Method and device for alleviating hotspot areas in computer rooms

Abstract

The present application discloses a method and device for alleviating hotspot areas in a computer room. The method includes: obtaining multiple groups of first operating parameters of multiple cabinets in the computer room collected by the target device; inputting the multiple groups of first operating parameters into a pre-trained hotspot area prediction model to obtain hotspot area prediction results output by the hotspot area prediction model, wherein the hotspot area prediction results are used to reflect the probability of each cabinet being located in the hotspot area; determining the moving path of the target device based on the hotspot area prediction results, and controlling the target device to patrol multiple cabinets according to the moving path, wherein the target device is configured with a fan unit, and the area where the cabinet with a greater probability of being located in the hotspot area is located has a higher priority in the moving path; adjusting the temperature parameters of the air conditioner in the area where the multiple cabinets are located based on the hotspot area prediction results. The present application solves the technical problem in the related art that the temperature of the computer room cannot be controlled intelligently and accurately, resulting in frequent local hotspots.

Description

Technical Field

[0001] The present application relates to the field of environmental monitoring and control technology, and in particular to a method and device for alleviating hotspot areas in a computer room. Background Art

[0002] At present, communication rooms use a large number of new high-heat-density rack-mounted servers, which makes the cold and hot air flows in the communication rooms turbulent, resulting in frequent local hot spots in the cabinets.

[0003] Usually, the related technology uses axial flow fans and mobile air conditioners to cool down the local hot spots in the computer room. However, the disadvantage of this method is that the air volume has been reduced when the cold air from the axial flow fans and mobile air conditioners reaches the remote cabinets. In addition, since there are many pipes under the elevated floor, there will be more obstacles to the transmission of cold air. Therefore, in actual application, this method consumes a lot of energy, has a poor cooling effect, and cannot achieve intelligent control of the temperature of the computer room.

[0004] To address the above-mentioned problems, no effective solution has been proposed yet. Summary of the invention

[0005] The embodiments of the present application provide a method and device for alleviating hotspot areas in a computer room, so as to at least solve the technical problem in the related art that the temperature of the computer room cannot be intelligently and accurately controlled, resulting in frequent occurrence of local hotspots.

[0006] According to one aspect of an embodiment of the present application, a method for alleviating hot spot areas in a computer room is provided, comprising: obtaining multiple groups of first operating parameters of multiple cabinets in the computer room collected by a target device; inputting the multiple groups of first operating parameters into a pre-trained hot spot area prediction model to obtain a hot spot area prediction result output by the hot spot area prediction model, wherein the hot spot area prediction result is used to reflect the probability of each cabinet being located in the hot spot area; determining a moving path of the target device based on the hot spot area prediction result, and controlling the target device to patrol multiple cabinets according to the moving path, wherein a fan unit is configured in the target device, and an area where a cabinet with a greater probability of being located in the hot spot area is located has a higher priority in the moving path; and adjusting the temperature parameters of air conditioners in the area where the multiple cabinets are located based on the hot spot area prediction result.

[0007] Optionally, the target device includes: an automatic guided vehicle, a micro control unit, a wireless communication unit, a sensor unit and a fan unit installed on the automatic guided vehicle, wherein the automatic guided vehicle is used to move according to a moving path; the sensor unit includes multiple sensors for collecting first operating parameters, and the first operating parameters include at least one of the following: cabinet front temperature, cabinet rear temperature, cabinet front wind speed, cabinet rear wind speed and cabinet load; the fan unit includes a matrix fan composed of multiple fans; the micro control unit is used to send multiple groups of first operating parameters to the power environment monitoring platform through the wireless communication unit, and receive control instructions sent by the power environment monitoring platform, and control the automatic guided vehicle and the fan unit to perform operations corresponding to the control instructions.

[0008] Optionally, in response to the temperature abnormality alarm information issued by the target cabinet, the target device is controlled to move to the area where the target cabinet is located, and a second operating parameter of the target cabinet is obtained; the second operating parameter is input into the hot spot area prediction model to obtain a first prediction probability output by the hot spot area prediction model; when the prediction probability is greater than a preset probability threshold, it is determined that the target cabinet is located in the hot spot area, the air supply volume of the fan unit is adjusted according to the temperature abnormality alarm information, and the temperature parameters of the air conditioner in the area where the target cabinet is located are adjusted according to the temperature abnormality alarm information.

[0009] Optionally, when the predicted probability is not greater than a preset probability threshold, it is determined that the target cabinet is not located in the hot spot area, and an abnormal prompt message is issued, where the abnormal prompt message is used to prompt that the temperature abnormality alarm message issued by the target cabinet is incorrect.

[0010] Optionally, a third operating parameter of the target device itself is obtained, wherein the third operating parameter includes at least one of the following: operating voltage, operating current and air output of the fan unit; when the third operating parameter does not meet a preset operating threshold range, the target device is controlled to turn off the fan unit and issue an alarm message, wherein the alarm message is used to prompt that the operating status of the target device is abnormal and requires maintenance.

[0011] Optionally, the training process of the hot spot area prediction model includes: determining a historical time period when a hot spot area exists in the computer room; obtaining multiple groups of fourth operating parameters of multiple cabinets in the historical time period from historical operating data, and using each group of fourth operating parameters as a group of training sample data, wherein the fourth operating parameters include at least one of the following: cabinet front temperature, cabinet rear temperature, cabinet front wind speed, cabinet rear wind speed and cabinet load; adding sample labels to each group of training sample data, and the sample labels are used to mark whether the cabinet corresponding to the training sample data is located in the hot spot area; training the initial model based on multiple groups of training sample data and sample labels to obtain a hot spot area prediction model, wherein the initial model is an XGBoost binary classification model.

[0012] Optionally, the initial model is trained based on multiple groups of training sample data and sample labels to obtain a hot spot area prediction model, including: for each group of training sample data, the training sample data is input into the XGBoost binary classification model to obtain a second prediction probability output by the XGBoost binary classification model; a target loss function is constructed based on the second prediction probability and the sample labels corresponding to the training sample data; multiple groups of training sample data are input into the XGBoost binary classification model in sequence for iterative training, and the model parameters of the XGBoost binary classification model are adjusted based on the target loss function to obtain the hot spot area prediction model.

[0013] According to another aspect of an embodiment of the present application, a device for alleviating hot spot areas in a computer room is also provided, including: an acquisition module, used to acquire multiple groups of first operating parameters of multiple cabinets in the computer room collected by a target device; a prediction module, used to input the multiple groups of first operating parameters into a pre-trained hot spot area prediction model, and obtain the hot spot area prediction results output by the hot spot area prediction model, wherein the hot spot area prediction results are used to reflect the probability of each cabinet being located in the hot spot area; a control module, used to determine a moving path of the target device based on the hot spot area prediction results, and control the target device to patrol multiple cabinets according to the moving path, wherein a fan unit is configured in the target device, and the area where the cabinet with a greater probability of being located in the hot spot area is located has a higher priority in the moving path; an adjustment module, used to adjust the temperature parameters of the air conditioners in the area where the multiple cabinets are located based on the hot spot area prediction results.

[0014] According to another aspect of an embodiment of the present application, a non-volatile storage medium is also provided, which includes a stored program, wherein the device where the non-volatile storage medium is located executes the above-mentioned method for alleviating hot spot areas in the computer room by running the program.

[0015] According to another aspect of an embodiment of the present application, an electronic device is further provided, comprising: a memory and a processor, wherein a computer program is stored in the memory, and the processor is configured to execute the above-mentioned method for alleviating hot spot areas in a computer room through the computer program.

[0016] In an embodiment of the present application, multiple groups of first operating parameters of multiple cabinets in a computer room collected by a target device are obtained; multiple groups of first operating parameters are input into a pre-trained hot spot prediction model to obtain hot spot prediction results output by the hot spot prediction model, wherein the hot spot prediction results are used to reflect the probability of each cabinet being located in a hot spot area; the moving path of the target device is determined based on the hot spot prediction results, and the target device is controlled to patrol multiple cabinets according to the moving path, wherein a fan unit is configured in the target device, and the cabinet with a greater probability of being located in a hot spot area has a higher priority in the moving path; the temperature parameters of the air conditioner in the area where the multiple cabinets are located are adjusted based on the hot spot prediction results. The operating data of multiple cabinets in the computer room intelligently collected by the target device ensures that the hot spot areas in the computer room can be accurately predicted later; the moving trajectory of the target device is determined by the hot spot prediction results, and the hot spot areas in the computer room are controlled in time; the temperature parameters of the air conditioner in the area where the multiple cabinets are located are adjusted by the hot spot prediction results, so as to realize intelligent control of the computer room temperature, thereby solving the technical problem that the computer room temperature cannot be intelligently and accurately controlled in the related technology, resulting in frequent local hot spots. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0018] Figure 1 It is a flow chart of a method for alleviating hotspot areas in a computer room according to an embodiment of the present application;

[0019] Figure 2 is a schematic diagram of an optional automatic guided vehicle according to an embodiment of the present application;

[0020] Figure 3 is a schematic diagram of an optional control module according to an embodiment of the present application;

[0021] Figure 4 is a schematic diagram of an optional trajectory of movement of a target device according to an embodiment of the present application;

[0022] Figure 5 It is a structural schematic diagram of a device for alleviating hot spot areas in a computer room according to an embodiment of the present application. DETAILED DESCRIPTION

[0023] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present application.

[0024] It should be noted that the terms "first", "second", etc. in the specification, claims and drawings of the present application are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0025] In order to better understand the embodiments of the present application, some nouns or terms that appear in the description of the embodiments of the present application are first translated and explained as follows:

[0026] Power environment monitoring system: The power environment monitoring system is designed for the equipment characteristics and working environment of various communication stations (including communication rooms, base stations, branch stations, module stations, etc.), and realizes functions such as "telemetry, telesignaling, remote control, and remote adjustment" for intelligent and non-intelligent equipment such as communication power supplies, battery packs, UPS (Uninterruptible Power Supply), generators, air conditioners, and environmental quantities such as temperature and humidity, smoke, ground water, and access control in the stations. The system makes full use of various transmission channel resources provided by communication transmission equipment, not only can it successfully realize multi-level network management, making unmanned stations a reality, but also uses channel resources efficiently, saving users a lot of channel resource investment and operation and maintenance investment, and reducing user operating costs; the monitoring center software can realize the operation of the Chinese graphical human-machine interface, and can realize full-parameter and all-round monitoring of all stations, greatly improving the user's maintenance and management efficiency.

[0027] XGBoost (eXtreme Gradient Boosting) algorithm: XGBoost algorithm is an efficient gradient boosting decision tree algorithm. It is an improvement on the original GBDT (Gradient Boosting Decision Tree) algorithm, which greatly improves the model effect. XGBoost algorithm realizes the learning of weak learners by optimizing the structural loss function, and uses the distributed forward additive model to integrate multiple weak learners into a strong learner. That is, multiple trees are used to make decisions together, and the result of each tree is the difference between the target value and the prediction results of all previous trees, and the results of all trees are accumulated to obtain the final result, so as to improve the effect of the entire model. In addition, XGBoost is integrated by many CART (Classification And Regression Tree), where CART is a general term for classification decision trees and regression decision trees, so XGBoost algorithm can also handle classification and regression problems.

[0028] Example 1

[0029] Since the computer room uses a large number of new high-heat-density rack-mounted servers, the hot and cold air flows in the computer room become disordered, resulting in frequent local hot spots in the cabinets.

[0030] Usually, the related technology uses axial flow fans and mobile air conditioners to cool down local hot spots in the computer room. However, the disadvantage of this method is that the air volume has been reduced when the cold air generated by the axial flow fans and mobile air conditioners reaches the remote cabinets. At the same time, because the elevated floor is full of pipes, there are more obstacles to the transmission of cold air. Therefore, in actual application, this method consumes more energy, has a poor cooling effect, and cannot intelligently control the temperature of the computer room.

[0031] In order to solve the above problems, the embodiment of the present application provides a method for alleviating hotspot areas in a computer room. The operation data of multiple cabinets in the computer room are intelligently collected by the target device to ensure that the hotspot areas in the computer room can be accurately predicted later; the movement trajectory of the target device is determined by the hotspot area prediction results, and the hotspot areas in the computer room are controlled in time; the temperature parameters of the air conditioners in the areas where the multiple cabinets are located are adjusted by the hotspot area prediction results, and the temperature of the computer room is intelligently controlled, thereby alleviating local hotspots in the computer room and improving the safety of the computer room environment.

[0032] It should be noted that the steps shown in the flowcharts of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and that, although a logical order is shown in the flowcharts, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0033] Figure 1is a flow chart of an optional method for alleviating hotspot areas in a computer room according to an embodiment of the present application, such as Figure 1 As shown, the method at least includes steps S102-S108, wherein:

[0034] Step S102: acquiring multiple groups of first operating parameters of multiple cabinets in a computer room collected by a target device.

[0035] Optionally, the target device includes: an automatic guided vehicle, a micro control unit, a wireless communication unit, a sensor unit and a fan unit installed on the automatic guided vehicle, wherein the automatic guided vehicle is used to move according to a moving path; the sensor unit includes multiple sensors for collecting first operating parameters, and the first operating parameters include at least one of the following: cabinet front temperature, cabinet rear temperature, cabinet front wind speed, cabinet rear wind speed and cabinet load; the fan unit includes a matrix fan composed of multiple fans; the micro control unit is used to send multiple groups of first operating parameters to the power environment monitoring platform through the wireless communication unit, and receive control instructions sent by the power environment monitoring platform, and control the automatic guided vehicle and the fan unit to perform operations corresponding to the control instructions.

[0036] Specifically, when the target device is moving along the determined moving trajectory, the temperature and humidity sensors collect the temperature in front of the cabinet and the temperature in the back of the cabinet, the wind speed collector collects the wind speed in front of the cabinet and the wind speed in the back of the cabinet, and the electric meter collector collects the load of the cabinet.

[0037] Figure 2 A schematic diagram of an optional automatic guided vehicle is shown, comprising Figure 2 It can be seen that the wireless communication unit 201, sensor unit 203 and fan unit 204 in the automatic guided vehicle 20 are all connected to the micro control unit 202. The micro control unit 202 is used to send the operating data collected by the sensor unit 203 to the power environment monitoring platform, and receive control instructions from the power environment monitoring platform to control the fan unit 204 and the automatic guided vehicle to perform operations corresponding to the control instructions, such as adjusting the speed of the fan unit, the start and stop and operating speed of the target equipment, and the air conditioning setting parameters, so as to ensure timely relief of hot spots in the computer room and improve the safety of the computer room environment. In addition, the wireless communication unit 201 realizes remote control of the automatic guided vehicle and controls the temperature parameters of the air conditioners in the areas where the cabinets in the computer room are located.

[0038] At the same time, the target device can be controlled to move on a safe magnetic track to control the automatic guided vehicle to perform forward, turning, backward and other actions normally.

[0039] In addition, the embodiment of the present application also provides an optional control module. Figure 3It is a schematic diagram of the control module, and the control module can realize the control function of the microcontroller unit 202. The control module adopts an STM32 chip, which includes a wind speed acquisition serial port, an electric meter acquisition serial port, a temperature and humidity acquisition serial port, a WIFI transparent transmission serial port, and a relay control serial port. The specific WIFI transparent transmission serial port is connected to the wireless communication unit for remotely adjusting the air-conditioning temperature in the computer room; the temperature and humidity acquisition serial port is connected to the temperature and humidity sensor, the wind speed acquisition serial port is connected to the wind speed collector, and the electric meter acquisition serial port is connected to the electric meter collector. The cabinet front temperature and cabinet rear temperature collected by the temperature and humidity sensor, the wind speed collector collects the cabinet front wind speed, the cabinet rear wind speed, and the electric meter collector collects the cabinet load through the wireless communication unit and sends it to the power environment monitoring platform. The control command sent by the power environment monitoring platform will be connected to the STM32 chip through the I / O port, and the fan unit and the start and stop and running speed of the automatic guide car are controlled by the relay.

[0040] Step S104, inputting multiple groups of first operating parameters into a pre-trained hot spot area prediction model to obtain a hot spot area prediction result output by the hot spot area prediction model, wherein the hot spot area prediction result is used to reflect the probability of each cabinet being located in the hot spot area.

[0041] In order to ensure that the hotspot area prediction model can output more accurate prediction results, it is necessary to train the initial model and select the hotspot area prediction model with the optimal hyperparameters, that is, to select the optimal number of initial models, decision tree depth and learning rate.

[0042] Optionally, the training process of the hot spot area prediction model includes: determining a historical time period when a hot spot area exists in the computer room; obtaining multiple groups of fourth operating parameters of multiple cabinets in the historical time period from historical operating data, and using each group of fourth operating parameters as a group of training sample data, wherein the fourth operating parameters include at least one of the following: cabinet front temperature, cabinet rear temperature, cabinet front wind speed, cabinet rear wind speed and cabinet load; adding sample labels to each group of training sample data, and the sample labels are used to mark whether the cabinet corresponding to the training sample data is located in the hot spot area; training the initial model based on multiple groups of training sample data and sample labels to obtain a hot spot area prediction model, wherein the initial model is an XGBoost binary classification model.

[0043] The initial model training process based on multiple groups of training sample data and sample labels can be divided into the following steps: first, for each group of training sample data, the training sample data is input into the XGBoost binary classification model to obtain the second prediction probability output by the XGBoost binary classification model; then, the target loss function is constructed based on the second prediction probability and the sample labels corresponding to the training sample data; finally, multiple groups of training sample data are input into the XGBoost binary classification model in turn for iterative training, and the model parameters of the XGBoost binary classification model are adjusted based on the target loss function to obtain the hot spot area prediction model.

[0044] Among them, the target loss function in the embodiment of the present application can be a cross entropy loss function.

[0045] Specifically, the training process of the hot spot area prediction model is as follows: first, collect historical data of n cabinets in the communication room in a certain past time period when they are in the hot spot area, where the historical data must contain the operation data of the hot spot cabinet. For example, by using temperature and humidity sensors to collect the temperature measured in front of the cabinet and the temperature measured in the back of the cabinet, using wind speed collectors to collect the wind speed in front of the cabinet and the wind speed in the back of the cabinet, and using electric meter collectors to collect the cabinet load. The collected historical data is normalized and cleaned to ensure the accuracy of subsequent prediction results, and artificially marked whether the cabinet is a hot spot area y, where the hot spot area is marked as 1 and the non-hot spot area is marked as 0). The processed historical data of each cabinet can be used as a set of training sample data. For the i-th cabinet, the expression of its training sample data is as follows:

[0046]

[0047] in, Indicates the temperature measured in front of the cabinet (℃); Indicates the temperature measured at the back of the cabinet (℃); Indicates the wind speed in front of the cabinet (m / s); Indicates the wind speed at the rear of the cabinet (m / s); e i Indicates the cabinet load (KW).

[0048] At the same time, add sample labels y for each set of training samples i , which is used to determine whether the cabinet corresponding to the standard training sample data is located in the hot spot area. Therefore, the data input into the XGBoost binary classification model can be expressed by the following formula:

[0049] T={(x1,y1),(x2,y2),…,(x i ,y i ),…,(x n ,y 1＝n )}

[0050] Then, the collected multiple sets of training sample data are input into the XGBoost binary classification model to obtain the second prediction probability output by the XGBoost binary classification model. The expression of the second prediction probability is as follows:

[0051]

[0052] Among them, f m (x i ) represents the prediction result of the mth XGBoost binary classification model for the ith cabinet. All the prediction results are superimposed to obtain the second prediction probability of the ith cabinet. When the second predicted probability When the probability is greater than the preset threshold, it is confirmed that the i-th cabinet is located in the hot spot area; otherwise, it is not located in the hot spot area.

[0053] The target loss function is constructed based on the second predicted probability and the sample label corresponding to the training sample data. The expression of the target loss function is as follows:

[0054]

[0055] Therefore, the objective function of the XGBoost binary classification model can be determined by the objective loss function and the regularization term, where the objective loss function is used to determine the training error, and the regularization function is used to determine the complexity of the decision tree to prevent the XGBoost binary classification model from overfitting and improve the generalization ability of the XGBoost binary classification model. Therefore, the objective function expression of the XGBoost binary classification model can be written as:

[0056]

[0057] Then, multiple sets of training sample data are sequentially input into the XGBoost binary classification model for iterative training to determine the optimal hyperparameters of the hotspot prediction model. For the tth round of training, the objective function expression of the hotspot prediction model is:

[0058]

[0059]

[0060] Among them, T is the number of leaf nodes of the decision tree, w j is the weight of each leaf node. Therefore, the XGBoost binary classification model can be organized according to the leaf node tree, so the objective function of the XGBoost binary classification model can be expressed as:

[0061]

[0062] Therefore, the objective function expression of the final hotspot area prediction model is:

[0063]

[0064] Among them, I represents the sample set under each leaf node; G j Represents the sum of the first-order derivatives of each leaf node; H j Represents the sum of the second-order derivatives of each leaf node.

[0065] After completing the training of the hotspot area prediction model, in an embodiment of the present application, the obtained multiple sets of first operating parameters are input into the hotspot area prediction model to obtain the hotspot area prediction results output by the hotspot area prediction model. The hotspot area prediction results can accurately reflect the probability of each cabinet in the computer room being located in the hotspot area.

[0066] Step S106, determining the moving path of the target device based on the hot spot area prediction result, and controlling the target device to patrol multiple cabinets according to the moving path, wherein the target device is configured with a fan unit, and the area where the cabinet with a greater probability of being located in the hot spot area is located has a higher priority in the moving path.

[0067] Step S108: adjusting the temperature parameters of the air conditioners in the area where the multiple cabinets are located based on the hot spot area prediction result.

[0068] Specifically, based on the hot spot area prediction model, it is determined whether there is a hot spot area in the hot channel of all cabinets in the computer room, and the hot spot area prediction result is determined. The optimal control strategy is automatically generated and issued through the wireless communication module and the power environment monitoring platform, the moving path of the target equipment in the computer room is determined, and multiple cabinets in the predicted hot spot area are monitored.

[0069] For example, Figure 4A schematic diagram of an optional target device movement is shown, wherein after the operating parameters of the rightmost row cabinet 1-2 are input into the hot spot area prediction model, the output hot spot area prediction result is 0.8, indicating that the probability of the cabinet being located in the hot spot area is the highest; after the operating parameters of the leftmost row cabinet 1-9 are input into the hot spot area prediction model, the output hot spot area prediction result is 0.2, indicating that the probability of the cabinet being located in the hot spot area is relatively low. Therefore, first determine the first path that reaches the rightmost row cabinet 1-2 as quickly as possible, and control the target device to stop driving when it reaches the cabinet position at the fastest speed along the first path. At this time, the fan speed of the fan unit is intelligently increased. At the same time, in order to quickly reduce the temperature of the cabinet, the temperature parameters of the air conditioner 1 in the area where the cabinet is located can also be reduced. Then determine the second path of the leftmost row cabinet 1-9, and control the target device to stop moving when it reaches the cabinet position along the second path at the fastest speed. At this time, the fan speed of the fan unit is intelligently increased. Therefore, the probability that this cabinet is located in the hot spot area is small, indicating that the temperature of the cabinet is not too high. Therefore, cooling can be achieved only by the fan unit of the target device. At this time, there is no need to adjust the temperature parameters of the air conditioner 3 in the area where the cabinet is located.

[0070] It should be noted that the execution order of step S106 and step S108 is interchangeable, that is, step S106 may be executed first, and then step S108 may be executed.

[0071] As an optional implementation method, when sensors of various cabinets in the computer room send out abnormal alarm prompts, in order to avoid unnecessary safety hazards to the computer room environment, it is necessary to promptly inspect the cabinets that send out abnormal alarm prompts.

[0072] Optionally, in response to the temperature abnormality alarm information issued by the target cabinet, the target device is controlled to move to the area where the target cabinet is located, and a second operating parameter of the target cabinet is obtained; the second operating parameter is input into the hot spot area prediction model to obtain a first prediction probability output by the hot spot area prediction model; when the prediction probability is greater than a preset probability threshold, it is determined that the target cabinet is located in the hot spot area, the air supply volume of the fan unit is adjusted according to the temperature abnormality alarm information, and the temperature parameters of the air conditioner in the area where the target cabinet is located are adjusted according to the temperature abnormality alarm information.

[0073] Optionally, when the predicted probability is not greater than a preset probability threshold, it is determined that the target cabinet is not located in the hot spot area, and an abnormal prompt message is issued, where the abnormal prompt message is used to prompt that the temperature abnormality alarm message issued by the target cabinet is incorrect.

[0074] For example, when the alarm temperature sensor of a cabinet in a communication room emits an abnormal temperature alarm tone, the automatic guided vehicle is controlled to move to the location of the cabinet. At this time, the temperature measured in front of the cabinet and the temperature measured in the back of the cabinet are collected through the temperature and humidity sensors in the target device; the wind speed collector is used to collect the wind speed at the front and back of the cabinet; the electric meter collector is used to collect the cabinet load (i.e., the second operating parameter), and the second operating parameter is input into the hot spot area prediction model, and the first prediction probability can be output. By comparing the first prediction probability with the preset probability threshold of 0.5, it is determined whether the cabinet is located in the hot spot area.

[0075] When the first predicted probability is greater than 0.5, the cabinet is marked as 1, indicating that the cabinet is located in the hot spot area. According to the positional relationship between the alarm temperature sensor and the predicted hot spot area, the air supply volume of the fan unit and the temperature parameters of the air conditioner in the area where the cabinet is located are adjusted to timely cool down the hot spot area where the cabinet is located; when the first predicted probability is not greater than 0.5, the cabinet is marked as 0, indicating that the target cabinet is not within the hot spot area. When the alarm temperature sensor of the cabinet sends out an abnormal alarm prompt sound, it is to remind relevant technical personnel that the abnormal temperature alarm prompt sound issued by the cabinet is incorrect, and it is necessary to pay attention to whether other reasons cause the cabinet to issue an alarm prompt.

[0076] As an optional implementation method, the target equipment may bring unnecessary safety hazards to the computer room environment due to overvoltage, overcurrent, and local airflow short circuit caused by excessive fan speed during operation. Therefore, it is necessary to receive the power supply voltage of the fan unit and the battery voltage built into the automatic guided vehicle through the power environment monitoring platform to determine whether their voltage status meets the threshold value, so as to check the operating status of the target equipment and ensure that the target equipment patrols the computer room in the best state.

[0077] Optionally, a third operating parameter of the target device itself is obtained, wherein the third operating parameter includes at least one of the following: operating voltage, operating current and air output of the fan unit; when the third operating parameter does not meet a preset operating threshold range, the target device is controlled to turn off the fan unit and issue an alarm message, wherein the alarm message is used to prompt that the operating status of the target device is abnormal and requires maintenance.

[0078] For example, the operating voltage, operating current and air output of the fan unit (that is, the third operating parameter) of the target device during operation can be obtained. When the operating voltage of the target device during operation does not meet the preset operating threshold voltage range, or the operating current of the target device during operation does not meet the preset operating threshold current range, etc., the fan unit of the target device can be controlled to shut down, and an alarm message can be issued in time to remind relevant technical personnel to check the target device in time to improve the service life of the target device. At the same time, the accuracy of the cabinet operation data collected by the target device can be ensured.

[0079] In an embodiment of the present application, multiple groups of first operating parameters of multiple cabinets in a computer room collected by a target device are obtained; multiple groups of first operating parameters are input into a pre-trained hot spot prediction model to obtain hot spot prediction results output by the hot spot prediction model, wherein the hot spot prediction results are used to reflect the probability of each cabinet being located in a hot spot area; the moving path of the target device is determined based on the hot spot prediction results, and the target device is controlled to patrol multiple cabinets according to the moving path, wherein a fan unit is configured in the target device, and the cabinet with a greater probability of being located in a hot spot area has a higher priority in the moving path; the temperature parameters of the air conditioner in the area where the multiple cabinets are located are adjusted based on the hot spot prediction results. The operating data of multiple cabinets in the computer room intelligently collected by the target device ensures that the hot spot areas in the computer room can be accurately predicted later; the moving trajectory of the target device is determined by the hot spot prediction results, and the hot spot areas in the computer room are controlled in time; the temperature parameters of the air conditioner in the area where the multiple cabinets are located are adjusted by the hot spot prediction results, so as to realize intelligent control of the computer room temperature, thereby solving the technical problem that the computer room temperature cannot be intelligently and accurately controlled in the related technology, resulting in frequent local hot spots.

[0080] Example 2

[0081] According to an embodiment of the present application, a device for alleviating hot spots in a computer room is also provided for implementing the method for alleviating hot spots in a computer room in Embodiment 1. Figure 5 As shown, the hot spot area relief device of the computer room includes at least an acquisition module 51, a prediction module 52, a control module 53 and an adjustment module 54, wherein:

[0082] The acquisition module 51 is used to acquire multiple groups of first operating parameters of multiple cabinets in the computer room collected by the target device.

[0083] Optionally, the target device includes: an automatic guided vehicle, a micro control unit, a wireless communication unit, a sensor unit and a fan unit installed on the automatic guided vehicle, wherein the automatic guided vehicle is used to move according to a moving path; the sensor unit includes multiple sensors for collecting first operating parameters, and the first operating parameters include at least one of the following: cabinet front temperature, cabinet rear temperature, cabinet front wind speed, cabinet rear wind speed and cabinet load; the fan unit includes a matrix fan composed of multiple fans; the micro control unit is used to send multiple groups of first operating parameters to the power environment monitoring platform through the wireless communication unit, and receive control instructions sent by the power environment monitoring platform, so as to control the automatic guided vehicle and the fan unit to perform operations corresponding to the control instructions.

[0084] Specifically, the acquisition module 51 acquires multiple groups of first operating parameters of multiple cabinets in the computer room during the moving track determined by the target device.

[0085] Among them, the temperature and humidity sensors are used to collect the temperature in front of the cabinet and the temperature in the back of the cabinet, the wind speed collector collects the wind speed in front of the cabinet and the wind speed in the back of the cabinet, and the electric meter collector collects the cabinet load.

[0086] As an optional implementation, the target device may be controlled to travel on a secure magnetic track so that the automated guided vehicle can normally perform actions such as moving forward, turning, and moving backward.

[0087] The prediction module 52 is used to input multiple groups of first operating parameters into a pre-trained hot spot area prediction model to obtain a hot spot area prediction result output by the hot spot area prediction model, wherein the hot spot area prediction result is used to reflect the probability of each cabinet being located in the hot spot area.

[0088] In order to ensure that the prediction module 52 can output a relatively accurate prediction result, it is necessary to train the initial model and select the hot spot area prediction model with the optimal hyperparameters, that is, to select the optimal number of initial models, decision tree depth and learning rate.

[0089] Optionally, the training process of the hot spot area prediction model includes: determining a historical time period when a hot spot area exists in the computer room; obtaining multiple groups of fourth operating parameters of multiple cabinets in the historical time period from historical operating data, and using each group of fourth operating parameters as a group of training sample data, wherein the fourth operating parameters include at least one of the following: cabinet front temperature, cabinet rear temperature, cabinet front wind speed, cabinet rear wind speed and cabinet load; adding sample labels to each group of training sample data, and the sample labels are used to mark whether the cabinet corresponding to the training sample data is located in the hot spot area; training the initial model based on multiple groups of training sample data and sample labels to obtain a hot spot area prediction model, wherein the initial model is an XGBoost binary classification model.

[0090] The initial model training process based on multiple groups of training sample data and sample labels can be divided into the following steps: first, for each group of training sample data, the training sample data is input into the XGBoost binary classification model to obtain the second prediction probability output by the XGBoost binary classification model; then, the target loss function is constructed based on the second prediction probability and the sample labels corresponding to the training sample data; finally, multiple groups of training sample data are input into the XGBoost binary classification model in turn for iterative training, and the model parameters of the XGBoost binary classification model are adjusted based on the target loss function to obtain the hot spot area prediction model.

[0091] Among them, the target loss function in the embodiment of the present application can be a cross entropy loss function.

[0092] Specifically, the training process of the hot spot area prediction model is as follows: first, collect historical data of n cabinets in the communication room in a certain past time period when they are in the hot spot area, where the historical data must contain the operation data of the hot spot cabinet. For example, by using temperature and humidity sensors to collect the temperature measured in front of the cabinet and the temperature measured in the back of the cabinet, using wind speed collectors to collect the wind speed in front of the cabinet and the wind speed in the back of the cabinet, and using electric meter collectors to collect the cabinet load. The collected historical data is normalized and cleaned to ensure the accuracy of subsequent prediction results, and artificially marked whether the cabinet is a hot spot area y, where the hot spot area is marked as 1 and the non-hot spot area is marked as 0). The processed historical data of each cabinet can be used as a set of training sample data. For the i-th cabinet, the expression of its training sample data is as follows:

[0093]

[0094] in, Indicates the temperature measured in front of the cabinet (℃); Indicates the temperature measured at the back of the cabinet (℃); Indicates the wind speed in front of the cabinet (m / s); Indicates the wind speed at the rear of the cabinet (m / s); e i Indicates the cabinet load (KW).

[0095] At the same time, add sample labels y for each set of training samples i , which is used to determine whether the cabinet corresponding to the standard training sample data is located in the hot spot area. Therefore, the data input into the XGBoost binary classification model can be expressed by the following formula:

[0096] T={(x1,y1),(x2,y2),…,(x i ,y i ),…,(x n ,y 1＝n )}

[0097] Then, the collected multiple sets of training sample data are input into the XGBoost binary classification model to obtain the second prediction probability output by the XGBoost binary classification model. The expression of the second prediction probability is as follows:

[0098]

[0099] Among them, f m (x i ) represents the prediction result of the mth XGBoost binary classification model for the ith cabinet. All the prediction results are superimposed to obtain the second prediction probability of the ith cabinet. When the second predicted probability When the probability is greater than the preset threshold, it is confirmed that the i-th cabinet is located in the hot spot area; otherwise, it is not located in the hot spot area.

[0100] The target loss function is constructed based on the second predicted probability and the sample label corresponding to the training sample data. The expression of the target loss function is as follows:

[0101]

[0102] Therefore, the objective function of the XGBoost binary classification model can be determined by the objective loss function and the regularization term, where the objective loss function is used to determine the training error, and the regularization function is used to determine the complexity of the decision tree to prevent the XGBoost binary classification model from overfitting and improve the generalization ability of the XGBoost binary classification model. Therefore, the objective function expression of the XGBoost binary classification model can be written as:

[0103]

[0104] Then, multiple sets of training sample data are sequentially input into the XGBoost binary classification model for iterative training to determine the optimal hyperparameters of the hotspot prediction model. For the tth round of training, the objective function expression of the hotspot prediction model is:

[0105]

[0106]

[0107] Among them, T is the number of leaf nodes of the decision tree, w j is the weight of each leaf node. Therefore, the XGBoost binary classification model can be organized according to the leaf node tree, so the objective function of the XGBoost binary classification model can be expressed as:

[0108]

[0109] Therefore, the objective function expression of the final hotspot area prediction model is:

[0110]

[0111] Among them, I represents the sample set under each leaf node; G j Represents the sum of the first-order derivatives of each leaf node; H j Represents the sum of the second-order derivatives of each leaf node.

[0112] After completing the training of the hotspot area prediction model, in an embodiment of the present application, the obtained multiple sets of first operating parameters are input into the hotspot area prediction model to obtain the hotspot area prediction results output by the hotspot area prediction model. The hotspot area prediction results can accurately reflect the probability of each cabinet in the computer room being located in the hotspot area.

[0113] The control module 53 is used to determine the moving path of the target device based on the hot spot area prediction result, and control the target device to patrol multiple cabinets according to the moving path, wherein the target device is configured with a fan unit, and the area where the cabinet with a greater probability of being located in the hot spot area is located has a higher priority in the moving path.

[0114] The adjustment module 54 is used to adjust the temperature parameters of the air conditioners in the area where the multiple cabinets are located based on the hot spot area prediction result.

[0115] Specifically, based on the hot spot area prediction model, it is determined whether there is a hot spot area in the hot channel of all the cabinets in the computer room, and the hot spot area prediction result is determined. The optimal control strategy is automatically generated and issued through the wireless communication module and the power environment monitoring platform, and the moving path of the target equipment in the computer room is determined. Patrol monitoring is carried out on multiple cabinets in the predicted hot spot area.

[0116] It should be noted that each module in the computer room hot spot area relief device in the embodiment of the present application corresponds one by one to each implementation step of the computer room hot spot area relief method in Example 1. Since a detailed description has been given in Example 1, some details not reflected in this embodiment can be referred to Example 1 and will not be repeated here.

[0117] Example 3

[0118] According to an embodiment of the present application, a non-volatile storage medium is also provided, which includes a stored program, wherein the device where the non-volatile storage medium is located executes the method for alleviating hot spot areas in the computer room in Example 1 by running the program.

[0119] Specifically, the device where the non-volatile storage medium is located implements the following steps by running the program: obtaining multiple groups of first operating parameters of multiple cabinets in the computer room collected by the target device; inputting the multiple groups of first operating parameters into a pre-trained hot spot area prediction model to obtain the hot spot area prediction results output by the hot spot area prediction model, wherein the hot spot area prediction results are used to reflect the probability of each cabinet being located in the hot spot area; determining the moving path of the target device based on the hot spot area prediction results, and controlling the target device to patrol the multiple cabinets according to the moving path, wherein a fan unit is configured in the target device, and the area where the cabinet with a greater probability of being located in the hot spot area is located has a higher priority in the moving path; adjusting the temperature parameters of the air conditioners in the area where the multiple cabinets are located based on the hot spot area prediction results.

[0120] Optionally, in response to the temperature abnormality alarm information issued by the target cabinet, the target device is controlled to move to the area where the target cabinet is located, and a second operating parameter of the target cabinet is obtained; the second operating parameter is input into the hot spot area prediction model to obtain a first prediction probability output by the hot spot area prediction model; when the prediction probability is greater than a preset probability threshold, it is determined that the target cabinet is located in the hot spot area, the air supply volume of the fan unit is adjusted according to the temperature abnormality alarm information, and the temperature parameters of the air conditioner in the area where the target cabinet is located are adjusted according to the temperature abnormality alarm information.

[0121] Optionally, when the predicted probability is not greater than a preset probability threshold, it is determined that the target cabinet is not located in the hot spot area, and an abnormal prompt message is issued, where the abnormal prompt message is used to prompt that the temperature abnormality alarm message issued by the target cabinet is incorrect.

[0122] According to an embodiment of the present application, a processor is also provided, which is used to run a program, wherein the method for alleviating hot spot areas in a computer room in Example 1 is executed when the program is running.

[0123] Specifically, when the program is running, the following steps are executed: obtaining multiple groups of first operating parameters of multiple cabinets in the computer room collected by the target device; inputting the multiple groups of first operating parameters into a pre-trained hot spot area prediction model to obtain a hot spot area prediction result output by the hot spot area prediction model, wherein the hot spot area prediction result is used to reflect the probability of each cabinet being located in the hot spot area; determining a moving path of the target device based on the hot spot area prediction result, and controlling the target device to patrol multiple cabinets according to the moving path, wherein a fan unit is configured in the target device, and the area where the cabinet with a greater probability of being located in the hot spot area is located has a higher priority in the moving path; adjusting the temperature parameters of the air conditioners in the area where the multiple cabinets are located based on the hot spot area prediction result.

[0124] Optionally, in response to the temperature abnormality alarm information issued by the target cabinet, the target device is controlled to move to the area where the target cabinet is located, and a second operating parameter of the target cabinet is obtained; the second operating parameter is input into the hot spot area prediction model to obtain a first prediction probability output by the hot spot area prediction model; when the prediction probability is greater than a preset probability threshold, it is determined that the target cabinet is located in the hot spot area, the air supply volume of the fan unit is adjusted according to the temperature abnormality alarm information, and the temperature parameters of the air conditioner in the area where the target cabinet is located are adjusted according to the temperature abnormality alarm information.

[0125] Optionally, when the predicted probability is not greater than a preset probability threshold, it is determined that the target cabinet is not located in the hot spot area, and an abnormal prompt message is issued, where the abnormal prompt message is used to prompt that the temperature abnormality alarm message issued by the target cabinet is incorrect.

[0126] According to an embodiment of the present application, an electronic device is also provided, which includes: a memory and a processor, wherein a computer program is stored in the memory, and the processor is configured to execute the method for alleviating hot spot areas in the computer room in Example 1 through the computer program.

[0127] Specifically, the processor is configured to execute the following steps through a computer program: obtaining multiple groups of first operating parameters of multiple cabinets in a computer room collected by a target device; inputting the multiple groups of first operating parameters into a pre-trained hot spot area prediction model to obtain a hot spot area prediction result output by the hot spot area prediction model, wherein the hot spot area prediction result is used to reflect the probability of each cabinet being located in the hot spot area; determining a moving path of the target device based on the hot spot area prediction result, and controlling the target device to patrol multiple cabinets according to the moving path, wherein a fan unit is configured in the target device, and the area where the cabinet with a greater probability of being located in the hot spot area is located has a higher priority in the moving path; adjusting the temperature parameters of the air conditioners in the area where the multiple cabinets are located based on the hot spot area prediction result.

[0128] Optionally, in response to the temperature abnormality alarm information issued by the target cabinet, the target device is controlled to move to the area where the target cabinet is located, and a second operating parameter of the target cabinet is obtained; the second operating parameter is input into the hot spot area prediction model to obtain a first prediction probability output by the hot spot area prediction model; when the prediction probability is greater than a preset probability threshold, it is determined that the target cabinet is located in the hot spot area, the air supply volume of the fan unit is adjusted according to the temperature abnormality alarm information, and the temperature parameters of the air conditioner in the area where the target cabinet is located are adjusted according to the temperature abnormality alarm information.

[0129] Optionally, when the predicted probability is not greater than a preset probability threshold, it is determined that the target cabinet is not located in the hot spot area, and an abnormal prompt message is issued, where the abnormal prompt message is used to prompt that the temperature abnormality alarm message issued by the target cabinet is incorrect.

[0130] The serial numbers of the above-mentioned embodiments of the present application are for description only and do not represent the advantages or disadvantages of the embodiments.

[0131] In the above embodiments of the present application, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, please refer to the relevant description of other embodiments.

[0132] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the device embodiments described above are only schematic. For example, the division of units can be a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of units or modules, which can be electrical or other forms.

[0133] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple units. Some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0134] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0135] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art or all or part of the technical solution, can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, server or network device, etc.) to perform all or part of the steps of each embodiment method of the present application. The aforementioned storage medium includes: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, disk or optical disk, etc. Various media that can store program codes.

[0136] The above are only preferred implementations of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present application. These improvements and modifications should also be regarded as the scope of protection of the present application.

Claims

1. A method for alleviating hotspot areas in a computer room, applied to a power environment monitoring platform, characterized in that: include: Acquire multiple groups of first operating parameters of multiple cabinets in a computer room collected by a target device; Inputting the plurality of groups of first operating parameters into a pre-trained hot spot area prediction model to obtain a hot spot area prediction result output by the hot spot area prediction model, wherein the hot spot area prediction result is used to reflect the probability that each cabinet is located in the hot spot area; Determine a moving path of the target device based on the hot spot area prediction result, and control the target device to patrol the multiple cabinets according to the moving path, wherein the target device is configured with a fan unit, and the cabinet with a greater probability of being located in the hot spot area has a higher priority in the moving path; The temperature parameters of the air conditioners in the area where the multiple cabinets are located are adjusted based on the hot spot area prediction result.

2. The method according to claim 1, characterized in that The target device includes: an automatic guided vehicle, a micro control unit, a wireless communication unit, a sensor unit and the fan unit installed on the automatic guided vehicle, wherein: The automatic guided vehicle is used to move according to the moving path; The sensor unit includes a plurality of sensors for collecting the first operating parameters, wherein the first operating parameters include at least one of the following: a cabinet front temperature, a cabinet rear temperature, a cabinet front wind speed, a cabinet rear wind speed, and a cabinet load; The fan unit includes a matrix fan composed of a plurality of fans; The micro control unit is used to send the multiple groups of first operating parameters to the power environment monitoring platform through the wireless communication unit, and receive control instructions sent by the power environment monitoring platform to control the automatic guided vehicle and the fan unit to perform operations corresponding to the control instructions.

3. The method according to claim 1, characterized in that The method further comprises: In response to the abnormal temperature alarm information issued by the target cabinet, controlling the target device to move to the area where the target cabinet is located, and obtaining the second operating parameter of the target cabinet; Inputting the second operating parameter into the hotspot area prediction model to obtain a first prediction probability output by the hotspot area prediction model; When the first predicted probability is greater than a preset probability threshold, it is determined that the target cabinet is located in a hot spot area, the air supply volume of the fan unit is adjusted according to the temperature abnormality alarm information, and the temperature parameters of the air conditioner in the area where the target cabinet is located are adjusted according to the temperature abnormality alarm information.

4. The method according to claim 3, characterized in that The method further comprises: When the first predicted probability is not greater than the preset probability threshold, it is determined that the target cabinet is not located in the hot spot area, and an abnormal prompt information is issued, where the abnormal prompt information is used to prompt that the temperature abnormality alarm information issued by the target cabinet is incorrect.

5. The method according to claim 1, characterized in that The method further comprises: Acquire a third operating parameter of the target device itself, wherein the third operating parameter includes at least one of the following: operating voltage, operating current, and air output of the fan unit; When the third operating parameter does not meet the preset operating threshold range, the target device is controlled to turn off the fan unit and send out an alarm message, where the alarm message is used to prompt that the operating state of the target device is abnormal and needs maintenance.

6. The method according to claim 1, characterized in that The training process of the hotspot area prediction model includes: Determine a historical time period when a hot spot area exists in the computer room; Acquire multiple groups of fourth operating parameters of the multiple cabinets in the historical time period from the historical operating data, and use each group of the fourth operating parameters as a group of training sample data, wherein the fourth operating parameters include at least one of the following: cabinet front temperature, cabinet rear temperature, cabinet front wind speed, cabinet rear wind speed and cabinet load; Adding a sample label to each group of the training sample data, wherein the sample label is used to mark whether the cabinet corresponding to the training sample data is located in a hot spot area; The initial model is trained based on multiple groups of the training sample data and the sample labels to obtain the hotspot area prediction model, wherein the initial model is an XGBoost binary classification model.

7. The method according to claim 6, characterized in that The initial model is trained based on multiple groups of the training sample data and the sample labels to obtain the hotspot area prediction model, including: For each group of the training sample data, inputting the training sample data into the XGBoost binary classification model to obtain a second prediction probability output by the XGBoost binary classification model; Constructing a target loss function based on the second predicted probability and the sample label corresponding to the training sample data; Inputting multiple groups of the training sample data into the XGBoost binary classification model in sequence for iterative training, and adjusting the model parameters of the XGBoost binary classification model based on the target loss function to obtain the hot spot area prediction model.

8. A device for alleviating hotspot areas in a computer room, characterized in that: include: An acquisition module, used to acquire multiple groups of first operating parameters of multiple cabinets in a computer room collected by a target device; A prediction module, used for inputting the plurality of groups of first operating parameters into a pre-trained hot spot area prediction model to obtain a hot spot area prediction result output by the hot spot area prediction model, wherein the hot spot area prediction result is used to reflect the probability that each cabinet is located in the hot spot area; a control module, configured to determine a moving path of the target device based on the hot spot area prediction result, and control the target device to patrol the multiple cabinets according to the moving path, wherein the target device is configured with a fan unit, and the cabinet with a greater probability of being located in the hot spot area has a higher priority in the moving path; An adjustment module is used to adjust the temperature parameters of the air conditioners in the area where the multiple cabinets are located based on the hot spot area prediction result.

9. A non-volatile storage medium, characterized in that: The non-volatile storage medium includes a stored program, wherein the device where the non-volatile storage medium is located executes the method for alleviating hotspot areas in a computer room as described in any one of claims 1 to 7 by running the program.

10. An electronic device, characterized in that: include: A memory and a processor, wherein the memory stores a computer program, and the processor is configured to execute the method for alleviating hotspot areas in a computer room as described in any one of claims 1 to 7 through the computer program.

Images