two-way convection enhanced heat dissipation fan. Background Art

[0002] In the field of electronic devices, with the continuous improvement of chip performance and the increasing integration of devices, the heat dissipation problem has become a key factor affecting the stability and service life of devices. The two-way convection enhanced heat dissipation fan generates different airflows by rotating clockwise and counterclockwise. Compared with traditional unidirectional fans, it can more effectively improve the heat dissipation efficiency and is widely used in the fields of servers, high-performance computers, industrial control devices, etc.

[0003] Currently, the control methods for two-way convection enhanced heat dissipation fans are mainly based on simple threshold settings or fixed operating modes. The threshold-based control method only compares the single temperature data collected by the temperature sensor with the preset temperature threshold, and adjusts the fan speed when the temperature exceeds the threshold. The control method of the fixed operating mode is that during the operation of the device, the fan always operates in a predetermined mode. The existing fan operation control method has a low matching degree with the heat dissipation demand, resulting in poor heat dissipation effect or energy waste, and the energy consumption and heat dissipation effect cannot reach a balance. Summary of the Invention

[0004] Aiming at the above technical problems, the purpose of this application is to provide a control method and control system for a two-way convection enhanced heat dissipation fan, aiming to solve the technical problems that the existing fan operation control method has a low matching degree with the heat dissipation demand, resulting in poor heat dissipation effect or energy waste, and the energy consumption and heat dissipation effect cannot reach a balance.

[0005] In a first aspect, an embodiment of this application provides a control method for a two-way convection enhanced heat dissipation fan, including:

[0006] For each set of heat dissipation condition characteristic data, control the two-way convection enhanced heat dissipation fan to execute various operating modes, and record the heat dissipation effect data under various operating modes; wherein, the heat dissipation condition characteristics include device temperature, device load, ambient temperature, and ambient humidity; the heat dissipation effect data includes the temperature drop amplitude of the device within a preset time and the fan energy consumption;

[0007] Based on the heat dissipation effect data, determine the optimal operating mode corresponding to each set of heat dissipation condition characteristic data;

[0008] Based on the heat dissipation condition characteristic data and the corresponding optimal operating mode, construct a decision tree model; wherein, each leaf node of the decision tree corresponds to an optimal operating mode;

[0009] Use the decision tree model to determine the optimal operating mode of the device under the current working condition;

[0010] Control the two-way convection enhanced heat dissipation fan to operate according to the optimal operation mode under the current working condition.

[0011] Further, the step of determining the optimal operation mode corresponding to each set of heat dissipation working condition characteristic data based on the heat dissipation effect data includes:

[0012] Based on the multi-objective optimization algorithm and the heat dissipation effect data, perform a Pareto optimal solution search between heat dissipation efficiency and energy consumption, and select the mode that is on the heat dissipation efficiency - energy consumption frontier curve and ranks first in terms of scoring as the optimal operation mode under this working condition.

[0013] Further, the construction method of the heat dissipation efficiency - energy consumption frontier curve includes:

[0014] For a set of heat dissipation working condition characteristic data, plot the heat dissipation efficiency and energy consumption data corresponding to all operation modes in a two-dimensional coordinate system to form a scatter plot, where the abscissa represents energy consumption and the ordinate represents heat dissipation efficiency;

[0015] In the scatter plot, identify the data points with the lowest energy consumption at a given heat dissipation efficiency level, and use these data points to generate the heat dissipation efficiency - energy consumption frontier curve.

[0016] Further, the score is calculated based on the maximum heat dissipation efficiency of all operation modes, the minimum heat dissipation efficiency of all operation modes, the heat dissipation efficiency of the target operation mode, the weight coefficient of heat dissipation efficiency, the maximum energy consumption of all operation modes, the minimum energy consumption of all operation modes, the energy consumption of the target operation mode, and the weight coefficient of energy consumption.

[0017] Further, the step of constructing a decision tree model based on the heat dissipation working condition characteristic data and the corresponding optimal operation mode includes:

[0018] S31. Calculate the information gain rate of each heat dissipation working condition characteristic, select the heat dissipation working condition characteristic with the largest information gain rate as the root node, and divide the data set into multiple subsets according to different values of the heat dissipation working condition characteristic with the largest gain rate;

[0019] S32. For each subset divided by the root node, repeat step S31, that is, recalculate the information gain rate of the remaining heat dissipation working condition characteristics on this subset, select the heat dissipation working condition characteristic with the largest information gain rate for division, and generate new subsets;

[0020] S33. Check whether each subset meets the stop condition. If it meets the stop condition, mark this subset as a leaf node and determine its corresponding optimal operation mode; if not, continue to recurse;

[0021] S34. Repeat steps S32 and S33 until all subsets meet the stop condition, and a decision tree model for determining the optimal operation mode is constructed.

[0022] Further, the step of using the decision tree model to determine the optimal operation mode of the device under the current working condition includes:

[0023] Obtain the heat dissipation condition characteristic data of the device under the current working condition;

[0024] Based on the heat dissipation condition characteristic data under the current working condition, determine the optimal operation mode of the device under the current working condition through the decision tree model.

[0025] In a second aspect, an embodiment of the present application provides a two-way convection enhanced heat dissipation fan control system, and the system includes:

[0026] A first control module, configured to control the two-way convection enhanced heat dissipation fan to execute various operation modes for each set of heat dissipation condition characteristic data, and record the heat dissipation effect data under various operation modes; wherein, the heat dissipation condition characteristics include device temperature, device load, ambient temperature, and ambient humidity; the heat dissipation effect data includes the temperature drop amplitude of the device within a preset time and the fan energy consumption;

[0027] A first determination module, configured to determine the optimal operation mode corresponding to each set of heat dissipation condition characteristic data based on the heat dissipation effect data;

[0028] A decision tree construction module, configured to construct a decision tree model based on the heat dissipation condition characteristic data and the corresponding optimal operation mode; wherein, each leaf node of the decision tree corresponds to an optimal operation mode;

[0029] A second determination module, configured to use the decision tree model to determine the optimal operation mode of the device under the current working condition;

[0030] A second control module, configured to control the two-way convection enhanced heat dissipation fan to operate according to the optimal operation mode under the current working condition.

[0031] Further, the first determination module is specifically configured to:

[0032] Based on the multi-objective optimization algorithm and the heat dissipation effect data, perform a Pareto optimal solution search between heat dissipation efficiency and energy consumption, and select the mode that is on the heat dissipation efficiency - energy consumption frontier curve and has the first score ranking as the optimal operation mode under this working condition.

[0033] Further, the construction method of the heat dissipation efficiency - energy consumption frontier curve includes:

[0034] For a set of heat dissipation condition characteristic data, plot the heat dissipation efficiency and energy consumption data corresponding to all operating modes in a two-dimensional coordinate system to form a scatter plot, where the abscissa represents the energy consumption and the ordinate represents the heat dissipation efficiency;

[0035] In the scatter plot, identify the data points with the lowest energy consumption at a given heat dissipation efficiency level, and use the data points to generate a heat dissipation efficiency - energy consumption frontier curve.

[0036] Furthermore, the score is calculated based on the maximum heat dissipation efficiency of all operating modes, the minimum heat dissipation efficiency of all operating modes, the heat dissipation efficiency of the target operating mode, the weight coefficient of the heat dissipation efficiency, the maximum energy consumption of all operating modes, the minimum energy consumption of all operating modes, the energy consumption of the target operating mode, and the weight coefficient of the energy consumption.

[0037] The embodiments of the present application have the following technical effects:

[0038] (1) Compared with the limitations brought by only controlling the fan based on a single temperature or a fixed mode in the traditional method, the embodiments of the present application consider the heat dissipation characteristic data such as the device temperature, device load, ambient temperature, and ambient humidity, and consider the heat dissipation effect data such as the temperature drop amplitude of the device within a preset time and the energy consumption of the fan. And based on the heat dissipation effect data, determine the optimal operating mode corresponding to each group of heat dissipation condition characteristic data, construct a decision tree model for predicting the optimal operating mode based on the heat dissipation condition characteristic data and the corresponding optimal operating mode, and use the decision tree model to determine the optimal operating mode of the device under the current working condition, which greatly improves the matching degree between the fan operation control method and the heat dissipation requirement, and balances the energy consumption and the heat dissipation effect.

[0039] (2) The relationship between the heat dissipation condition characteristics (device temperature, device load, ambient temperature, and ambient humidity) of the device and the optimal operating mode of the fan is often non - linear. And the present application uses a decision tree to automatically discover complex patterns and non - linear relationships in the data, and can accurately determine the appropriate fan operating mode under various working conditions.

[0040] (3) The training and prediction processes of the decision tree are relatively simple, and the computational complexity is low. When constructing the decision tree, only sorting and comparison operations need to be performed on the data, and the model training can be quickly completed for large - scale heat dissipation condition data. When predicting, only node judgments need to be made along the path of the decision tree, and the optimal operating mode of the fan under the current working condition can be determined in a short time, meeting the requirements of real - time control. Brief Description of the Drawings

[0041] To more clearly illustrate the technical solutions of this application, the following will briefly introduce the accompanying drawings required for the implementation. Obviously, the accompanying drawings in the following description are only some embodiments of this application. For those of ordinary skill in the art, without creative efforts, other accompanying drawings can be obtained based on these drawings.

[0042] Figure 1 It is a schematic flowchart of a two-way convection enhanced heat dissipation fan control method provided by an embodiment of this application;

[0043] Figure 2 It is a schematic structural diagram of a two-way convection enhanced heat dissipation fan control system provided by another embodiment of this application. Specific embodiments

[0044] In order to make the purpose, technical solutions and advantages of this application more clear, the following further details this application in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain this application and are not used to limit this application.

[0045] Those skilled in the art of this technology can understand that unless specifically stated, the singular forms "a", "an", "the above" and "the" used herein may also include the plural forms. It should be further understood that the term "including" used in the specification of this application means the presence of features, integers, steps, operations, elements, modules and / or components, but does not exclude the presence or addition of one or more other features, integers, steps, operations, elements, modules, components and / or their groups. It should be understood that when we say an element is "connected" or "coupled" to another element, it can be directly connected or coupled to other elements, or there may also be intermediate elements. In addition, the "connection" or "coupling" used herein may include wireless connection or wireless coupling. The phrase "and / or" used herein includes all or any module and all combinations of one or more related listed items.

[0046] Those skilled in the art of this technology can understand that unless otherwise defined, all terms (including technical terms and scientific terms) used herein have the same meaning as the general understanding of those of ordinary skill in the field to which this application belongs. It should also be understood that terms such as those defined in a general dictionary should be understood to have a meaning consistent with the meaning in the context of the prior art, and will not be interpreted with an idealized or overly formal meaning unless specifically defined as here.

[0047] Please refer to Figure 1 , an embodiment of this application provides a two-way convection enhanced heat dissipation fan control method, and the method includes:

[0048] S1. For each set of heat dissipation condition characteristic data, control the bidirectional convection enhanced cooling fan to execute various operating modes, and record the heat dissipation effect data under various operating modes. Among them, the heat dissipation condition characteristics include device temperature, device load, ambient temperature, and ambient humidity. The heat dissipation effect data includes the temperature drop amplitude of the device within a preset time and the fan energy consumption.

[0049] S2. Based on the heat dissipation effect data, determine the optimal operating mode corresponding to each set of heat dissipation condition characteristic data.

[0050] S3. Based on the heat dissipation condition characteristic data and the corresponding optimal operating mode, construct a decision tree model. Among them, each leaf node of the decision tree corresponds to an optimal operating mode.

[0051] S4. Use the decision tree model to determine the optimal operating mode of the device under the current working condition.

[0052] S5. Control the bidirectional convection enhanced cooling fan to operate according to the optimal operating mode under the current working condition.

[0053] In the embodiment of the present application, the device temperature is the temperature value detected by the temperature sensor set at a specified position inside the device. The device temperature may specifically include the CPU temperature and the GPU temperature. The ambient temperature is the temperature value detected by the temperature sensor set at a specified position outside the device. The ambient humidity is the temperature value detected by the temperature sensor set at a specified position outside the device. The device load includes indicators such as CPU usage rate, GPU usage rate, and memory occupancy rate. The bidirectional convection enhanced cooling fan includes two fans. These two fans are on the same axis and face opposite directions. The operating modes of the bidirectional convection enhanced cooling fan are divided according to the rotation speed of the fan, the rotation direction of the fan, and the number of working fans. In the embodiment of the present application, for each set of heat dissipation condition data, specifically, drive the bidirectional convection enhanced cooling fan to execute various operating modes through the PWM speed regulation module and the H-bridge drive circuit, and record the heat dissipation effect data under various modes. Among them, the PWM speed regulation module is used to control the rotation speed of each fan in the bidirectional convection enhanced cooling fan, and the H-bridge drive circuit is used to control the rotation direction of each fan in the bidirectional convection enhanced cooling fan. The PWM speed regulation module and the H-bridge drive circuit cooperate with each other, so that the bidirectional convection enhanced cooling fan can execute operating modes with different rotation speeds such as clockwise, counterclockwise, and bidirectional.

[0054] The embodiment of the present application has the following technical effects:

[0055] (1) Compared with the limitations brought by only controlling the fan based on a single temperature or a fixed mode in the traditional method, the embodiment of the present application takes into account the heat dissipation characteristic data such as device temperature, device load, ambient temperature, and ambient humidity, and also takes into account the heat dissipation effect data such as the temperature drop amplitude of the device within a preset time and the fan energy consumption. Based on the heat dissipation effect data, the optimal operation mode corresponding to each group of heat dissipation condition characteristic data is determined, and a decision tree model for predicting the optimal operation mode is constructed based on the heat dissipation condition characteristic data and the corresponding optimal operation mode. Using the decision tree model to determine the optimal operation mode of the device under the current working condition greatly improves the matching degree between the fan operation control method and the heat dissipation requirement, and balances the energy consumption and the heat dissipation effect.

[0056] (2) The relationship between the heat dissipation condition characteristics (device temperature, device load, ambient temperature, and ambient humidity) of the device and the optimal operation mode of the fan is often non-linear. The decision tree adopted in the present application can automatically discover complex patterns and non-linear relationships in the data, and can accurately determine the appropriate fan operation mode under complex working conditions.

[0057] (3) The training and prediction processes of the decision tree are relatively simple, and the computational complexity is low. When constructing the decision tree, only sorting and comparison operations need to be performed on the data, and the model training can be quickly completed for large-scale heat dissipation condition data. When predicting, only node judgments need to be made along the path of the decision tree, and the optimal operation mode of the fan under the current working condition can be determined in a short time, meeting the requirements of real-time control.

[0058] In one embodiment, the step of determining the optimal operation mode corresponding to each group of heat dissipation condition characteristic data based on the heat dissipation effect data includes:

[0059] Based on the multi-objective optimization algorithm and the heat dissipation effect data, perform a Pareto optimal solution search between heat dissipation efficiency and energy consumption, and select the mode that is on the heat dissipation efficiency - energy consumption frontier curve and ranks first in terms of score as the optimal operation mode under this working condition.

[0060] In the embodiment of the present application, for each group of heat dissipation condition characteristic data, control the two-way convection enhanced heat dissipation fan to execute various operation modes, and record the heat dissipation effect data under various operation modes. In order to construct the heat dissipation efficiency - energy consumption frontier curve, it is necessary to calculate the heat dissipation efficiency under different operation modes according to the recorded data. Specifically, various methods can be used to quantify the heat dissipation efficiency. For example, based on the temperature drop amplitude of the device and the preset time, calculate the temperature drop rate per unit time, and use this as an index of the heat dissipation efficiency. Assume that the initial temperature of the device is T0, and the temperature drops to T1 after the preset time t, then the heat dissipation efficiency .

[0061] Multi-objective optimization algorithms (such as NSGA-II) are methods for solving optimization problems with multiple conflicting objectives simultaneously. In the embodiments of this application, the objectives are to optimize the heat dissipation efficiency and energy consumption simultaneously. For example, increasing the fan speed may improve the heat dissipation efficiency, but at the same time increase the energy consumption. There is a certain contradiction between these two objectives, and the multi-objective optimization algorithm is to find a balance in this contradiction. The Pareto optimal solution refers to a solution in a multi-objective optimization problem where there is no other feasible solution that can improve the value of other objective functions without reducing the value of at least one objective function. Simply put, under the current conditions, there is no way to improve one objective without harming other objectives. Its search process is to traverse various different fan operation modes, including different speed combinations, forward and reverse rotation combinations, etc., calculate the heat dissipation efficiency and energy consumption in each mode, and then find those modes that can no longer improve the heat dissipation efficiency and reduce the energy consumption simultaneously by changing the operation mode. The solutions corresponding to these modes are the Pareto optimal solutions.

[0062] In one embodiment, the method for constructing the heat dissipation efficiency - energy consumption frontier curve includes:

[0063] For a set of heat dissipation condition characteristic data, plot the heat dissipation efficiency and energy consumption data corresponding to all operation modes in a two-dimensional coordinate system to form a scatter plot, where the abscissa represents the energy consumption and the ordinate represents the heat dissipation efficiency;

[0064] In the scatter plot, identify the data points with the lowest energy consumption at a given heat dissipation efficiency level, and use these data points to generate the heat dissipation efficiency - energy consumption frontier curve.

[0065] In the embodiments of this application, the method for constructing the heat dissipation efficiency - energy consumption frontier curve is as follows: Plot the heat dissipation efficiency and energy consumption data corresponding to all operation modes in a two-dimensional coordinate system to form a scatter plot, where the abscissa represents the energy consumption and the ordinate represents the heat dissipation efficiency. In the scatter plot, identify the data points with the lowest energy consumption at a given heat dissipation efficiency level. These points constitute the heat dissipation efficiency - energy consumption frontier curve, also known as the Pareto frontier. Some mathematical algorithms can be used to identify these frontier points, such as relevant techniques in multi-objective optimization algorithms like the fast non-dominated sorting algorithm (NSGA-II). This algorithm will sort all data points and identify the data points that are in a non-dominated position (that is, there is no other point that is better than it in both the heat dissipation efficiency and energy consumption indicators). Connecting these points forms the heat dissipation efficiency - energy consumption frontier curve.

[0066] In the embodiments of this application, through the Pareto optimal solution search between the heat dissipation efficiency and the energy consumption based on the multi-objective optimization algorithm and the heat dissipation effect data, and selecting the mode that is on the heat dissipation efficiency - energy consumption frontier curve and ranks first in terms of scoring as the optimal operation mode under this condition, it is possible to achieve an accurate dynamic balance between the heat dissipation performance and the energy consumption.

[0067] Further, if there are some discontinuities or large fluctuations between the leading points, the curve can also be further optimized and adjusted according to the actual situation. For example, some abnormal points that deviate significantly from the overall trend can be removed, or some parts of the curve can be corrected according to domain knowledge to improve data reliability.

[0068] In one embodiment, the score is calculated based on the maximum heat dissipation efficiency of all operating modes, the minimum heat dissipation efficiency of all operating modes, the heat dissipation efficiency of the target operating mode, the weight coefficient of the heat dissipation efficiency, the maximum energy consumption of all operating modes, the minimum energy consumption of all operating modes, the energy consumption of the target operating mode, and the weight coefficient of the energy consumption.

[0069] In the embodiments of the present application, specifically, the calculation formula of the score is:

[0070] ;

[0071] Where represents the score, which is used to measure the quality of each operating mode. represents the heat dissipation efficiency of the target operating mode. represents the minimum heat dissipation efficiency of all operating modes. represents the maximum heat dissipation efficiency of all operating modes. represents the minimum energy consumption of all operating modes. represents the maximum energy consumption of all operating modes. represents the energy consumption value of the target operating mode. represents the weight coefficient of the heat dissipation efficiency. represents the weight coefficient of the energy consumption.

[0072] In the embodiments of the present application, through this term, the value of the heat dissipation efficiency is normalized to between 0 and 1 to reflect the relative level of the heat dissipation efficiency of this operating mode among all modes. This term normalizes the energy consumption value to between 0 and 1, but it is measured by the relative reduction degree of the energy consumption, that is, the lower the energy consumption, the higher the value of this term.

[0073] The embodiments of the present application can better achieve the precise dynamic balance between the heat dissipation performance and the energy consumption based on the score formula.

[0074] In one embodiment, the steps of constructing a decision tree model based on the heat dissipation condition characteristic data and the corresponding optimal operating mode include:

[0075] S31. Calculate the information gain ratio of each heat dissipation condition feature, select the heat dissipation condition feature with the largest information gain ratio as the root node, and divide the data set into multiple subsets according to different values of the heat dissipation condition feature with the largest gain ratio;

[0076] S32. For each subset divided by the root node, repeat step S31, that is, recalculate the information gain ratio of the remaining heat dissipation condition features on this subset, select the heat dissipation condition feature with the largest information gain ratio for division, and generate new subsets;

[0077] S33. Check whether each subset meets the stop condition. If it meets the stop condition, mark this subset as a leaf node and determine its corresponding optimal operation mode; if not, continue to recurse;

[0078] S34. Repeat step S32 and step S33 until all subsets meet the stop condition, and construct a decision tree model for determining the optimal operation mode.

[0079] In the embodiment of the present application, first, calculate the information gain ratio for each heat dissipation condition feature (equipment temperature, equipment load, ambient temperature, and ambient humidity). Then compare the information gain ratios of all heat dissipation condition features, select the heat dissipation condition feature with the largest information gain ratio as the root node of the decision tree. For each subset (sub-node) divided by the root node, repeat the above steps, that is, recalculate the information gain ratio of the remaining heat dissipation condition features on this subset, select the heat dissipation condition feature with the largest information gain ratio for division, and generate new subsets (sub-nodes). Subsequently, check whether each subset (sub-node) meets the stop condition, such as whether the number of samples in the node is less than a preset threshold, the sample purity reaches a certain degree (such as all samples in the node belong to the same category), the depth of the tree reaches the preset value, etc. If it meets the stop condition, mark this sub-node as a leaf node and determine its corresponding optimal operation mode; if not, continue to recursively divide. Repeat step S32 and step S33 until all sub-nodes meet the stop condition. At this time, the complete decision tree is constructed and can be used to determine the optimal operation mode for new data. The embodiment of the present application uses the decision tree model to discover the complex relationship between the heat dissipation condition features (equipment temperature, equipment load, ambient temperature, and ambient humidity) and the optimal operation mode of the fan, and can accurately determine the appropriate fan operation mode under various working conditions. Through the decision tree, the optimal operation mode of the fan under the current working condition can be determined in a short time to meet the requirements of real-time control.

[0080] In one embodiment, the step of using the decision tree model to determine the optimal operation mode of the device under the current working condition includes:

[0081] Obtain the heat dissipation condition feature data of the device under the current working condition;

[0082] Based on the heat dissipation condition characteristic data under the current working condition, determine the optimal operation mode of the device under the current working condition through the decision tree model.

[0083] In the embodiment of the present application, after the decision tree model is constructed, the decision tree model can be used to determine the optimal operation mode of the device under the current working condition. Specifically, input the heat dissipation condition characteristics of the device under the current working condition, including data such as device temperature, device load, ambient temperature, and ambient humidity, into the decision tree model. Starting from the root node, screen the data according to the characteristic judgment condition corresponding to the root node. For example, if the root node is the device temperature and the set judgment condition is "whether the device temperature is greater than A °C", when the input current device temperature data meets this condition, the data will enter the next node along the corresponding branch; otherwise, it will enter another branch. At each intermediate node, the model will repeat the above judgment process, and gradually classify the data according to the characteristics and judgment conditions corresponding to the node. As the data continuously descends in the decision tree, it will eventually reach the leaf node. And each leaf node corresponds to an optimal operation mode, which is determined for a specific combination of working condition characteristics after being trained with a large amount of historical data, and can achieve an optimal balance of energy consumption while ensuring the heat dissipation effect. Thus, through the layer-by-layer judgment and screening of the decision tree model, the optimal operation mode of the device under the current working condition is successfully determined.

[0084] In one embodiment, after the step of controlling the two-way convection enhanced heat dissipation fan to operate according to the optimal operation mode under the current working condition, the method further includes:

[0085] Regularly collect new heat dissipation condition characteristic data and corresponding optimal operation mode data, and update and optimize the decision tree model.

[0086] In the embodiment of the present application, by regularly updating and optimizing the decision tree model, it is possible to avoid the problem of reduced prediction accuracy caused by the possible change in the relationship between the heat dissipation condition and the optimal operation mode due to the long-term increase in the device usage time and the long-term change in environmental conditions.

[0087] In one embodiment, after the step of controlling the two-way convection enhanced heat dissipation fan to operate according to the optimal operation mode under the current working condition, the method further includes:

[0088] Send the heat dissipation effect data of the optimal operation mode under the current working condition to the display screen for display.

[0089] In one embodiment, after the step of controlling the two-way convection enhanced heat dissipation fan to operate according to the optimal operation mode under the current working condition, the method further includes:

[0090] Using a camera, detect whether there are people in the space where the device is located based on a target detection algorithm;

[0091] If there are, recognize the facial expressions of the detected people and detect the noise of the device;

[0092] If the noise of the device is greater than a preset threshold and the expression of the detected person is dissatisfied, gradually reduce the rotation speed of the two-way convection enhanced cooling fan based on the cooling efficiency.

[0093] In the embodiments of the present application, a camera is installed on the device to clearly capture the picture of the space where the device is located. At the same time, a noise detection sensor is installed to accurately measure the noise generated during the operation of the device. The recognition of the facial expressions of the detected people can use a facial expression recognition model, such as an expression recognition network based on deep learning, which can be trained using a publicly available facial expression dataset. In the embodiments of the present application, when it is detected that the device noise exceeds the preset threshold and the person's expression is dissatisfied, based on the cooling efficiency, and on the premise of meeting the cooling requirements, gradually reduce the fan rotation speed, which can not only meet the cooling requirements but also effectively reduce the interference of noise to people. Among them, during the process of gradually reducing the fan rotation speed, the facial expressions of the detected people are monitored in real time to judge whether the mood of the detected person improves. The embodiments of the present application gradually reduce the fan rotation speed by comprehensively considering the cooling efficiency and the mood of the detected person, which can not only meet the cooling requirements but also effectively reduce the interference of noise to people.

[0094] The embodiments of the present application also provide a two-way convection enhanced cooling fan control system, and the system includes:

[0095] The first control module 1 is used to control the two-way convection enhanced cooling fan to execute various operating modes for each set of cooling condition characteristic data, and record the cooling effect data under various operating modes; among them, the cooling condition characteristics include device temperature, device load, ambient temperature, and ambient humidity; the cooling effect data includes the device temperature drop amplitude and fan energy consumption within a preset time;

[0096] The first determination module 2 is used to determine the optimal operating mode corresponding to each set of cooling condition characteristic data based on the cooling effect data;

[0097] The decision tree construction module 3 is used to construct a decision tree model based on the cooling condition characteristic data and the corresponding optimal operating mode; among them, each leaf node of the decision tree corresponds to an optimal operating mode;

[0098] The second determination module 4 is used to use the decision tree model to determine the optimal operating mode of the device under the current working condition;

[0099] The second control module 5 is used to control the two-way convection enhanced heat dissipation fan to operate according to the optimal operation mode under the current working condition.

[0100] In one embodiment, the first determination module is specifically configured to:

[0101] Based on the multi-objective optimization algorithm and the heat dissipation effect data, perform a Pareto optimal solution search between heat dissipation efficiency and energy consumption, and select the mode that is on the heat dissipation efficiency - energy consumption frontier curve and has the first ranking in terms of scoring as the optimal operation mode under this working condition.

[0102] In one embodiment, the method for constructing the heat dissipation efficiency - energy consumption frontier curve includes:

[0103] For a set of heat dissipation working condition characteristic data, plot the heat dissipation efficiency and energy consumption data corresponding to all operation modes in a two-dimensional coordinate system to form a scatter plot, where the abscissa represents energy consumption and the ordinate represents heat dissipation efficiency;

[0104] In the scatter plot, identify the data points with the lowest energy consumption at a given heat dissipation efficiency level, and use these data points to generate the heat dissipation efficiency - energy consumption frontier curve.

[0105] In one embodiment, the scoring is calculated based on the maximum heat dissipation efficiency of all operation modes, the minimum heat dissipation efficiency of all operation modes, the heat dissipation efficiency of the target operation mode, the weight coefficient of heat dissipation efficiency, the maximum energy consumption of all operation modes, the minimum energy consumption of all operation modes, the energy consumption of the target operation mode, and the weight coefficient of energy consumption.

[0106] In one embodiment, constructing the decision tree model based on the heat dissipation working condition characteristic data and the corresponding optimal operation mode includes:

[0107] S31. Calculate the information gain rate of each heat dissipation working condition characteristic, select the heat dissipation working condition characteristic with the largest information gain rate as the root node, and divide the data set into multiple subsets according to different values of the heat dissipation working condition characteristic with the largest gain rate;

[0108] S32. For each subset divided by the root node, repeat step S31, that is, recalculate the information gain rate of the remaining heat dissipation working condition characteristics on this subset, select the heat dissipation working condition characteristic with the largest information gain rate for division, and generate new subsets;

[0109] S33. Check whether each subset meets the stop condition. If it meets the stop condition, mark this subset as a leaf node and determine its corresponding optimal operation mode; if not, continue to recurse;

[0110] S34. Repeat steps S32 and S33 until all subsets meet the stop condition, and a decision tree model for determining the optimal operation mode is constructed.

[0111] In one embodiment, the step of using the decision tree model to determine the optimal operation mode of the device under the current working condition includes:

[0112] Obtain the heat dissipation condition characteristic data of the device under the current working condition;

[0113] Based on the heat dissipation condition characteristic data under the current working condition, determine the optimal operation mode of the device under the current working condition through the decision tree model.

[0114] Those of ordinary skill in the art can understand that all or part of the processes of implementing the methods in the above embodiments can be completed by instructing relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above methods. Among them, any reference to a memory, storage, database, or other medium provided in this application can include non-volatile and / or volatile memories. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or an external cache. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (SSRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0115] It should be noted that in this article, the term "including", "comprising", or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, device, article, or method including a series of elements not only includes those elements, but also includes other elements not explicitly listed, or further includes elements inherent to such a process, device, article, or method. Without further limitation, an element defined by the statement "including one..." does not exclude the existence of another identical element in the process, device, article, or method including that element.

[0116] The above are only the preferred embodiments of the present application, and do not limit the patent scope of the present application. Any equivalent structure or equivalent process transformation made by using the content of the specification and drawings of the present application, or directly or indirectly applied in other related technical fields, shall be equally included in the patent protection scope of the present application.

Claims

1. A two-way convection enhanced cooling fan control method, characterized in that: The method comprises: For each set of heat dissipation condition characteristic data, the bidirectional convection enhanced heat dissipation fan is controlled to execute various operation modes, and the heat dissipation effect data under various operation modes are recorded; wherein the heat dissipation condition characteristics include device temperature, device load, ambient temperature and ambient humidity; the heat dissipation effect data includes the temperature drop of the device within a preset time and the fan energy consumption; Determine the optimal operation mode corresponding to each set of heat dissipation condition characteristic data based on the heat dissipation effect data; Constructing a decision tree model based on the heat dissipation condition characteristic data and the corresponding optimal operation mode; wherein each leaf node of the decision tree corresponds to an optimal operation mode; Determine the optimal operation mode of the equipment under the current working conditions by using the decision tree model; The bidirectional convection enhanced cooling fan is controlled to operate in an optimal operating mode under the current operating condition.

2. The bidirectional convection enhanced cooling fan control method according to claim 1, characterized in that: The step of determining the optimal operation mode corresponding to each set of heat dissipation condition characteristic data based on the heat dissipation effect data comprises: Based on the multi-objective optimization algorithm and the heat dissipation effect data, a Pareto optimal solution search is performed between heat dissipation efficiency and energy consumption, and a mode that is on the frontier curve of heat dissipation efficiency-energy consumption and ranks first in score is selected as the optimal operating mode under this working condition.

3. The bidirectional convection enhanced cooling fan control method according to claim 2, characterized in that: The method for constructing the heat dissipation efficiency-energy consumption frontier curve includes: For a set of heat dissipation condition characteristic data, the heat dissipation efficiency and energy consumption data corresponding to all operation modes are plotted in a two-dimensional coordinate system to form a scatter plot, where the abscissa represents energy consumption and the ordinate represents heat dissipation efficiency; In the scatter plot, data points with the lowest energy consumption at a given heat dissipation efficiency level are identified, and a heat dissipation efficiency-energy consumption frontier curve is generated using the data points.

4. The bidirectional convection enhanced cooling fan control method according to claim 2, characterized in that: The score is calculated based on the maximum heat dissipation efficiency of all operating modes, the minimum heat dissipation efficiency of all operating modes, the heat dissipation efficiency of the target operating mode, the weight coefficient of the heat dissipation efficiency, the maximum energy consumption of all operating modes, the minimum energy consumption of all operating modes, the energy consumption of the target operating mode and the weight coefficient of the energy consumption.

5. The bidirectional convection enhanced cooling fan control method according to claim 1, characterized in that: The step of constructing a decision tree model based on the heat dissipation condition characteristic data and the corresponding optimal operation mode includes: S31, calculating the information gain rate of each heat dissipation condition feature, selecting the heat dissipation condition feature with the largest information gain rate as the root node, and dividing the data set into multiple subsets according to different values ​​of the heat dissipation condition feature with the largest gain rate; S32, for each subset divided from the root node, repeat step S31, that is, recalculate the information gain rate of the remaining heat dissipation condition features on the subset, select the heat dissipation condition feature with the largest information gain rate for division, and generate a new subset; S33, check whether each subset meets the stopping condition. If the stopping condition is met, mark the subset as a leaf node and determine the corresponding optimal operation mode; if not, continue recursion; S34, repeating step S32 and step S33 until all subsets meet the stopping condition, and constructing a decision tree model for determining the optimal operation mode.

6. The bidirectional convection enhanced cooling fan control method according to claim 1, characterized in that: The step of using the decision tree model to determine the optimal operating mode of the equipment under the current working conditions includes: Obtain the heat dissipation characteristic data of the equipment under the current working condition; Based on the heat dissipation operating condition characteristic data under the current operating condition, the optimal operating mode of the device under the current operating condition is determined through the decision tree model.

7. A two-way convection enhanced cooling fan control system, characterized in that: The system comprises: The first control module is used to control the two-way convection enhanced cooling fan to execute various operation modes for each set of cooling condition characteristic data, and record the cooling effect data under various operation modes; wherein the cooling condition characteristics include device temperature, device load, ambient temperature and ambient humidity; the cooling effect data includes the temperature drop of the device within a preset time and the fan energy consumption; A first determination module, configured to determine an optimal operation mode corresponding to each set of heat dissipation condition characteristic data based on the heat dissipation effect data; A decision tree construction module, used to construct a decision tree model based on the heat dissipation condition characteristic data and the corresponding optimal operation mode; wherein each leaf node of the decision tree corresponds to an optimal operation mode; A second determination module, used to determine the optimal operation mode of the equipment under the current working conditions by using the decision tree model; The second control module is used to control the bidirectional convection enhanced cooling fan to operate according to the optimal operating mode under the current working condition.

8. The two-way convection enhanced cooling fan control system according to claim 7, characterized in that: The first determining module is specifically used for: Based on the multi-objective optimization algorithm and the heat dissipation effect data, a Pareto optimal solution search is performed between heat dissipation efficiency and energy consumption, and a mode that is on the frontier curve of heat dissipation efficiency-energy consumption and ranks first in score is selected as the optimal operating mode under this working condition.

9. The two-way convection enhanced cooling fan control system according to claim 8, characterized in that: The method for constructing the heat dissipation efficiency-energy consumption frontier curve includes: For a set of heat dissipation condition characteristic data, the heat dissipation efficiency and energy consumption data corresponding to all operation modes are plotted in a two-dimensional coordinate system to form a scatter plot, where the abscissa represents energy consumption and the ordinate represents heat dissipation efficiency; In the scatter plot, data points with the lowest energy consumption at a given heat dissipation efficiency level are identified, and a heat dissipation efficiency-energy consumption frontier curve is generated using the data points.

10. The two-way convection enhanced cooling fan control system according to claim 8, characterized in that: The score is calculated based on the maximum heat dissipation efficiency of all operating modes, the minimum heat dissipation efficiency of all operating modes, the heat dissipation efficiency of the target operating mode, the weight coefficient of the heat dissipation efficiency, the maximum energy consumption of all operating modes, the minimum energy consumption of all operating modes, the energy consumption of the target operating mode and the weight coefficient of the energy consumption.

Images