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

Abstract

The present application discloses a heat dissipation device control method, apparatus, equipment and storage medium, which relate to the field of heat dissipation technology. Actual operation characteristic data of the heat dissipation device are collected at preset time intervals, and feature processing is performed on the actual operation characteristic data to obtain time series characteristic data of the heat dissipation device. When a preset number of time series characteristic data are continuously collected, predicted operation characteristic data are obtained based on the time series characteristic data of the heat dissipation device through a time series prediction model. The predicted operation characteristic data and the time series characteristic data are used as inputs of a reinforcement learning model to generate a control strategy for the heat dissipation device. The heat dissipation device is controlled according to the control strategy. Therefore, the technical problem of reduced heat dissipation effect on the equipment load can be solved, and the heat dissipation effect can be improved.

Description

Technical Field

[0001] The present application relates to the field of heat dissipation technology, and in particular to a heat dissipation device control method, device, equipment and storage medium. Background Art

[0002] The cooling system is a key component to ensure the stable operation of the equipment. The effectiveness of the cooling system's control strategy directly affects the hardware life, energy consumption and reliability of the equipment.

[0003] Current cooling systems employ a fixed-threshold control strategy, which sets a preset outlet temperature threshold. When the actual temperature exceeds the threshold, the fan automatically operates at full speed. However, in these technologies, the temperature of the device load changes dynamically. Fixed temperature thresholds can lead to overheating or underheating in response to varying load temperatures, resulting in reduced cooling effectiveness for the device load. Summary of the Invention

[0004] The present application provides a heat dissipation device control method, apparatus, device and storage medium to at least solve the problem of reduced heat dissipation effect on device loads in related technologies.

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

[0006] Collecting actual operating characteristic data of the heat dissipation device at preset time intervals;

[0007] Perform feature processing on the actual operation feature data to obtain time series feature data corresponding to the heat dissipation equipment;

[0008] When a preset number of time series feature data are continuously obtained, the time series feature data are input into the trained time series prediction model to output the predicted operation feature data of the heat dissipation equipment within a future preset time period;

[0009] Input the predicted operation feature data and time series feature data into the trained reinforcement learning model and output the control strategy of the cooling equipment;

[0010] Control the cooling equipment according to the control strategy.

[0011] The present application also provides a heat dissipation device control device, comprising:

[0012] A collection module, used to collect actual operating characteristic data of the heat dissipation device at preset time intervals;

[0013] A feature processing module is used to perform feature processing on actual operation feature data to obtain time series feature data corresponding to the heat dissipation device;

[0014] The first output module is configured to input the time series feature data into a trained time series prediction model when a preset number of time series feature data are continuously acquired, and output predicted operation feature data of the heat dissipation device within a preset time period in the future;

[0015] The second output module is used to input the predicted operation feature data and time series feature data into the trained reinforcement learning model and output the control strategy of the heat dissipation device;

[0016] The control module is used to control the heat dissipation device according to the control strategy.

[0017] The present application also provides an electronic device, comprising: a memory for storing a computer program; and a processor for implementing the steps of any of the above-mentioned heat dissipation device control methods when executing the computer program.

[0018] The present application also provides a computer-readable storage medium, in which a computer program is stored. When the computer program is executed by a processor, the steps of any of the above-mentioned heat dissipation device control methods are implemented.

[0019] The present application also provides a computer program product, including a computer program, which implements the steps of any of the above-mentioned heat dissipation device control methods when executed by a processor.

[0020] Through the present application, the actual operation characteristic data of the heat dissipation device is collected according to a preset time interval, and the actual operation characteristic data is feature processed to obtain the time series characteristic data of the heat dissipation device. When a preset number of time series characteristic data are continuously collected, the predicted operation characteristic data is obtained according to the time series characteristic data of the heat dissipation device through the time series prediction model. The predicted operation characteristic data and the time series characteristic data are used as inputs of the reinforcement learning model to generate a control strategy for the heat dissipation device, and the heat dissipation device is controlled according to the control strategy. Therefore, the technical problem of reduced heat dissipation effect on the equipment load can be solved, and the heat dissipation effect is improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] In order to more clearly illustrate the embodiments of the present application, the following is a brief introduction to the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0022] Figure 1 A schematic diagram of the system structure of a computer device provided in an embodiment of the present application;

[0023] Figure 2 A flow chart of a heat dissipation device control method provided in an embodiment of the present application;

[0024] Figure 3 A schematic diagram of the structure of a heat dissipation device control device provided in an embodiment of the present application;

[0025] Figure 4 This is a schematic diagram of the structure of the electronic device provided in this application. DETAILED DESCRIPTION

[0026] The following will be combined with the accompanying drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making any creative work are within the scope of protection of this application.

[0027] It should be noted that, in the description of this application, the terms "comprises," "includes," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or device comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or elements inherent to such process, method, article, or device. The terms "first," "second," etc., in this application are used to distinguish similar objects, and are not used to describe a particular order or sequence.

[0028] In order to solve the problem of reduced heat dissipation effect on equipment load in related technologies, the embodiments of the present application propose the following technical concepts: the inventors consider collecting actual operation characteristic data of the heat dissipation device at preset time intervals, consider performing feature processing on the actual operation characteristic data to obtain time series characteristic data of the heat dissipation device, and when continuously collecting a preset number of time series characteristic data, consider using a time series prediction model to predict and output the predicted operation characteristic data of the heat dissipation device, use the predicted operation characteristic data and the time series characteristic data as inputs of the reinforcement learning model, generate a control strategy for the heat dissipation device through the reinforcement learning model, control the heat dissipation device, and improve the heat dissipation effect.

[0029] In order to enable those skilled in the art to better understand the present application, the present application is further described in detail below with reference to the accompanying drawings and specific implementation methods.

[0030] In conjunction with the specific application environment architecture or specific hardware architecture on which the execution of the heat dissipation device control method depends, the specific application environment architecture or specific hardware architecture is described herein.

[0031] refer to Figure 1 , Figure 1 This is a schematic diagram of the system structure of the computer device provided in the embodiment of the present application. Figure 1As shown, the computer device includes: a receiving device 101, a processing device 102 and a display device 103.

[0032] It is understood that the structure illustrated in the embodiment of the present application does not constitute a specific limitation on the heat dissipation device control method. In other feasible embodiments of the present application, the above architecture may include more or fewer components than shown, or combine or split certain components, or arrange the components differently. The specific configuration can be determined based on the actual application scenario and is not limited here. Figure 1 The components shown can be implemented in hardware, software, or a combination of software and hardware.

[0033] In a specific implementation process, the receiving device 101 may be an input / output interface or a communication interface, and may collect actual operation characteristic data of the heat dissipation device.

[0034] The processing device 102 may generate a control strategy for the heat dissipation device.

[0035] The display device 103 can be used to display the control strategy of the above-mentioned heat dissipation device.

[0036] The display device may also be a touch screen display, which is used to receive user instructions while displaying the above-mentioned content to achieve operational interaction with the user.

[0037] It should be understood that the above-mentioned processing device can be implemented by a processor reading instructions in a memory and executing the instructions, or it can be implemented by a chip circuit.

[0038] In addition, the network architecture and business scenarios described in the embodiments of the present application are intended to more clearly illustrate the technical solutions of the embodiments of the present application, and do not constitute a limitation on the technical solutions provided in the embodiments of the present application. Ordinary technicians in this field can know that with the evolution of network architecture and the emergence of new business scenarios, the technical solutions provided in the embodiments of the present application are also applicable to similar technical problems.

[0039] Figure 2 A flow chart of a heat dissipation device control method provided in an embodiment of the present application is shown in FIG. Figure 2 As shown, an embodiment of the present application provides a heat dissipation device control method, which is described in detail as follows:

[0040] In this embodiment, devices to which the heat dissipation device control method is applied include but are not limited to: computers, servers, switches, and storage devices.

[0041] S201: Collecting actual operating characteristic data of a heat dissipation device at a preset time interval.

[0042] In this embodiment, actual operation characteristic data of the heat dissipation device is collected through a data collection process.

[0043] In this embodiment, different heat dissipation devices have different preset time intervals.

[0044] In this embodiment, the actual operating characteristic data includes, but is not limited to, air inlet temperature, air outlet temperature, whole machine power consumption, and fan speed.

[0045] S202: Perform feature processing on the actual operation feature data to obtain time series feature data corresponding to the heat dissipation device.

[0046] Specifically, the temperature difference is calculated based on the air outlet temperature and the air inlet temperature, the speed change rate is calculated based on the fan speed, and the power consumption change rate is calculated based on the overall power consumption. After normalization processing, time series feature data is obtained.

[0047] In this embodiment, the time series feature data includes but is not limited to air inlet temperature, air outlet temperature, whole machine power consumption, fan speed, temperature difference information, speed change rate, and power consumption change rate.

[0048] S203: When a preset number of time series feature data are continuously acquired, the time series feature data are input into a trained time series prediction model, and predicted operation feature data of the heat dissipation device within a future preset time period is output.

[0049] Specifically, a sliding time window is selected, a preset number of continuous time series feature data within the sliding time window are obtained, the time series feature data are input into a time series prediction model, and predicted operation feature data of the heat dissipation equipment within a future preset time period are output.

[0050] In this embodiment, the time series feature data 30 seconds before the current moment is collected.

[0051] In this embodiment, the predicted operating characteristic data is the predicted air outlet temperature at the current moment.

[0052] S204: Input the predicted operation feature data and the time series feature data into the trained reinforcement learning model, and output a control strategy for the heat dissipation device.

[0053] Specifically, the predicted air outlet temperature and time series feature data are input into the trained reinforcement learning model to output the control strategy of the heat dissipation equipment.

[0054] S205: Control the heat dissipation device according to the control strategy.

[0055] In this embodiment, the control strategy includes, but is not limited to, a speed adjustment strategy for the heat dissipation device, a heat dissipation duration strategy, and the number of enabled heat dissipation devices.

[0056] It can be seen from the above embodiments that the actual operation characteristic data of the heat dissipation device is collected at preset time intervals, and the actual operation characteristic data is subjected to feature processing to obtain time series characteristic data of the heat dissipation device. When a preset number of time series characteristic data are continuously collected, predicted operation characteristic data is obtained based on the time series characteristic data of the heat dissipation device through a time series prediction model. The predicted operation characteristic data and the time series characteristic data are used as inputs of the reinforcement learning model to generate a control strategy for the heat dissipation device. The heat dissipation device is controlled according to the control strategy, thereby improving the heat dissipation effect.

[0057] In one embodiment of the present application, step S202 includes:

[0058] S2021: Obtain the air outlet temperature and air inlet temperature in the actual operation characteristic data, and calculate the temperature difference information based on the air outlet temperature and the air inlet temperature.

[0059] In this embodiment, the air outlet temperature is recorded as .

[0060] In this embodiment, the air inlet temperature is recorded as .

[0061] Specifically, the calculation method of temperature difference information is:

[0062]

[0063] Where, represents the temperature difference at time t-1; represents the outlet temperature at time t-1; Indicates the air inlet temperature at time t-1.

[0064] S2022: Obtain the fan speed from the actual operation characteristic data, and calculate the speed change rate based on the fan speed.

[0065] In this embodiment, the rotational speed change rate is an indicator of the degree of fluctuation of the fan in a time series, and is the standard deviation of the fan rotational speed over a period of time.

[0066] In this embodiment, the speed change rate is recorded as .

[0067] In this embodiment, if the speed change rate is large, it means that the fan speed is frequently adjusted; if the speed change rate is small, it means that the fan speed is stable.

[0068] In this embodiment, a window length of 5 time steps is selected, and the standard deviation of the fan speed within the window length is recorded.

[0069] In this embodiment, five time steps cover the feedback cycle after the fan speed is adjusted.

[0070] In this embodiment, if the fan speed fluctuation is greater than 200 rpm and the adjustment frequency is greater than 3 times per minute, it is determined to be a high-frequency oscillation, and smoothing control is triggered to limit the step size of a single adjustment.

[0071] Among them, the formula for smooth control speed adjustment is:

[0072]

[0073] Where, Indicates the adjusted speed; Indicates the speed before adjustment.

[0074] S2023: Obtain the power consumption of the entire machine from the actual operation characteristic data, and calculate the power consumption change rate based on the power consumption of the entire machine.

[0075] In this embodiment, the calculation formula for the power consumption change rate is:

[0076]

[0077] Where, Indicates the rate of change of power consumption; Indicates the current power consumption of the entire device; Represents the interval with time T The power consumption of the whole device during the sampling interval; Indicates the sampling interval, which is 1 second.

[0078] In this embodiment, when it is detected that the power consumption change rate is greater than 100 W / s, it is determined to be a sudden load increase, and the time series prediction model is triggered to predict the temperature change in the next 30 seconds.

[0079] In this embodiment, when the standard deviation of the power consumption change rate, that is, the load fluctuation rate, is greater than 50W / When the time window is shortened, it is adjusted from 30 seconds to 20 seconds.

[0080] S2024: Normalize the temperature difference information, the rotation speed change rate, the power consumption change rate, and the actual operation characteristic data to obtain time series characteristic data of the heat dissipation device.

[0081] In this embodiment, the minimum and maximum normalization method is used to normalize the data.

[0082] The normalization formula is:

[0083]

[0084] Where, represents unnormalized data; represents the minimum value of the unnormalized data; represents the maximum value of the unnormalized data; Represents the normalized data.

[0085] It can be seen from the above embodiments that by obtaining the outlet temperature and the inlet temperature, the temperature difference information is calculated, by obtaining the fan speed, the speed change rate is calculated, by obtaining the power consumption of the entire machine, the power consumption change rate is calculated, the temperature difference information, the speed change rate, the power consumption change rate and the actual operation characteristic data are normalized, and the time series characteristic data is constructed as the input of the time series prediction model, thereby avoiding the single source of the operation characteristic data of the heat dissipation equipment.

[0086] In one embodiment of the present application, before step S204, the method further includes:

[0087] S301: Obtaining a training set state vector of predicted operation feature data and time series feature data of a heat dissipation device.

[0088] In this embodiment, a deep Q-network algorithm is used as the reinforcement learning algorithm.

[0089] Among them, the deep Q network is a model-free reinforcement learning algorithm that predicts the expected reward of taking a certain action in a certain state by learning a state-action-value function.

[0090] S302: Constructing a heat dissipation action according to the predicted operation feature data of the heat dissipation device and the training set state vector of the time series feature data.

[0091] In this embodiment, the heat dissipation action is the adjustment step of the fan speed.

[0092] Among them, the adjustment range is ±5%~±20%.

[0093] Specifically, the adjustment step size of the fan speed is discretized.

[0094] Exemplarily, the adjustment steps are divided into -20%, -15%, -10%, -5%, 0, +5%, +10%, +15% and +20%.

[0095] S303: Execute a heat dissipation action, and generate a reward value corresponding to the heat dissipation action according to the reward function.

[0096] In this embodiment, the reward function is the feedback provided by the actions taken by the agent in different states.

[0097] In this embodiment, the formula for generating the reward value corresponding to the heat dissipation action according to the reward function is:

[0098]

[0099] Where, Represents the reward value; Indicates the predicted outlet temperature; Indicates the maximum temperature threshold for safe operation of the server; Indicates the adjusted speed; Indicates the speed before adjustment; Indicates the adjusted power consumption; Indicates the power consumption before adjustment.

[0100] S304: storing the training set state vector of the predicted operation feature data and time series feature data of the heat dissipation device, the heat dissipation action, and the reward value corresponding to the heat dissipation action into the experience pool of the reinforcement learning model.

[0101] Specifically, the heat dissipation action is optimized according to the size of the reward value, which is used as experience for the next round of reinforcement learning to optimize the heat dissipation action for the next round.

[0102] S305: Update the state vector, heat dissipation action, and reward value in the experience pool according to a preset number of iterations to generate updated reinforcement learning model parameters.

[0103] Specifically, iterative learning is performed according to a preset number of iterations to obtain the heat dissipation action with the maximum reward value, and to obtain updated reinforcement learning model parameters.

[0104] S306: Determine the reinforcement learning model according to the updated reinforcement learning model parameters.

[0105] Specifically, if the updated reinforcement learning model parameters satisfy the training stop strategy of the reinforcement learning, the reinforcement learning model is determined according to the reinforcement learning model parameters.

[0106] From the above embodiments, it can be seen that by obtaining the state vector of the training set, constructing the heat dissipation action, calculating the reward value for performing the heat dissipation action under the condition of the training set state vector, storing the state vector, heat dissipation action and reward value in the experience pool of the reinforcement learning model, iterative learning, and training to obtain the reinforcement learning model, and formulating the heat dissipation strategy through the reinforcement learning model to improve the accuracy of heat dissipation.

[0107] In one embodiment of the present application, after step S306, the following steps are further included:

[0108] S307: Obtain device specification information of the heat dissipation device.

[0109] In this embodiment, the device specification information includes but is not limited to the number of running devices, device power consumption, and device type.

[0110] S308: Determine the heat dissipation type according to the device specification information of the heat dissipation device.

[0111] In this embodiment, the heat dissipation type includes but is not limited to a high heat dissipation requirement type, an energy-saving type, and a low mechanical stress type.

[0112] S309: Adjust the temperature deviation penalty weight, the speed change penalty weight, and the power consumption fluctuation penalty weight in the reward function according to the heat dissipation type to adapt to the heat dissipation type.

[0113] In this embodiment, the temperature deviation penalty is a measure of the degree of deviation between the predicted temperature and the set safety temperature.

[0114] In this embodiment, the speed variation penalty suppresses frequent fluctuations in the fan speed, thereby reducing mechanical loss and noise.

[0115] In this embodiment, the power consumption fluctuation penalty is the additional power consumption caused by the rotation speed adjustment.

[0116] In this embodiment, the temperature deviation penalty is a high priority target, the speed change penalty is a medium priority target, and the power consumption fluctuation penalty is a low priority target.

[0117] For example, for a device with low mechanical stress, the rotation speed change penalty weight is increased and the fan rotation speed is reduced to adapt to the device.

[0118] From the above embodiment, it can be seen that by obtaining the device specification information of the heat dissipation device, the heat dissipation type is determined, and the temperature deviation penalty weight, speed change penalty weight, and power consumption fluctuation penalty weight in the reward function are adjusted according to different heat dissipation types to adapt to different types of heat dissipation devices, so as to facilitate the reinforcement learning model to generate a more accurate heat dissipation strategy.

[0119] In one embodiment of the present application, before step S203, the following steps are further included:

[0120] S401: Obtain a training dataset of time series feature data of a heat dissipation device.

[0121] In this embodiment, the training data set includes but is not limited to the air outlet temperature at historical moments, the air inlet temperature at historical moments, the fan speed at historical moments, and the whole machine power consumption at historical moments.

[0122] S402: Create a time series prediction model for heat dissipation equipment.

[0123] In this embodiment, the time series prediction model includes three long short-term memory network layers and one fully connected layer.

[0124] S403: Input the training data set of the time series feature data of the heat dissipation device into the long short-term memory network layer of the time series prediction model, and output the hidden features.

[0125] In this embodiment, the first LSTM layer includes 64 neurons, and the return parameter is true. The second LSTM layer includes 32 neurons, and the return parameter is true. The third LSTM layer includes 16 neurons, and the return parameter is false.

[0126] In this embodiment, the return parameter is recorded as return_sequences.

[0127] Among them, if return_sequences=true, it means that the return parameter returns the output of each time step; if return_sequences=false, it means that the return parameter only returns the output of the last time step.

[0128] S404: Input the hidden features into the fully connected layer of the time series prediction model, and output the predicted running feature data corresponding to the training data set.

[0129] In this embodiment, the fully connected layer includes one neuron.

[0130] Specifically, the output of the fully connected layer is denormalized to obtain the predicted temperature value of the air outlet

[0131] S405: If the predicted operating characteristic data corresponding to the training data set meets the preset early stopping strategy, a time series prediction model is determined according to the predicted operating characteristic data corresponding to the training data set.

[0132] Specifically, the mean square error of the validation dataset is calculated. If the mean square error of the validation dataset does not decrease for 10 consecutive rounds, the model parameters determined by the training dataset meet the early stopping strategy. The corresponding model parameters are obtained based on the predicted running feature data output by the model, and the model parameters are determined as the model parameters of the trained time series prediction model.

[0133] In this embodiment, an edge-cloud hierarchical control architecture is adopted, the time series prediction model is run through the edge node, the data of the time series prediction model is aggregated through the cloud, and the update strategy is issued through the cloud.

[0134] From the above embodiments, it can be seen that by obtaining a training data set, a time series prediction model for the heat dissipation equipment is created, the training data set is input into the long short-term memory network layer, the hidden features are output, and the predicted operation feature data corresponding to the training data set is output through the fully connected layer according to the hidden features. It is judged whether the predicted operation feature data meets the early stopping strategy to determine the time series prediction model. The operation feature data is predicted in advance through the time series prediction model to provide a data basis for formulating the heat dissipation strategy.

[0135] In one embodiment of the present application, after step S405, the following steps are further included:

[0136] S406: If the predicted operation feature data corresponding to the training data set does not satisfy the preset early stopping strategy, the initial learning rate and number of iterations of the optimizer are set.

[0137] In this embodiment, the Adam optimizer is selected as the optimizer, the initial learning rate is set to 0.001, and the number of iterations is set to 100.

[0138] S407: Adjusting the model parameters of the time series prediction model according to the initial learning rate and the number of iterations to obtain at least one set of prediction model parameters.

[0139] In this example, the batch size is set to 32.

[0140] In this embodiment, the batch size is the number of samples used to calculate gradients and update model parameters.

[0141] S408: Apply at least one set of prediction model parameters to the time series prediction model to obtain at least one set of prediction operation characteristic data.

[0142] Specifically, the model parameters are applied to the time series prediction model, and the time series feature data is predicted by the time series prediction model to obtain the predicted operation feature data output corresponding to each model parameter.

[0143] S409: Calculate the mean square error of at least one set of predicted operation characteristic data and a validation data set of the heat dissipation device time series characteristic data.

[0144] In this embodiment, the formula for calculating the mean square error is:

[0145]

[0146] Where, represents the mean square error; Indicates the actual air outlet temperature; Indicates the predicted outlet temperature; N indicates the number of training samples.

[0147] S410: If the mean square error satisfies a preset early stopping strategy, a time series prediction model is determined according to at least one set of prediction operation feature data.

[0148] In this embodiment, the preset early stopping strategy is the mean square error of the validation set. When the mean square error of the validation set does not decrease for 10 consecutive rounds, training is automatically stopped.

[0149] Specifically, if the mean square error of the validation set does not decrease for 10 consecutive rounds, the model parameters corresponding to the minimum mean square error of the validation set are saved to determine the time series prediction model.

[0150] It can be seen from the above embodiment that if the predicted operating characteristic data does not meet the early stopping strategy, the initial learning rate and number of iterations are set, the model parameters are adjusted, and the predicted operating characteristic data output by the obtained multiple model parameters are used to calculate the mean square error of the validation data set. If the mean square error meets the early stopping strategy, it is determined to be a time series prediction model. By calculating the mean square error of the validation data set, overfitting of the training model is prevented.

[0151] In one embodiment of the present application, before step S203, the following steps are further included:

[0152] S501: Acquire a historical speed adjustment dataset and a historical temperature change delay dataset of a heat dissipation device.

[0153] In this embodiment, the historical temperature change delay is the temperature change delay caused by the heat dissipation device adjusting the rotation speed according to the rotation speed adjustment strategy.

[0154] S502: Establishing a mapping relationship between the rotation speed and the temperature change according to the historical rotation speed adjustment data set and the historical temperature change delay data set.

[0155] In this embodiment, a convolutional neural network is used to extract features from historical speed and temperature sequences, identify delay distributions corresponding to different speed change amplitudes, and establish a mapping relationship.

[0156] S503: Generate compensation data for the time series prediction model according to the mapping relationship between the rotation speed and the temperature change.

[0157] In this embodiment, the compensation data is used to compensate for the prediction of the air outlet temperature by the time series prediction model.

[0158] For example, when the speed changes by 20%, the delay of temperature drop is 8-12 seconds. By introducing a 10-second delay compensation time point, the prediction accuracy of the time series prediction model for the air outlet temperature is improved.

[0159] It can be seen from the above embodiments that by obtaining historical data sets of speed adjustment and temperature change delays, establishing a mapping relationship between speed and temperature change, and calculating the compensation of the time series prediction model based on the mapping relationship, the physical delay in temperature response after fan speed adjustment is avoided, thereby improving the accuracy of predicting the air outlet temperature.

[0160] In one embodiment of the present application, after step S205, the following steps are further included:

[0161] S206: Collecting operational characteristic data of the heat dissipation device after executing the control strategy.

[0162] Specifically, after executing the control strategy, the length of the time window is set, and the operating characteristic data of the heat dissipation device within the time window is collected.

[0163] S207: Generate a temperature curve of the heat dissipation device according to the operation characteristic data of the heat dissipation device after executing the control strategy.

[0164] In this embodiment, the horizontal axis of the temperature curve of the heat dissipation device is the sampling time in the time window, and the vertical axis is the temperature value of the heat dissipation device.

[0165] S208: If the temperature curve of the heat dissipation device exceeds a temperature curve threshold of the heat dissipation device, an alarm message is generated.

[0166] Specifically, the drawn temperature curve of the heat dissipation device is compared with the temperature curve threshold of the heat dissipation device, sampling points exceeding the temperature threshold are obtained, the temperature and status information of the corresponding sampling points are obtained, and the alarm information is recorded and generated.

[0167] From the above embodiments, it can be seen that by collecting the operating characteristic data of the heat dissipation device after executing the control strategy, a temperature curve is drawn, and the drawn temperature curve is compared with the temperature curve threshold of the heat dissipation device. An alarm message is generated for the part that exceeds the threshold, prompting the staff to manually adjust the heat dissipation strategy.

[0168] Through the description of the above implementation methods, those skilled in the art can clearly understand that the method according to the above embodiment can be implemented by means of software plus the necessary general hardware platform, and of course it can also be implemented by hardware, but in many cases the former is a better implementation method.

[0169] Figure 3 This is a schematic diagram of the structure of the heat dissipation device control device provided in the embodiment of the present application. Figure 3 As shown, an embodiment of the present application further provides a heat dissipation device control device 30 , including: an acquisition module 301 , a feature processing module 302 , a first output module 303 , a second output module 304 and a control module 305 .

[0170] The collection module 301 is used to collect actual operation characteristic data of the heat dissipation device at every preset time interval;

[0171] A feature processing module 302 is used to perform feature processing on the actual operation feature data to obtain time series feature data corresponding to the heat dissipation device;

[0172] The first output module 303 is configured to input the time series feature data into the trained time series prediction model when a preset number of time series feature data are continuously acquired, and output the predicted operation feature data of the heat dissipation device within a preset time period in the future;

[0173] The second output module 304 is used to input the predicted operation feature data and the time series feature data into the trained reinforcement learning model and output the control strategy of the heat dissipation device;

[0174] The control module 305 is used to control the heat dissipation device according to the control strategy.

[0175] In one embodiment of the present application, the feature processing module 302 includes:

[0176] The first acquisition unit is used to acquire the air outlet temperature and the air inlet temperature in the actual operation characteristic data, and calculate the temperature difference information according to the air outlet temperature and the air inlet temperature.

[0177] The second acquiring unit is configured to acquire the fan speed from the actual operation characteristic data, and calculate the speed change rate according to the fan speed.

[0178] The third acquiring unit is configured to acquire the whole machine power consumption from the actual operation characteristic data, and calculate the power consumption change rate according to the whole machine power consumption.

[0179] The normalization processing unit is used to normalize the temperature difference information, the speed change rate, the power consumption change rate and the actual operation characteristic data to obtain the time series characteristic data of the heat dissipation device.

[0180] In one embodiment of the present application, the heat dissipation device control device 30 further includes:

[0181] The first acquisition module is used to obtain the training set state vector of the predicted operation feature data and time series feature data of the heat dissipation device.

[0182] A building module is used to build a cooling action based on the predicted operating feature data of the cooling device and the training set state vector of the time series feature data.

[0183] The execution module is used to execute the heat dissipation action and generate a reward value corresponding to the heat dissipation action according to the reward function.

[0184] The storage module is used to store the training set state vector of the predicted operation feature data and time series feature data of the heat dissipation device, the heat dissipation action and the reward value corresponding to the heat dissipation action into the experience pool of the reinforcement learning model.

[0185] The first generation module is used to update the state vector, heat dissipation action and reward value in the experience pool according to a preset number of iterations to generate updated reinforcement learning model parameters.

[0186] The first determination module is used to determine the reinforcement learning model according to the updated reinforcement learning model parameters.

[0187] In one embodiment of the present application, the heat dissipation device control device 30 further includes:

[0188] The second acquisition module is used to obtain device specification information of the heat dissipation device.

[0189] The second determining module is configured to determine the heat dissipation type according to the device specification information of the heat dissipation device.

[0190] The first adjustment module is used to adjust the temperature deviation penalty weight, the speed change penalty weight, and the power consumption fluctuation penalty weight in the reward function according to the heat dissipation type to adapt to the heat dissipation type.

[0191] In one embodiment of the present application, the heat dissipation device control device 30 further includes:

[0192] The third acquisition module is used to obtain a training data set of time series feature data of the heat dissipation device.

[0193] Create a module for creating a time series forecasting model for cooling equipment.

[0194] The third output module is used to input the training data set of the time series feature data of the heat dissipation equipment into the long short-term memory network layer of the time series prediction model, and output hidden features.

[0195] The fourth output module is used to input the hidden features into the fully connected layer of the time series prediction model and output the predicted running feature data corresponding to the training data set.

[0196] The third determination module is configured to determine a time series prediction model based on the predicted operation characteristic data corresponding to the training data set if the predicted operation characteristic data corresponding to the training data set meets a preset early stopping strategy.

[0197] In one embodiment of the present application, the heat dissipation device control device 30 further includes:

[0198] The setting module is used to set the initial learning rate and number of iterations of the optimizer if the predicted running feature data corresponding to the training dataset does not meet the preset early stopping strategy.

[0199] The second adjustment module is used to adjust the model parameters of the time series prediction model according to the initial learning rate and the number of iterations to obtain at least one set of prediction model parameters.

[0200] The application module is used to apply at least one set of prediction model parameters to the time series prediction model to obtain at least one set of prediction operation characteristic data.

[0201] The calculation module is used to calculate the mean square error of at least one set of prediction operation characteristic data and the validation data set of the heat dissipation device time series characteristic data.

[0202] The fourth determination module is configured to determine a time series prediction model based on at least one set of prediction operation feature data if the mean square error satisfies a preset early stopping strategy.

[0203] In one embodiment of the present application, the heat dissipation device control device 30 further includes:

[0204] The fourth acquisition module is used to acquire a historical speed adjustment data set and a historical temperature change delay data set of the heat dissipation device.

[0205] A module is established to establish a mapping relationship between the rotation speed and the temperature change according to the historical rotation speed adjustment data set and the historical temperature change delay data set.

[0206] The second generating module is used to generate compensation data of the time series prediction model according to the mapping relationship between the rotation speed and the temperature change.

[0207] For the description of the features in the embodiment corresponding to the heat dissipation device control apparatus, reference can be made to the relevant description of the embodiment corresponding to the heat dissipation device control method, which will not be repeated here.

[0208] Figure 4 This is a schematic diagram of the structure of the electronic device provided in this application. Figure 4 As shown, the electronic device 40 provided in this embodiment includes: at least one processor 401 and a memory 402. Optionally, the electronic device 40 further includes a communication component 403. The processor 401, the memory 402 and the communication component 403 are connected via a bus.

[0209] In a specific implementation process, at least one processor 401 executes the computer-executable instructions stored in the memory 402 , so that the at least one processor 401 executes the above-mentioned heat dissipation device control method embodiment.

[0210] The specific implementation process of the processor 401 can be found in the above method embodiment. Its implementation principle and technical effects are similar and will not be repeated here in this embodiment.

[0211] In the above embodiments, it should be understood that the processor may be a central processing unit (CPU), other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), etc. A general-purpose processor may be a microprocessor or any conventional processor. The steps of the method disclosed in the application may be directly executed by a hardware processor or by a combination of hardware and software modules within the processor.

[0212] The memory may include random access memory (RAM) and may also include non-volatile memory (NVM), such as at least one disk storage.

[0213] A bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus. Buses can be categorized as address buses, data buses, and control buses. For ease of illustration, the buses in the drawings of this application are not limited to just one bus or just one type of bus.

[0214] An embodiment of the present application further provides a computer-readable storage medium, in which a computer program is stored. The computer program is configured to execute the steps of any of the above-mentioned heat dissipation device control method embodiments when running.

[0215] In an exemplary embodiment, the computer-readable storage medium may include, but is not limited to, various media that can store computer programs, such as a USB flash drive, a read-only memory (ROM), a random access memory (RAM), a mobile hard disk, a magnetic disk, or an optical disk.

[0216] An embodiment of the present application further provides a computer program product, which includes a computer program. When the computer program is executed by a processor, the steps in any of the above-mentioned heat dissipation device control method embodiments are implemented.

[0217] An embodiment of the present application further provides another computer program product, including a non-volatile computer-readable storage medium, wherein the non-volatile computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of any of the above-mentioned heat dissipation device control method embodiments are implemented.

[0218] Professionals may further appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the above description has generally described the components and steps of each example according to their functions. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professionals and technicians may use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0219] The above is a detailed introduction to a heat dissipation device control method, device, equipment and storage medium provided by the present application. Specific examples are used herein to illustrate the principles and implementation methods of the present application. The description of the above embodiments is only used to help understand the method and core ideas of the present application. It should be pointed out that for ordinary technicians in this technical field, without departing from the principles of the present application, several improvements and modifications can be made to the present application, and these improvements and modifications also fall within the scope of protection of the claims of the present application.

Claims

1. A heat dissipation device control method, characterized in that: include: Collecting actual operating characteristic data of the heat dissipation device at preset time intervals; Obtaining the air outlet temperature and the air inlet temperature in the actual operation characteristic data, and calculating temperature difference information based on the air outlet temperature and the air inlet temperature; Obtaining the fan speed in the actual operation characteristic data, and calculating the speed change rate based on the fan speed; Obtaining the entire machine power consumption from the actual operation characteristic data, and calculating the power consumption change rate based on the entire machine power consumption; Normalizing the temperature difference information, the rotational speed change rate, the power consumption change rate, and the actual operation characteristic data to obtain time series characteristic data of the heat dissipation device; When a preset number of time series feature data are continuously acquired, the time series feature data are input into the trained time series prediction model to output predicted operation feature data of the heat dissipation device within a future preset time period; Inputting the predicted operation feature data and the time series feature data into a trained reinforcement learning model, and outputting a control strategy for the heat dissipation device; controlling the heat dissipation device according to the control strategy; Before inputting the time series feature data into the trained time series prediction model, the method further includes: Obtaining a training data set of time series feature data of the heat dissipation device; Creating a time series prediction model for the heat dissipation device; Inputting the training data set of the time series feature data of the heat dissipation device into the long short-term memory network layer of the time series prediction model, and outputting hidden features; Inputting the hidden features into the fully connected layer of the time series prediction model, and outputting the predicted running feature data corresponding to the training data set; If the predicted operating characteristic data corresponding to the training data set satisfies a preset early stopping strategy, a time series prediction model is determined based on the predicted operating characteristic data corresponding to the training data set; wherein the preset early stopping strategy indicates that the mean square error between the predicted operating characteristic data corresponding to the training data set and the validation data set of the time series characteristic data has not decreased for consecutive preset rounds; Before inputting the time series feature data into the trained time series prediction model, the method further includes: Obtain historical speed adjustment data sets and historical temperature change delay data sets of the heat dissipation equipment; Establishing a mapping relationship between speed and temperature change according to the historical speed adjustment data set and the historical temperature change delay data set; Compensation data of a time series prediction model is generated according to the mapping relationship between the rotation speed and the temperature change.

2. The heat dissipation device control method according to claim 1, characterized in that: Before inputting the predicted operation feature data and the time series feature data into the trained reinforcement learning model, the method further includes: Obtaining a training set state vector of the predicted operation characteristic data and time series characteristic data of the heat dissipation device; Constructing a heat dissipation action based on the training set state vector of the predicted operation characteristic data and time series characteristic data of the heat dissipation device; Executing the heat dissipation action, and generating a reward value corresponding to the heat dissipation action according to a reward function; Storing the predicted operation feature data and the training set state vector of the time series feature data of the heat dissipation device, the heat dissipation action, and the reward value corresponding to the heat dissipation action in an experience pool of a reinforcement learning model; Update the state vector, heat dissipation action, and reward value in the experience pool according to a preset number of iterations to generate updated reinforcement learning model parameters; A reinforcement learning model is determined according to the updated reinforcement learning model parameters.

3. The heat dissipation device control method according to claim 2, characterized in that: The formula for generating the reward value corresponding to the heat dissipation action according to the reward function is: Where, Represents the reward value; Indicates the predicted outlet temperature; Indicates the maximum temperature threshold for safe operation of the server; Indicates the adjusted speed; Indicates the speed before adjustment; Indicates the adjusted power consumption; Indicates the power consumption before adjustment.

4. The heat dissipation device control method according to claim 2, characterized in that: After determining the reinforcement learning model according to the updated reinforcement learning model parameters, the method further includes: Obtain device specification information of heat dissipation equipment; Determining a heat dissipation type according to device specification information of the heat dissipation device; The temperature deviation penalty weight, the rotation speed change penalty weight, and the power consumption fluctuation penalty weight in the reward function are adjusted according to the heat dissipation type to adapt to the heat dissipation type.

5. The heat dissipation device control method according to claim 1, characterized in that: After determining the time series prediction model according to the predicted operation characteristic data corresponding to the training data set, the method further includes: If the predicted running feature data corresponding to the training data set does not meet the preset early stopping strategy, the initial learning rate and number of iterations of the optimizer are set; Adjusting the model parameters of the time series prediction model according to the initial learning rate and the number of iterations to obtain at least one set of prediction model parameters; Applying the at least one set of prediction model parameters to a time series prediction model to obtain at least one set of prediction operation characteristic data; Calculating a mean square error between the at least one set of predicted operating characteristic data and a validation data set of the heat dissipation device time series characteristic data; If the mean square error satisfies the preset early stopping strategy, a time series prediction model is determined according to the at least one set of prediction operation feature data.

6. A heat dissipation device control device, characterized in that: include: A collection module, used to collect actual operating characteristic data of the heat dissipation device at preset time intervals; a feature processing module, configured to obtain the air outlet temperature and the air inlet temperature in the actual operation feature data, and calculate temperature difference information based on the air outlet temperature and the air inlet temperature; obtain the fan speed in the actual operation feature data, and calculate the speed change rate based on the fan speed; obtain the whole machine power consumption in the actual operation feature data, and calculate the power consumption change rate based on the whole machine power consumption; and normalize the temperature difference information, the speed change rate, the power consumption change rate, and the actual operation feature data to obtain time series feature data of the heat dissipation device; A first output module is configured to input a preset number of time series feature data into a trained time series prediction model when the preset number of time series feature data is continuously acquired, and output predicted operation feature data of the heat dissipation device within a preset time period in the future; A second output module is configured to input the predicted operation characteristic data and the time series characteristic data into a trained reinforcement learning model, and output a control strategy for the heat dissipation device; A control module for controlling the heat dissipation device according to the control strategy; A third acquisition module is used to obtain a training data set of time series feature data of the heat dissipation device; A creation module for creating a time series prediction model for the heat dissipation device; A third output module is configured to input the training data set of the time series feature data of the heat dissipation device into the long short-term memory network layer of the time series prediction model and output hidden features; a fourth output module, configured to input the hidden features into a fully connected layer of the time series prediction model and output predicted running feature data corresponding to the training data set; A third determination module is configured to determine a time series prediction model based on the predicted operating characteristic data corresponding to the training data set if the predicted operating characteristic data corresponding to the training data set satisfies a preset early stopping strategy; wherein the preset early stopping strategy indicates that the mean square error between the predicted operating characteristic data corresponding to the training data set and a validation data set of the time series characteristic data has not decreased for consecutive preset rounds; A fourth acquisition module is used to acquire a historical speed adjustment data set and a historical temperature change delay data set of the heat dissipation device; An establishing module, configured to establish a mapping relationship between the rotational speed and the temperature change according to the historical rotational speed adjustment data set and the historical temperature change delay data set; The second generating module is used to generate compensation data of the time series prediction model according to the mapping relationship between the rotation speed and the temperature change.

7. An electronic device, characterized in that: include: Memory for storing computer programs; A processor, configured to implement the steps of the heat dissipation device control method according to any one of claims 1 to 5 when executing the computer program.

8. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, wherein the computer program, when executed by a processor, implements the steps of the heat dissipation device control method according to any one of claims 1 to 5.

Images