Air conditioning operation control method, device, electronic equipment and storage medium

Abstract

The embodiment of the present invention provides an air conditioner operation control method, device, electronic device and storage medium, the method comprising: obtaining the importance of a target area and the first historical collected data of the target area in the past first preset time period; inputting the first historical collected data into a multi-index prediction model, predicting the predicted load rate of the air conditioner in a second preset time period following the first preset time period, the predicted computing resource usage rate of the server, and the predicted temperature difference between the temperature of the server and the ambient temperature, and controlling the operation of the air conditioner according to the importance, predicted load rate, predicted computing resource usage rate and predicted temperature difference. In the embodiment of the present invention, multi-dimensional parameters are predicted based on the multi-index prediction model, and when it is found that the hair temperature is about to rise, the cooling intensity of the air conditioner is increased in advance for effective cooling, and when it is found that the hair temperature is about to decrease, the cooling intensity of the air conditioner is reduced in advance to avoid waste of air conditioning energy consumption.

Description

Technical Field

[0001] The present invention relates to the technical field of air conditioning control, and in particular to an air conditioning operation control method, an air conditioning operation control device, an electronic device and a computer-readable storage medium. Background Art

[0002] The main maintenance mode of the computer room air conditioner is still based on manual inspection and regulation. The computer room maintenance regulations are used as the basis, and the various parameters of the air conditioner in the computer room are set in combination with the subjective feelings of the inspectors, temperature alarm events in the computer room, customer requirements and other factors.

[0003] However, the existing air conditioning maintenance mode cannot respond quickly to rapid changes in temperature, especially for large data center (IDC, Internet Data Center) computer rooms, which often cannot respond quickly to local high temperatures and set the temperature too low, resulting in energy waste. Summary of the invention

[0004] The embodiment of the present invention provides an air conditioner operation control method, device, electronic device and storage medium to solve the problem that the existing air conditioner maintenance mode cannot quickly respond to rapid changes in temperature, sets the temperature too low, and causes energy waste.

[0005] The embodiment of the present invention discloses an air conditioner operation control method, the method comprising:

[0006] Obtaining the importance of the target area and first historical collected data of the target area in the past first preset time period; wherein the first historical collected data includes a historical load rate of the air conditioner in the target area, a historical computing resource usage rate of the server, and a historical temperature difference between the temperature of the server and the ambient temperature in the target area;

[0007] Inputting the first historical collected data into a multi-index prediction model to obtain a predicted load rate of the air conditioner, a predicted computing resource usage rate of the server, and a predicted temperature difference between the temperature of the server and the ambient temperature in a second preset time period following the first preset time period; the multi-index prediction model is trained using the second historical collected data of the target area;

[0008] The operation of the air conditioner is controlled according to the importance degree, the predicted load rate, the predicted computing resource usage rate, and the predicted temperature difference.

[0009] Optionally, the obtaining of the importance of the target area and the first historical collected data of the target area in the past first preset time period includes:

[0010] Construct a wide table of server performance, wide table of temperature and humidity signals, wide table of air conditioning operation status, and wide table of static space resource services;

[0011] The server performance wide table is used to collect the temperature of servers in each area and the computing resource usage rate of the servers; the temperature and humidity signal wide table is used to collect the ambient temperature of each area of ​​the computer room; the air conditioner operation wide table is used to collect the working current and compressor pressure of the air conditioner in each area; the space resource service static wide table is used to collect the importance of the service in each area;

[0012] The importance of the target area is obtained from the spatial resource business static wide table; the working current and compressor pressure of the air conditioner in the target area in the past first preset time period are obtained from the air conditioner operation status wide table, and the historical load rate of the air conditioner in the target area is determined based on the working current and the compressor pressure; the historical computing resource usage rate of the server in the target area in the past first preset time period is obtained from the server performance wide table; the temperature of the server in the target area and the ambient temperature of the target area in the past first preset time period are obtained from the server performance wide table and the temperature and humidity signal wide table respectively, and the historical temperature difference between the temperature of the server in the target area and the ambient temperature of the target area is calculated based on the temperature of the server and the ambient temperature of the target area.

[0013] Optionally, the multi-index prediction model includes a load rate prediction model, a computing resource usage rate prediction model and a temperature difference prediction model, and the first historical collection data is input into the multi-index prediction model to predict a predicted load rate of the air conditioner, a predicted computing resource usage rate of the server, and a predicted temperature difference between the temperature of the server and the ambient temperature in a second preset time period following the first preset time period, including:

[0014] Inputting the historical load rate into the load rate prediction model to obtain a predicted load rate of the air conditioner within the second preset time period;

[0015] Inputting the historical computing resource usage rate into the computing resource usage rate prediction model to obtain the predicted computing resource usage rate of the server within the second preset time period;

[0016] The historical temperature difference is input into the temperature difference prediction model to obtain a predicted temperature difference between the temperature of the server and the ambient temperature within the second preset time period.

[0017] Optionally, the method further comprises:

[0018] Acquire first training sample data, second training sample data, and third training sample data; wherein the first training sample data is the historical load rate of the air conditioner in the second historical collection data of the target area, the second training sample data is the historical computing resource usage rate of the server in the second historical collection data of the target area, and the third training sample data is the historical temperature difference between the temperature of the server in the second historical collection data of the target area and the ambient temperature in the target area;

[0019] Training the load rate prediction model to be trained using the first training sample data to obtain a trained load rate prediction model;

[0020] The computing resource usage prediction model to be trained is trained by using the second training sample data to obtain a trained computing resource usage prediction model;

[0021] The temperature difference prediction model to be trained is trained by using the third training sample data to obtain a trained temperature difference prediction model.

[0022] Optionally, the operation of the air conditioner is controlled according to the importance degree, the predicted load rate, the predicted computing resource usage rate and the predicted temperature difference, including:

[0023] determining an air conditioning control strategy according to the importance degree, the predicted load rate, the predicted computing resource usage rate, and the predicted temperature difference;

[0024] The operation of the air conditioner is controlled by the air conditioner control strategy.

[0025] Optionally, the importance level is divided into four categories: extremely high, high, medium and low, and the predicted load rate, the predicted computing resource usage rate and the predicted temperature difference are all divided into three categories: high, medium and low.

[0026] Optionally, the control strategy includes at least:

[0027] When the importance level is extremely high, the predicted load rate is high, and the predicted temperature difference is high, increasing the cooling capacity of the air conditioner;

[0028] When the importance level is high, the predicted load rate is high, and the predicted temperature difference is high, the cooling capacity of the air conditioner is increased; when the importance level is high, the predicted load rate is high, and the predicted temperature difference is medium, the cooling capacity of the air conditioner is maintained; when the importance level is high, the predicted load rate is high, and the predicted temperature difference is low, the cooling capacity of the air conditioner is reduced;

[0029] When the importance level is high, the predicted load rate is medium, and the predicted temperature difference is high, the cooling capacity of the air conditioner is maintained; when the importance level is high, the predicted load rate is medium, and the predicted temperature difference is medium or low, the cooling capacity of the air conditioner is reduced;

[0030] When the importance level is medium, the predicted load rate is high, and the predicted temperature difference is high, the cooling capacity of the air conditioner is increased; when the importance level is medium, the predicted load rate is high, and the predicted temperature difference is medium or low, the cooling capacity of the air conditioner is reduced.

[0031] The embodiment of the present invention further discloses an air conditioner operation control device, comprising:

[0032] A data acquisition module, used to acquire the importance of a target area and first historical collected data of the target area in the past first preset time period; wherein the first historical collected data includes a historical load rate of the air conditioner in the target area, a historical computing resource usage rate of the server, and a historical temperature difference between the temperature of the server and the ambient temperature in the target area;

[0033] a result prediction module, for inputting the first historical collection data into a multi-index prediction model, and obtaining a predicted load rate of the air conditioner, a predicted computing resource usage rate of the server, and a predicted temperature difference between the temperature of the server and the ambient temperature in a second preset time period following the first preset time period; the multi-index prediction model is obtained by training the second historical collection data of the target area;

[0034] An operation control module is used to control the operation of the air conditioner according to the importance degree, the predicted load rate, the predicted computing resource usage rate and the predicted temperature difference.

[0035] Optionally, the data acquisition module includes:

[0036] The table construction submodule is used to construct a server performance wide table, a temperature and humidity signal wide table, an air conditioning operation status wide table, and a space resource business static wide table;

[0037] The data collection submodule is used to collect the temperature of the servers in each area and the computing resource usage rate of the servers through the server performance wide table; collect the ambient temperature of each area of ​​the computer room through the temperature and humidity signal wide table; collect the working current and compressor pressure of the air conditioners in each area through the air conditioner operation wide table; collect the importance of the services in each area through the space resource service static wide table;

[0038] The data acquisition submodule is used to obtain the importance of the target area from the spatial resource business static wide table; obtain the working current and compressor pressure of the air conditioner in the target area in the past first preset time period from the air conditioner operation status wide table, and determine the historical load rate of the air conditioner in the target area based on the working current and compressor pressure; obtain the historical computing resource usage rate of the server in the target area in the past first preset time period from the server performance wide table; obtain the temperature of the server in the target area and the ambient temperature of the target area in the past first preset time period from the server performance wide table and the temperature and humidity signal wide table respectively, and calculate the historical temperature difference between the temperature of the server in the target area and the ambient temperature of the target area based on the temperature of the server and the ambient temperature of the target area.

[0039] Optionally, the multi-index prediction model includes a load rate prediction model, a computing resource usage prediction model and a temperature difference prediction model, and the result prediction module includes:

[0040] A load rate prediction submodule, used for inputting the historical load rate into the load rate prediction model to obtain a predicted load rate of the air conditioner within the second preset time period;

[0041] A usage rate prediction submodule, used for inputting the historical computing resource usage rate into the computing resource usage rate prediction model to obtain the predicted computing resource usage rate of the server within the second preset time period;

[0042] The temperature difference prediction submodule is used to input the historical temperature difference into the temperature difference prediction model to obtain the predicted temperature difference between the temperature of the server and the ambient temperature within the second preset time period.

[0043] Optionally, it also includes:

[0044] A sample data acquisition submodule is used to acquire first training sample data, second training sample data and third training sample data; wherein the first training sample data is the historical load rate of the air conditioner in the second historical collection data of the target area, the second training sample data is the historical computing resource usage rate of the server in the second historical collection data of the target area, and the third training sample data is the historical temperature difference between the temperature of the server in the second historical collection data of the target area and the ambient temperature in the target area;

[0045] A first model training submodule is used to train the load rate prediction model to be trained using the first training sample data to obtain a trained load rate prediction model;

[0046] A second model training submodule is used to train the computing resource usage prediction model to be trained by using the second training sample data to obtain a trained computing resource usage prediction model;

[0047] The third model training submodule is used to train the temperature difference prediction model to be trained by using the third training sample data to obtain a trained temperature difference prediction model.

[0048] Optionally, the operation control module includes

[0049] a strategy determination submodule, configured to determine an air conditioning control strategy according to the importance degree, the predicted load rate, the predicted computing resource usage rate and the predicted temperature difference;

[0050] The operation control submodule is used to control the operation of the air conditioner through the air conditioner control strategy.

[0051] Optionally, the importance level is divided into four categories: extremely high, high, medium and low, and the predicted load rate, the predicted computing resource usage rate and the predicted temperature difference are all divided into three categories: high, medium and low.

[0052] Optionally, the control strategy includes at least:

[0053] When the importance level is extremely high, the predicted load rate is high, and the predicted temperature difference is high, increasing the cooling capacity of the air conditioner;

[0054] When the importance level is high, the predicted load rate is high, and the predicted temperature difference is high, the cooling capacity of the air conditioner is increased; when the importance level is high, the predicted load rate is high, and the predicted temperature difference is medium, the cooling capacity of the air conditioner is maintained; when the importance level is high, the predicted load rate is high, and the predicted temperature difference is low, the cooling capacity of the air conditioner is reduced;

[0055] When the importance level is high, the predicted load rate is medium, and the predicted temperature difference is high, the cooling capacity of the air conditioner is maintained; when the importance level is high, the predicted load rate is medium, and the predicted temperature difference is medium or low, the cooling capacity of the air conditioner is reduced;

[0056] When the importance level is medium, the predicted load rate is high, and the predicted temperature difference is high, the cooling capacity of the air conditioner is increased; when the importance level is medium, the predicted load rate is high, and the predicted temperature difference is medium or low, the cooling capacity of the air conditioner is reduced.

[0057] The embodiment of the present invention further discloses an electronic device, comprising a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other via the communication bus;

[0058] The memory is used to store computer programs;

[0059] The processor is used to implement the method described in the embodiment of the present invention when executing the program stored in the memory.

[0060] The embodiment of the present invention further discloses a computer-readable storage medium having instructions stored thereon, which, when executed by one or more processors, enables the processors to execute the method described in the embodiment of the present invention.

[0061] The embodiments of the present invention include the following advantages: multi-dimensional parameters in the target area in the future are predicted through a multi-indicator prediction model, and the operation of the air conditioner is accurately controlled in advance based on the predicted multi-dimensional parameters. Specifically, when it is predicted that the temperature of the target area (the temperature of the server and / or the ambient temperature) will rise rapidly in the future, the cooling intensity of the air conditioner can be increased in advance for effective cooling to avoid server failure caused by the rapid rise in the temperature of the target area. When it is predicted that the temperature of the target area will drop in the future, the cooling intensity of the air conditioner can be reduced to avoid setting the temperature of the air conditioner too low, resulting in a waste of air conditioning energy consumption, thereby achieving accurate control of the air conditioner through multi-dimensional prediction parameters, reducing the energy consumption of the computer room, and improving the degree of automation of air conditioning maintenance.

[0062] In addition, when controlling the operation of the air conditioner, the importance of the target area also needs to be considered. When the cooling capacity supplied by the air conditioner is limited, the cooling capacity can be supplied preferentially to the target areas with high importance to prevent the failure of the server in the target areas with high importance, resulting in the inability to process important business in time and causing huge economic losses. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] Figure 1 is a flowchart of a method for controlling air conditioning operation provided in an embodiment of the present invention;

[0064] Figure 2 is a flow chart of a data association analysis provided in an embodiment of the present invention;

[0065] Figure 3 is a flow chart of data transmission provided in an embodiment of the present invention;

[0066] Figure 4 is a structural block diagram of an air conditioning operation control device provided in an embodiment of the present invention;

[0067] Figure 5 It is a block diagram of an electronic device provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0068] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0069] With the increasing development of AI technology, artificial intelligence has gradually begun to touch all aspects of life. Various industries are expected to be closely integrated with artificial intelligence and subvert the traditional development model. As the "dual carbon" policy becomes an important foundation for green transformation, as a network operator of mobile Internet, it is also constantly exploring new models for green and low-carbon transformation of computer rooms.

[0070] At present, the main maintenance mode of the air conditioner in the operator's computer room is still mainly based on manual inspection and regulation. Based on the computer room maintenance regulations, the various parameters of the air conditioner in the computer room are set in combination with the subjective feelings of the inspectors, temperature alarm events in the computer room, customer requirements and other factors. The existing air conditioning maintenance mode cannot respond quickly to rapid changes in temperature, especially for large data center computer rooms. It is often unable to respond quickly to local high temperatures. The temperature is often set too low, resulting in energy waste. Operators urgently need an energy-saving technology that can quickly identify the temperature of each area in the computer room and automatically adjust the air conditioner to improve the overall efficiency of the computer room and gradually transform to a green computer room.

[0071] In order to adapt to the development of low-carbon transformation of large data centers, a computer room air conditioning control method based on regional temperature has emerged: regional temperature air conditioning control technology collects the ambient temperature of each area of ​​the computer room and centrally aggregates it into local group control, so that operator maintenance personnel can combine the on-site temperature conditions and perform differentiated maintenance and configuration of computer room air conditioning according to business importance, rather than simply controlling all air conditioners in a computer room equally based on the computer room as a maintenance unit, thereby improving the overall energy efficiency utilization rate of the computer room. At present, operator computer room maintenance has adopted a maintenance mode based on regional temperature.

[0072] In traditional communication rooms, the local dynamic environment monitoring system (SS, Supervision System) provides the data source for temperature analysis in each area; in large data centers, the group control system and HVAC system provide the data source for temperature analysis in each area; in base stations and access networks, the provincial convergence system provides the data source for regional temperature analysis. The real-time temperature performance data contained in these systems is similar. Regional temperature air conditioning control technology collects temperature data in the room and matches it to the corresponding air conditioner, which serves as the basis for differentially adjusting the air conditioners in each area of ​​the same room.

[0073] Therefore, the accuracy, timeliness and completeness of the temperature performance data collected in each area are the key to implementing fine-tuning of the air conditioning in the computer room. At present, the regional temperature air conditioning control technology mainly uses the following steps to control the air conditioning:

[0074] 1) Computer room temperature collection: connect to the local dynamic environment monitoring system to obtain real-time data and location data of all temperature sensors in the computer room;

[0075] 2) Associate the air conditioners in the computer room: Get the location information of all the air conditioners in the computer room, associate them with the temperature in the previous step and form a regional group;

[0076] 3) Adjust the air conditioner set temperature: average the real-time temperature values ​​of each area and adjust the air conditioner according to the pre-set threshold value.

[0077] However, after a period of actual use, the current maintenance mode of computer room air conditioning control based on regional temperature was found to be effective at the beginning, but bottlenecks would soon appear, making it impossible to further reduce carbon emissions. The specific difficulties are as follows:

[0078] 1) There are not enough temperature measurement points in the area or the distance between them is far, making it difficult to accurately control the air conditioner

[0079] At present, the rapidly developing big data, blockchain and other businesses have higher and higher requirements for terminal bandwidth, and the number of computer rooms established and maintained by operators is also increasing day by day. However, due to historical reasons, most of them are old computer rooms, and hardware facilities, especially sensor measurement points, are missing. A computer room of 100 square meters has only 3 or 4 temperature measurement points, and they are all installed on the roof of the computer room, which is a certain distance from the actual business racks. It is impossible to accurately collect the temperature of the business racks and it is difficult to accurately control the terminal air conditioners. Relying on these collected data, the maintenance personnel of the computer room often prioritize the safety of the computer room business and set the air conditioner to a lower temperature, resulting in energy waste.

[0080] 2) The single rack in the IDC room is highly service-intensive and heats up quickly in a short period of time, so the air conditioner cannot maintain a constant temperature.

[0081] With the development of cloud computing, 5G, the Internet of Things and other industries, the explosive growth of global data volume and data traffic has led to rapid business growth. The business of computer rooms, especially IDC computer rooms, has become more and more diverse. High-energy consumption and high-power racks are constantly used in computer rooms. The temperature rises faster and faster in a short period of time. The air conditioner needs to be set to a lower temperature to suppress the temperature increase trend, which consumes a lot of energy. For example, a computer room had only 30 businesses on the shelves a few days ago. After 10 days, it was fully loaded with 100 racks and was ready to continue to increase. At this time, the air conditioner in the area needs to be set to a lower temperature and consume more energy to suppress the temperature increase and ensure the normal operation of the business.

[0082] 3) The air conditioning control mode is single and lacks flexibility

[0083] In the era of mobile Internet, the number of high-computing-power servers has increased significantly, and the requirements for the computer room environment are getting higher and higher. A single air-conditioning control mode can no longer meet the energy-saving needs of multiple business scenarios in large data centers. There is an urgent need for an air-conditioning energy-saving mode that can be flexibly customized according to business scenarios.

[0084] Especially for the different requirements of various types of customers in IDC data centers, the temperature and air conditioning control method based on a single area can no longer support the further requirements for "dual carbon" and lacks the ability to control important business segments. For example: in the same computer room or even the same rack, the temperature of certain important racks is given priority protection, and the temperature of certain racks is lowered to reduce energy consumption.

[0085] Due to the above reasons, the current single-area temperature air-conditioning control method has problems such as inability to accurately control the air-conditioning, high energy consumption, single form and lack of flexibility.

[0086] Based on this, an air-conditioning operation control method, device, electronic device and storage medium are proposed in the embodiments of the present invention to solve the above-mentioned technical problems.

[0087] The air conditioning operation control method proposed in the embodiment of the present invention requires upgrading and reconstructing the on-site dynamic environment (dynamic environment monitoring system), resource system, IDC business system, on-site airflow organization, and data center. On-site maintenance personnel need to define each scene label once, and during the initialization process, they only need to model the five control schemes once. The specific requirements are as follows:

[0088] 1. Transformation of data field association and positioning in dynamic environment network management, resource system, and IDC business system

[0089] A temperature and humidity sensor with the PositionSen attribute field needs to be added to the dynamic environment monitoring system. It mainly stores the location information formed by the spatial information association of the resource system, and forms the relative position association data RelativePo together with the real-time temperature and humidity data collected at different time points, and associates and combines other wide tables.

[0090] Add the resource system space resource location information Position field, and combine it with static resource data such as IDC business racks and frames to form relative position association data RelativePo.

[0091] The PositionSer field, the position parameter of the rack where the server belongs to the IDC business system, is added. The relative position association data RelativePo is formed by combining the collected data such as CPU temperature, CPU load, data flow, motherboard temperature, and fan speed.

[0092] 2. Modification of air flow organization in the computer room area

[0093] The transformation of the airflow organization in each area of ​​the computer room must ensure that each area does not interfere with each other, the airflow organization is good, and all racks, cabinets, and floor baffles are complete.

[0094] 3. Transformation of the data center platform and system interfaces, label definition, modeling and analysis

[0095] The external interface of the data center was modified to connect the dynamic environment, resource system, and IDC business system with real-time data. The four types of labels were defined so that they could be automatically generated based on the collected data. The data modeling was modified based on the DeepAR (RNN+Embedding) algorithm to predict the heat generation of each area in the computer room.

[0096] Reference Figure 1 , shows a flow chart of the steps of a blade number detection method provided in an embodiment of the present invention, which may specifically include the following steps:

[0097] Step 101: Obtain the importance of a target area and first historical collected data of the target area within a first preset time period in the past.

[0098] The first historical collection data includes the historical load rate of the air conditioner in the target area, the historical computing resource usage rate of the server, and the historical temperature difference between the temperature of the server and the ambient temperature in the target area.

[0099] The computer room can usually be divided into multiple areas, such as 2 areas, 3 areas, etc. Each area is equipped with corresponding servers and air conditioners, and the air conditioners are used to cool the servers. The importance of servers varies according to the different businesses they process. The importance of the area to which the server belongs is differentiated according to the importance of the business they process. The more important the business, the higher the importance of the area to which the server belongs. The less important the business, the lower the importance of the area to which the server belongs.

[0100] The load rate of the air conditioner can be obtained by a comprehensive evaluation of the operating parameters of the air conditioner, such as working current, working voltage, compressor pressure, etc.; the computing resource utilization rate can be the hardware device utilization rate on the server (such as CPU utilization rate, GPU utilization rate); the server temperature is the internal temperature of the server (such as CPU temperature, motherboard temperature), the historical temperature difference between the server temperature and the ambient temperature in the target area, that is, the difference between the internal temperature of the server and the external ambient temperature where it is located.

[0101] During the operation of the air conditioner in each area of ​​the computer room, real-time data in each area can be periodically collected, including: CPU temperature, CPU load, data flow, motherboard temperature in the server, ambient temperature, humidity data, return air temperature and humidity during air conditioner operation, remote on / off control signal, temperature value setting signal data, etc. Thus, the first historical collection data of the target area can be obtained from the real-time data.

[0102] Specifically, before controlling the operation of the air conditioner in the target area, first obtain the importance of the target area and the first historical collected data of the target area in the past first preset time period, for example, obtain the first historical collected data of the target area in the past week, two weeks or one month. The first historical collected data includes the historical load rate of the air conditioner in the target area, the historical computing resource usage rate of the server, and the historical temperature difference between the temperature of the server and the ambient temperature in the target area.

[0103] Step 102: Input the first historical collected data into a multi-index prediction model to obtain a predicted load rate of the air conditioner, a predicted computing resource usage rate of the server, and a predicted temperature difference between the temperature of the server and the ambient temperature in a second preset time period following the first preset time period.

[0104] Among them, the multi-index prediction model is trained through the second historical collection data of the target area.

[0105] Specifically, the first historical collected data, such as the historical load rate of the air conditioner in the target area within the first preset time period in the past, the historical computing resource usage rate of the server, and the historical temperature difference between the temperature of the server and the ambient temperature in the target area, are input into the multi-indicator prediction model. The predicted load rate of the air conditioner, the predicted computing resource usage rate of the server, and the predicted temperature difference between the temperature of the server and the ambient temperature in the second preset time period following the first preset time period can be predicted. The second preset time period can be 10 minutes, half an hour, one hour, etc. following the first preset time period.

[0106] Step 103: Controlling the operation of the air conditioner according to the importance degree, the predicted load rate, the predicted computing resource usage rate and the predicted temperature difference.

[0107] Specifically, after predicting the predicted load rate of the air conditioner, the predicted computing resource usage rate of the server, and the predicted temperature difference between the server temperature and the ambient temperature within the second preset time period, within the next second preset time period, the operation of the air conditioner can be controlled in advance according to the importance of the target area, the predicted load rate of the air conditioner in the target area, the predicted computing resource usage rate of the server, and the predicted temperature difference between the server temperature and the ambient temperature of the target area. For example, before the temperature of the target area is about to change rapidly, corresponding effective cooling measures can be adopted through the air conditioner. On the one hand, the temperature of the target area can be prevented from rising rapidly. On the other hand, when it is predicted that the temperature of the target area will drop within the next second preset time period, the temperature of the air conditioner can be adjusted up to avoid setting the temperature of the air conditioner too low, resulting in waste of air conditioning energy.

[0108] In an embodiment of the present invention, a multi-dimensional parameter of a target area in the future is predicted through a multi-index prediction model, and the operation of the air conditioner is accurately controlled in advance based on the predicted multi-dimensional parameters. Specifically, when it is predicted that the temperature of the target area (the temperature of the server and / or the ambient temperature) will rise rapidly in the future, the cooling intensity of the air conditioner can be increased in advance for effective cooling to avoid server failure caused by the rapid rise in the temperature of the target area. When it is predicted that the temperature of the target area will drop in the future, the cooling intensity of the air conditioner can be reduced to avoid setting the temperature of the air conditioner too low, resulting in a waste of air conditioning energy, thereby achieving accurate control of the air conditioner through multi-dimensional prediction parameters, reducing the energy consumption of the computer room, and improving the degree of automation of air conditioning maintenance.

[0109] In addition, when controlling the operation of the air conditioner, the importance of the target area also needs to be considered. When the cooling capacity supplied by the air conditioner is limited, the cooling capacity can be supplied preferentially to the target areas with high importance to prevent the failure of the server in the target areas with high importance, resulting in the inability to process important business in time and causing huge economic losses.

[0110] Based on the above embodiment, a variant embodiment of the above embodiment is proposed. It should be noted that in order to make the description concise, only the differences from the above embodiment are described in the variant embodiment.

[0111] In an optional embodiment of the present invention, step 101 includes: constructing a server performance wide table, a temperature and humidity signal wide table, an air conditioning operation wide table, and a space resource business static wide table; collecting the temperature of each area server and the computing resource utilization rate of the server through the server performance wide table; collecting the ambient temperature of each area of ​​the computer room through the temperature and humidity signal wide table; collecting the working current and compressor pressure of the air conditioner in each area through the air conditioning operation wide table; collecting the importance of each area business through the space resource business static wide table; obtaining the importance of the target area from the space resource business static wide table; obtaining the working current and compressor pressure of the air conditioner in the target area in the past first preset time period from the air conditioning operation wide table, and determining the historical load rate of the air conditioner in the target area according to the working current and the compressor pressure; obtaining the historical computing resource utilization rate of the server in the target area in the past first preset time period from the server performance wide table; obtaining the temperature of the server in the target area and the ambient temperature of the target area in the past first preset time period from the server performance wide table and the temperature and humidity signal wide table respectively, and calculating the historical temperature difference between the temperature of the server in the target area and the ambient temperature of the target area according to the temperature of the server and the ambient temperature of the target area.

[0112] In order to simplify the complex temperature field of the computer room area, three types of real-time data collection wide tables are created: server performance wide table, temperature and humidity signal wide table, air conditioning operation status wide table, and a static wide table of computer room space resource data. The parameter Position describing the positioning information in the air conditioning resource is used to identify the location information of each device in each wide table.

[0113] Server performance wide table: used to store business server self-test data, especially CPU temperature (server temperature), CPU load (computer resource utilization), data flow, motherboard temperature, fan speed and other collected data as well as server rack position parameters (PositionSer).

[0114] Temperature and humidity signal wide table: used to store temperature signal data (ambient temperature) of each area of ​​the computer room, stored in the order of collection time, and define the two-dimensional positioning parameters (PositionSen) of the temperature and humidity area.

[0115] Air conditioning operation status wide table: used to store the operation and control parameters of the air conditioner in the computer room, including working current, return air temperature and humidity, compressor pressure, remote power on / off control signal, temperature value setting assignment signal, etc., and define the two-dimensional positioning parameters (PositionAirCon) of the air conditioning area.

[0116] Space resource business static wide table: used to store all space resource location information (Position) and static attributes of the computer room, including: equipment name, code, type, location, business name, business importance (importance), computer room name and other fields, to form a resource wide table for associated use.

[0117] Use the PositionSer, PositionSen, PositionAirCon and Position attributes in the above four tables to place the business server, temperature and humidity sensor, air conditioner and computer room static resource data in the same computer room space to form relative position association data RelativePo.

[0118] Therefore, the importance of the target area can be obtained from the static wide table of space resource services.

[0119] The operating current and compressor pressure of the air conditioner in the target area in the past first preset time period are obtained from the air conditioner operation status table, and the historical load rate of the air conditioner in the target area is determined according to the operating current and the compressor pressure.

[0120] The historical computing resource usage of the servers in the target area within the first preset time period in the past is obtained from the server performance wide table.

[0121] The temperature of the server in the target area and the ambient temperature of the target area in the first preset time period in the past are obtained from the server performance wide table and the temperature and humidity signal wide table respectively, and the historical temperature difference between the temperature of the server in the target area and the ambient temperature of the target area is calculated based on the temperature of the server and the ambient temperature of the target area.

[0122] It should be noted that the server performance wide table, temperature and humidity signal wide table, air conditioning operation status wide table and space resource business static wide table can be collected in real time or at intervals of preset time periods, for example, once every hour. The specific settings can be made according to actual needs, and this is not limited in the embodiments of the present invention.

[0123] In the above embodiments, static wide tables, dynamic wide tables, and resource wide tables are used to associate and locate the native sensors such as the motherboard and CPU in the business server with the resource rack, and effectively associate them with the ambient temperature sensors monitored by the dynamic environment monitoring system to form a temperature measurement point field, which makes up for the pain points of insufficient and remote temperature measurement points in traditional computer rooms.

[0124] In an optional embodiment of the present invention, the multi-indicator prediction model includes a load rate prediction model, a computing resource utilization prediction model and a temperature difference prediction model, and the step 102 includes: inputting the historical load rate into the load rate prediction model to obtain the predicted load rate of the air conditioner within the second preset time period; inputting the historical computing resource utilization into the computing resource utilization prediction model to obtain the predicted computing resource utilization of the server within the second preset time period; inputting the historical temperature difference into the temperature difference prediction model to obtain the predicted temperature difference between the temperature of the server and the ambient temperature within the second preset time period.

[0125] Specifically, the multi-index prediction model includes a load rate prediction model, a computing resource usage prediction model and a temperature difference prediction model, and predicts multi-dimensional parameters of the target area within a second preset time period in the future.

[0126] Specifically, the historical load rate of the air conditioner in the past first preset time period is input into the load rate prediction model to obtain the predicted load rate of the air conditioner in the second preset time period following the first preset time period.

[0127] By inputting the historical computing resource usage of the server in the past first preset time period into the computing resource usage prediction model, the predicted computing resource usage of the server in the second preset time period following the first preset time period is obtained.

[0128] By inputting the historical temperature difference between the temperature of the server and the ambient temperature in the target area in the past first preset time period into the temperature difference prediction model, the predicted temperature difference between the temperature of the server and the ambient temperature in a second preset time period following the first preset time period is obtained.

[0129] In the above embodiment, the load rate prediction model, the computing resource utilization prediction model and the temperature difference prediction model are used to predict the predicted load rate of the air conditioner in the target area within the second preset time period in the future, the predicted computing resource utilization rate of the server, and the predicted temperature difference between the server temperature and the ambient temperature, so that the operation of the air conditioner can be accurately predicted based on the predicted parameters.

[0130] In an optional embodiment of the present invention, the method also includes: obtaining first training sample data, second training sample data and third training sample data; wherein the first training sample data is the historical load rate of the air conditioner in the second historical collection data of the target area, the second training sample data is the historical computing resource usage rate of the server in the second historical collection data of the target area, and the third training sample data is the historical temperature difference between the temperature of the server in the second historical collection data of the target area and the ambient temperature in the target area; training the load rate prediction model to be trained by the first training sample data to obtain a trained load rate prediction model; training the computing resource usage prediction model to be trained by the second training sample data to obtain a trained computing resource usage prediction model; training the temperature difference prediction model to be trained by the third training sample data to obtain a trained temperature difference prediction model.

[0131] Among them, the second historical collected data can be data collected from the target area in any time period, and the second historical collected data at least includes the historical load rate of the air conditioner in the target area, the historical computing resource utilization rate of the server in the target area, and the historical temperature difference between the temperature of the server in the target area and the ambient temperature of the target area.

[0132] The load rate prediction model to be trained is trained: the first training sample data can be the historical load rate of the air conditioner in any time period (such as time period A) in the second historical collection data, which also includes the actual load rate of the air conditioner in time period B following time period A. The first training sample data is input into the load rate prediction model to be trained, and the predicted load rate of the air conditioner in time period B is predicted. Then, the loss value between the predicted load rate and the actual load rate in time period B is calculated. When the loss value is less than a preset threshold or the loss value converges, it is determined that the load rate prediction model training is completed. Otherwise, the load rate prediction model to be trained continues to be trained through the first training sample data.

[0133] The computing resource utilization prediction model to be trained is trained: the second training sample data can be the historical computing resource utilization of the server in any time period (such as time period A) in the second historical collection data, which also includes the actual computing resource utilization of the server in time period B following time period A. The second training sample data is input into the computing resource utilization prediction model to be trained, and the predicted computing resource utilization of the server in time period B is predicted. Then, the loss value between the predicted computing resource utilization and the actual computing resource utilization in time period B is calculated. When the loss value is less than a preset threshold or the loss value converges, it is determined that the training of the computing resource utilization prediction model is completed. Otherwise, the computing resource utilization prediction model to be trained continues to be trained using the second training sample data.

[0134] The temperature difference prediction model to be trained is trained: the third training sample data can be the historical temperature difference between the temperature of the server in any time period (such as time period A) in the second historical collection data and the ambient temperature in the target area, which also includes the actual temperature difference between the temperature of the server in time period B following time period A and the ambient temperature in the target area. By inputting the third training sample data into the temperature difference prediction model to be trained, the predicted temperature difference between the temperature of the server in time period B and the ambient temperature in the target area is predicted, and then the loss value of the predicted temperature difference and the actual temperature difference in time period B is calculated. When the loss value is less than a preset threshold, or the loss value converges, it is determined that the training of the temperature difference prediction model is completed, otherwise the temperature difference prediction model to be trained continues to be trained through the third training sample data.

[0135] In an optional embodiment of the present invention, step 103 includes: determining an air conditioning control strategy based on the importance level, the predicted load rate, the predicted computing resource usage rate and the predicted temperature difference; and controlling the operation of the air conditioner through the air conditioning control strategy.

[0136] Specifically, after predicting the predicted load rate of the air conditioner, the predicted computing resource utilization rate of the server, and the predicted temperature difference between the temperature of the server and the ambient temperature in the target area in the second preset time period in the future through a multi-indicator prediction model, the air conditioning control strategy is determined based on the importance of the target area, the predicted load rate, the predicted computing resource utilization rate and the predicted temperature difference, so as to control the operation of the air conditioner according to the control strategy.

[0137] In an optional embodiment of the present invention, the importance level is divided into four categories: extremely high, high, medium and low, and the predicted load rate, the predicted computing resource usage rate and the predicted temperature difference are all divided into three categories: high, medium and low.

[0138] Specifically, a multi-dimensional and multi-time period scenario label set can be formed based on four key scenarios: business importance (degree of importance), temperature trend (temperature difference), CPU utilization trend (computing resource utilization), and air conditioning operation quality (predicted load rate).

[0139] 1) Business-important scenario labels: Use the business importance data in the static wide table data of the spatial resource business to identify the regional labels of RelativePo, and mark them according to the business importance into four categories: B extremely high (less than 5%), B high (15%), B medium (60%), and B low (20%).

[0140] 2) Temperature trend label: Use the temperature and humidity data in the temperature and humidity signal wide table to associate with the server temperature in the server performance wide table to form a detailed temperature field map, record the data at each moment (hour), and divide it into three categories according to the difference between the equipment and the temperature and humidity measurement points on an hourly basis: T high (above 5 degrees), T medium (2-5 degrees), and T low (within 2 degrees).

[0141] 3) CPU usage trend label: Use the CPU usage factor that most affects the server temperature rise in the server performance table to record the CPU usage at each moment (hour), and divide it into three categories per hour: C high (above 80%), C medium (40%-80%), and C low (below 40%)

[0142] 4) Air conditioner operation status label: Use the air conditioner operation parameters in the air conditioner operation status table, combined with the current air conditioner working current and compressor pressure to comprehensively evaluate the current load rate of the air conditioner and record it. It is divided into three categories on an hourly basis: K high (air conditioner load reaches more than 80%), K medium (air conditioner load is between 50% and 80%), and K low (below 50%).

[0143] It should be noted that the above numerical values ​​for dividing into extremely high, high, medium and low are only for exemplary description, and the marking division method can be specifically set according to actual needs, which is not limited in the embodiments of the present invention.

[0144] In an optional embodiment of the present invention, the control strategy at least includes: when the importance level is extremely high, the predicted load rate is high, and the predicted temperature difference is high, increasing the cooling capacity of the air conditioner; when the importance level is high, the predicted load rate is high, and the predicted temperature difference is high, increasing the cooling capacity of the air conditioner; when the importance level is high, the predicted load rate is high, and the predicted temperature difference is medium, maintaining the cooling capacity of the air conditioner; when the importance level is high, the predicted load rate is high, and the predicted temperature difference is low, Reduce the cooling capacity of the air conditioner; when the importance level is high, the predicted load rate is medium, and the predicted temperature difference is high, maintain the cooling capacity of the air conditioner; when the importance level is high, the predicted load rate is medium, and the predicted temperature difference is medium or low, reduce the cooling capacity of the air conditioner; when the importance level is medium, the predicted load rate is high, and the predicted temperature difference is high, increase the cooling capacity of the air conditioner; when the importance level is medium, the predicted load rate is high, and the predicted temperature difference is medium or low, reduce the cooling capacity of the air conditioner.

[0145] Specifically, there may be multiple control strategies for the air conditioner, and the corresponding control strategy may be determined according to the category of the importance of the target area, the category of the load rate of the air conditioner, the category of the predicted computing resources of the server, and the category of the predicted temperature difference between the temperature of the server and the ambient temperature, and then the operation of the air conditioner may be controlled by the control strategy. As an example, the control strategy may be as follows:

[0146] 1) For the equipment in the 5% area with extremely high business-critical scenarios, the air conditioning cooling capacity is prioritized at all times to ensure the highest stability of the equipment in this area. Based on the two scenario tags of CPU usage and temperature trend, the cooling capacity of this area is increased when C is high and T is high to ensure the optimal environment in this area.

[0147] 2) For the B high and T high scenarios, during the C high time period, set the air conditioning temperature in advance (e.g., about 40 minutes) and increase the cooling capacity as soon as possible to ensure that the regional temperature rise does not increase on a large scale before responding in advance; during the C medium time period, maintain the cooling supply unchanged, monitor the temperature changes in the area at all times, and ensure the temperature supply of cooling capacity in the area; during the C low time period, combined with the regional temperature conditions, appropriately reduce the cooling capacity while ensuring that T high does not further increase, and give priority to supplying cooling capacity to other areas.

[0148] 3) For the B high and T medium scenario, in the C high period, maintain the cooling supply unchanged, monitor the temperature changes in the area at all times, and ensure that the temperature supply of cooling in the area is guaranteed when T medium does not further upgrade to T high; in the C medium period, combine the regional temperature conditions and appropriately reduce the cooling capacity while ensuring that the regional temperature does not rise further. In the C low period, further reduce the cooling supply, monitor the temperature changes in the area at all times, and ensure that the temperature rise is controllable.

[0149] 4) For the B medium and T high scenarios, during the C high time period, set the air conditioning temperature in advance (e.g., about 20 minutes) and gradually increase the cooling capacity to ensure that the regional temperature rise does not increase on a large scale before responding in advance; during the C medium time period, maintain the cooling supply unchanged, and monitor the temperature changes in the area at all times. If there is a temperature drop trend, slowly reduce the cooling supply synchronously; during the C low time period, reduce the cooling capacity based on the regional temperature conditions, and give priority to supplying cooling capacity to other areas.

[0150] 5) For other combined scenarios, based on the operation of the above four logics, while ensuring that the temperature field does not cause equipment alarms, the cooling capacity is preferentially supplied to other scenarios. The changes in the temperature field in the area are monitored at all times. While ensuring that the temperature field does not become extremely high, the cooling capacity is continuously reduced to reduce carbon emissions.

[0151] Of course, as the business load changes, the four scene tags will be updated and associated synchronously according to the actual collected values ​​to objectively conform to the actual situation of each computer room. The logic of the five types of air conditioner control can be customized to add, modify, and delete operations according to the actual situation of the maintenance personnel. Only any combination and trigger of the four tags is needed to accurately control the air conditioner on an hourly basis. It is simple to implement, with less on-site modification, easy for maintenance personnel to set up, and can be quickly started.

[0152] In order for those skilled in the art to better understand the embodiments of the present invention, an exemplary description is given below through an embodiment:

[0153] 1. Define the wide table associated with the monitoring of business equipment in the computer room

[0154] In order to simplify the complex temperature field of the computer room area, three types of real-time data collection wide tables, namely server performance wide table, temperature and humidity signal wide table, and air conditioning operation status wide table, and a static wide table of computer room space resource data were created. The parameter Position that describes the positioning information in the space resource is used to identify the location information of each device in each wide table.

[0155] Use the PositionSer, PositionSen, PositionAirCon and Position attributes in the above four tables to place the business server, temperature and humidity sensor, air conditioner and computer room static resource data together in the same computer room space to form relative position association data RelativePo for the next step of data association and label call.

[0156] 2. Collect real-time data such as CPU and temperature, associate them, and record them by time period

[0157] The first step is to periodically collect real-time data in the wide table, including: CPU temperature, CPU load, data flow, motherboard temperature in the server performance wide table, temperature and humidity data in the temperature and humidity signal, return air temperature and humidity, remote power on / off control signal, and temperature value setting assignment signal data in the air conditioning operation status wide table.

[0158] The second step is to associate all temperature data to form a temperature measurement point field. According to the spatial relative position attribute RelativePo, each device in the computer room is divided into regions (generally a computer room is divided into three regions). RelativePo is used to associate the motherboard and CPU of the nearest server to obtain real-time temperature data, which is associated and merged with the dynamic environment temperature and humidity collection data, solving the problem of insufficient temperature measurement points in old computer rooms.

[0159] The third step is to form a multi-dimensional and multi-time period scenario label set based on the four key scenarios of business importance, temperature trend, CPU usage trend, and air conditioning operation quality.

[0160] A, Business-important scenario labels: Use the business importance data in the static wide table data of the spatial resource business to identify the regional labels of RelativePo, and mark them according to the business importance into four categories: B extremely high (less than 5%), B high (15%), B medium (60%), and B low (20%)

[0161] B. Temperature trend label: Use the temperature and humidity data in the temperature and humidity signal wide table to associate with the server temperature in the server performance wide table to form a detailed temperature field map, record the data at each moment (hour), and divide it into three categories according to the difference between the device and the temperature and humidity measurement point per hour: T high (above 5 degrees), T medium (2-5 degrees), T low (within 2 degrees)

[0162] C, CPU usage trend label: Use the CPU usage factor that most affects the server temperature rise in the server performance table, record the CPU usage at each moment (hour), and divide it into three categories per hour: C high (above 80%), C medium (40%-80%), C low (below 40%)

[0163] D. Air conditioner operation quality label: Use the air conditioner operation parameters in the air conditioner operation status table, combined with the current air conditioner working current and compressor pressure to comprehensively evaluate the current load rate of the air conditioner and record it. It is divided into three categories per hour: K high (air conditioner load reaches more than 80%), K medium (air conditioner load is between 50% and 80%), and K low (below 50%).

[0164] 3. Build a data model to learn and output air conditioning control strategies in real time based on labels of different time periods

[0165] By combining four scenario label sets, namely business-important scenario labels, temperature trend labels, CPU usage trend labels, and air conditioning operation quality labels, a data model is built using a machine learning algorithm. Ultimately, the historical time data of massive time periods is analyzed into usable labels (T, C, K) that can predict the next time period, thereby achieving real-time and precise control of air conditioning to reduce carbon emissions.

[0166] The main algorithm models are as follows:

[0167] The DeepAR (RNN+Embedding) algorithm is used to predict the heat generation of each area in the computer room based on the data of each time period. By continuously collecting data from three wide tables of equipment performance, accumulating at least one week as a historical data set, according to the computer room space conditions, sampling point distribution and different equipment models in the resource static data table, a cooling influence model, an equipment energy consumption and cooling capacity association model, a computer room temperature prediction model and a cold station control logic model are established respectively, and the cooling capacity redundancy is evaluated. The RNN network algorithm is used to train the historical data, build a multi-index prediction model, and generate a predictable label (T, C, K) value for the next time period.

[0168] Data model process: Algorithm input: current cooling parameter configuration, temperature sensor number, temperature values ​​of the past 24 hours, etc., and static space data (first historical collection data). Algorithm output: under the current cooling configuration parameters, the expected T, C, and K values ​​of each temperature sensor in the next 10 minutes.

[0169] The algorithm randomly samples the training data, taking the randomly selected time point t as the original input data and t~t+L as the output. Taking the training sample generated at time t as an example, the input is: the relevant features of the current time point, and the LSTM (Long Short Term Memory) output corresponding to each feature at the previous time point; the model output is: the true value at time t. Repeatedly learn the samples at time t+1~t+L; repeat the random sampling data for model training.

[0170] 4. Determine the air conditioning control strategy based on the predicted T, C, K and B value logic to control the operation of the air conditioner through the control strategy.

[0171] The detailed correlation analysis process can be found in Figure 2 , shows a flow chart of a data association analysis provided in an embodiment of the present invention.

[0172] Dynamic wide table: Collect 3 wide tables of dynamic performance of equipment in the computer room (server performance wide table, temperature and humidity signal wide table, air conditioning operation status wide table). These devices are generally located in a relatively fixed position in the computer room when the office is established, and the relative position information RelativePo can be quickly formed through PositionSer, PositionSen, and PositionAirCon.

[0173] Static wide table: Maintain a dynamic wide table, taken from the resource system, mainly storing space resource business, using position information Position and 3 dynamic wide tables to quickly establish relative position information

[0174] Define 4 types of key tags: Based on the four key scenarios of business importance, temperature trend, CPU usage trend, and air conditioning operation quality, form 4 multi-dimensional and multi-time scenario tag sets

[0175] Machine learning: Use the DeepAR (RNN+Embedding) algorithm to build a mathematical model with historical collection parameters as input to generate predicted label values ​​for logical analysis.

[0176] Logical analysis: Based on the final predicted B, T, C, and K values, the computer room air conditioner is precisely controlled in five scenarios, and the logical analysis rules can be customized.

[0177] For detailed data transmission process, please refer to the attached Figure 3 , shows a flow chart of data transmission provided in an embodiment of the present invention.

[0178] 1. Complete the temperature and humidity signal wide table and air conditioning operation status wide table data collection through the unified data collection platform B interface, and obtain the server performance wide table performance data collection through the business network management. Each table contains a unique field (PositionSer, PositionSen, PositionAirCon) that identifies the position and the characteristic values ​​of each type of equipment. Use the POST protocol to form an interface, use the room name as the input parameter, and call the collection device attributes in the same room.

[0179] 2. Obtain the computer room space resource location information Position and equipment static attribute list from the resource system, including: equipment name, code, type, location, business name, business importance, computer room name and other fields, to form a resource wide table for associated use.

[0180] 3. Combine the historical performance values ​​and current performance values ​​in the three performance wide tables with the resource space data to form a massive time period label set. Use the DeepAR (RNN+Embedding) algorithm to build a data model and generate a predictable final label value for the next time period.

[0181] 4. According to the final multi-dimensional tag value, logical analysis is performed in five scenarios, and the air conditioning control strategy for the next time period is finally output, and the control command is issued through the B interface.

[0182] In the above embodiment, by introducing business rack server temperature monitoring data, CPU load data, historical temperature (regional measurement point + server self-test) data, air conditioning operation data, computer room space resource data, etc. for multi-dimensional association and labeling, a multi-dimensional and precise air conditioning management and control strategy is implemented based on multiple labels, and the machine learning DeepAR algorithm is used to predict the server heating curve. The active pre-intervention method of controlling the air conditioning to lower the temperature in advance replaces the original post-management method of consuming more energy to set a lower temperature to lower the regional temperature after the regional temperature field rises, thereby realizing a more precise and flexible method of labeling each air conditioner, further saving energy consumption in the computer room and reducing carbon emissions.

[0183] Compared with the existing most similar technology, the technical solution proposed in the embodiment of the present invention has the following advantages:

[0184] 1) In the prior art, the number of temperature measurement points in the area is insufficient or the distance is far, making it difficult to accurately control the air conditioner. In the embodiments of the present invention, static wide tables, dynamic wide tables, and resource wide tables are used to associate the original sensors such as the motherboard and CPU in the business server with the resource rack, and effectively associate them with the dynamic environment monitoring ambient temperature sensor to form a temperature field, which makes up for the pain points of insufficient and far-flung temperature measurement points in traditional computer rooms. Further combined with the combination of four types of tags, the regional temperature thermal field can be used to accurately control the air conditioner and improve the cooling efficiency.

[0185] 2) The existing technology is aimed at the high business density of a single rack in an IDC computer room, which heats up quickly in a short period of time and cannot be suppressed by air conditioners at a constant temperature. In the embodiment of the present invention, the historical server CPU usage curve is used, combined with the historical ambient temperature rise curve as the main method, and the DeepAR (RNN+Embedding) algorithm is used to predict the heat generation of each area in the computer room according to the data of each time period. It can effectively detect the increase in heat generation in advance, and accurately control the air conditioner in advance to gradually suppress it in advance, so as to ensure that when the regional temperature rise does not increase on a large scale, it can respond in advance, reduce the overall energy consumption of refrigeration in the area, and further reduce carbon emissions.

[0186] 3) The existing scheme controls the air conditioning mode in a single form and lacks flexibility. In the embodiment of the present invention, a multi-dimensional tag analysis model is adopted, which can be updated and associated synchronously according to the actual collected values ​​to objectively meet the actual conditions of each computer room. The logic of the five types of air conditioning control can be customized to add, modify, and delete operations according to the actual conditions of the maintenance personnel. Only any combination and trigger of the four tags are needed to accurately control the air conditioning by the hour. The implementation is simple, with less on-site modification, easy for maintenance personnel to set up, and can be quickly started.

[0187] It should be noted that, for the sake of simplicity, the method embodiments are described as a series of action combinations, but those skilled in the art should be aware that the embodiments of the present invention are not limited by the order of the actions described, because according to the embodiments of the present invention, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions involved are not necessarily required by the embodiments of the present invention.

[0188] Reference Figure 4 , shows a structural block diagram of an air-conditioning operation control device provided in an embodiment of the present invention. The air-conditioning operation control device may specifically include the following modules:

[0189] The data acquisition module 401 is used to acquire the importance of the target area and the first historical collected data of the target area in the past first preset time period; wherein the first historical collected data includes the historical load rate of the air conditioner in the target area, the historical computing resource usage rate of the server, and the historical temperature difference between the temperature of the server and the ambient temperature in the target area;

[0190] The result prediction module 402 is used to input the first historical collection data into a multi-index prediction model to obtain a predicted load rate of the air conditioner, a predicted computing resource usage rate of the server, and a predicted temperature difference between the temperature of the server and the ambient temperature in a second preset time period following the first preset time period; the multi-index prediction model is trained by the second historical collection data of the target area;

[0191] The operation control module 403 is used to control the operation of the air conditioner according to the importance degree, the predicted load rate, the predicted computing resource usage rate and the predicted temperature difference.

[0192] In an optional embodiment of the present invention, the data acquisition module includes:

[0193] The table construction submodule is used to construct a server performance wide table, a temperature and humidity signal wide table, an air conditioning operation status wide table, and a space resource business static wide table;

[0194] The data collection submodule is used to collect the temperature of the servers in each area and the computing resource usage rate of the servers through the server performance wide table; collect the ambient temperature of each area of ​​the computer room through the temperature and humidity signal wide table; collect the working current and compressor pressure of the air conditioners in each area through the air conditioner operation wide table; collect the importance of the services in each area through the space resource service static wide table;

[0195] The data acquisition submodule is used to obtain the importance of the target area from the spatial resource business static wide table; obtain the working current and compressor pressure of the air conditioner in the target area in the past first preset time period from the air conditioner operation status wide table, and determine the historical load rate of the air conditioner in the target area based on the working current and compressor pressure; obtain the historical computing resource usage rate of the server in the target area in the past first preset time period from the server performance wide table; obtain the temperature of the server in the target area and the ambient temperature of the target area in the past first preset time period from the server performance wide table and the temperature and humidity signal wide table respectively, and calculate the historical temperature difference between the temperature of the server in the target area and the ambient temperature of the target area based on the temperature of the server and the ambient temperature of the target area.

[0196] In an optional embodiment of the present invention, the multi-index prediction model includes a load rate prediction model, a computing resource usage prediction model and a temperature difference prediction model, and the result prediction module includes:

[0197] A load rate prediction submodule, used for inputting the historical load rate into the load rate prediction model to obtain a predicted load rate of the air conditioner within the second preset time period;

[0198] A usage rate prediction submodule, used for inputting the historical computing resource usage rate into the computing resource usage rate prediction model to obtain the predicted computing resource usage rate of the server within the second preset time period;

[0199] The temperature difference prediction submodule is used to input the historical temperature difference into the temperature difference prediction model to obtain the predicted temperature difference between the temperature of the server and the ambient temperature within the second preset time period.

[0200] In an optional embodiment of the present invention, it also includes:

[0201] A sample data acquisition module is used to acquire first training sample data, second training sample data and third training sample data; wherein the first training sample data is the historical load rate of the air conditioner in the second historical collection data of the target area, the second training sample data is the historical computing resource usage rate of the server in the second historical collection data of the target area, and the third training sample data is the historical temperature difference between the temperature of the server in the second historical collection data of the target area and the ambient temperature in the target area;

[0202] A first model training module, used for training the load rate prediction model to be trained by using the first training sample data to obtain a trained load rate prediction model;

[0203] A second model training module is used to train the computing resource usage prediction model to be trained using the second training sample data to obtain a trained computing resource usage prediction model;

[0204] The third model training module is used to train the temperature difference prediction model to be trained by using the third training sample data to obtain a trained temperature difference prediction model.

[0205] In an optional embodiment of the present invention, the operation control module includes

[0206] a strategy determination submodule, configured to determine an air conditioning control strategy according to the importance degree, the predicted load rate, the predicted computing resource usage rate and the predicted temperature difference;

[0207] The operation control submodule is used to control the operation of the air conditioner through the air conditioner control strategy.

[0208] In an optional embodiment of the present invention, the importance level is divided into four categories: extremely high, high, medium and low, and the predicted load rate, the predicted computing resource usage rate and the predicted temperature difference are all divided into three categories: high, medium and low.

[0209] In an optional embodiment of the present invention, the control strategy at least includes:

[0210] When the importance level is extremely high, the predicted load rate is high, and the predicted temperature difference is high, increasing the cooling capacity of the air conditioner;

[0211] When the importance level is high, the predicted load rate is high, and the predicted temperature difference is high, the cooling capacity of the air conditioner is increased; when the importance level is high, the predicted load rate is high, and the predicted temperature difference is medium, the cooling capacity of the air conditioner is maintained; when the importance level is high, the predicted load rate is high, and the predicted temperature difference is low, the cooling capacity of the air conditioner is reduced;

[0212] When the importance level is high, the predicted load rate is medium, and the predicted temperature difference is high, the cooling capacity of the air conditioner is maintained; when the importance level is high, the predicted load rate is medium, and the predicted temperature difference is medium or low, the cooling capacity of the air conditioner is reduced;

[0213] When the importance level is medium, the predicted load rate is high, and the predicted temperature difference is high, the cooling capacity of the air conditioner is increased; when the importance level is medium, the predicted load rate is high, and the predicted temperature difference is medium or low, the cooling capacity of the air conditioner is reduced.

[0214] As for the device embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0215] In addition, an embodiment of the present invention further provides an electronic device, including: a processor, a memory, and a computer program stored in the memory and executable on the processor. When the computer program is executed by the processor, the various processes of the above-mentioned data acquisition method embodiment are implemented, and the same technical effect can be achieved. To avoid repetition, it will not be described here.

[0216] The embodiment of the present invention further provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, each process of the above-mentioned data acquisition method embodiment is implemented, and the same technical effect can be achieved. To avoid repetition, it is not repeated here. The computer-readable storage medium is, for example, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0217] Figure 5 A schematic diagram of the hardware structure of an electronic device for implementing various embodiments of the present invention.

[0218] The electronic device 500 includes but is not limited to: a radio frequency unit 501, a network module 502, an audio output unit 503, an input unit 504, a sensor 505, a display unit 506, a user input unit 507, an interface unit 508, a memory 509, a processor 510, and a power supply 511. Those skilled in the art will appreciate that Figure 5 The electronic device structure shown in the figure does not constitute a limitation on the electronic device, and the electronic device may include more or fewer components than shown in the figure, or combine certain components, or arrange the components differently. In the embodiments of the present invention, the electronic device includes but is not limited to a mobile phone, a tablet computer, a laptop computer, a PDA, a vehicle-mounted terminal, a wearable device, and a pedometer.

[0219] It should be understood that in the embodiment of the present invention, the radio frequency unit 501 can be used for receiving and sending signals during information transmission or communication. Specifically, after receiving downlink data from the base station, it is sent to the processor 510 for processing; in addition, uplink data is sent to the base station. Generally, the radio frequency unit 501 includes but is not limited to an antenna, at least one amplifier, a transceiver, a coupler, a low noise amplifier, a duplexer, etc. In addition, the radio frequency unit 501 can also communicate with the network and other devices through a wireless communication system.

[0220] The electronic device provides users with wireless broadband Internet access through the network module 502, such as helping users to send and receive emails, browse web pages, and access streaming media.

[0221] The audio output unit 503 can convert the audio data received by the RF unit 501 or the network module 502 or stored in the memory 509 into an audio signal and output it as sound. Moreover, the audio output unit 503 can also provide audio output related to a specific function performed by the electronic device 500 (for example, a call signal reception sound, a message reception sound, etc.). The audio output unit 503 includes a speaker, a buzzer, a receiver, etc.

[0222] The input unit 504 is used to receive audio or video signals. The input unit 504 may include a graphics processor (GPU) 5041 and a microphone 5042. The graphics processor 5041 processes the image data of a static picture or video obtained by an image capture device (such as a camera) in a video capture mode or an image capture mode. The processed image frame can be displayed on the display unit 506. The image frame processed by the graphics processor 5041 can be stored in the memory 509 (or other storage medium) or sent via the radio frequency unit 501 or the network module 502. The microphone 5042 can receive sound and can process such sound into audio data. The processed audio data can be converted into a format output that can be sent to a mobile communication base station via the radio frequency unit 501 in the case of a telephone call mode.

[0223] The electronic device 500 also includes at least one sensor 505, such as a light sensor, a motion sensor, and other sensors. Specifically, the light sensor includes an ambient light sensor and a proximity sensor, wherein the ambient light sensor can adjust the brightness of the display panel 5061 according to the brightness of the ambient light, and the proximity sensor can turn off the display panel 5061 and / or the backlight when the electronic device 500 is moved to the ear. As a type of motion sensor, the accelerometer sensor can detect the magnitude of acceleration in each direction (generally three axes), and can detect the magnitude and direction of gravity when stationary, which can be used to identify the posture of the electronic device (such as horizontal and vertical screen switching, related games, magnetometer posture calibration), vibration recognition related functions (such as pedometer, tapping), etc.; the sensor 505 can also include a fingerprint sensor, a pressure sensor, an iris sensor, a molecular sensor, a gyroscope, a barometer, a hygrometer, a thermometer, an infrared sensor, etc., which will not be repeated here.

[0224] The display unit 506 is used to display information input by the user or information provided to the user. The display unit 506 may include a display panel 5061, which may be configured in the form of a liquid crystal display (LCD), an organic light-emitting diode (OLED), or the like.

[0225] The user input unit 507 can be used to receive input digital or character information, and to generate key signal input related to user settings and function control of the electronic device. Specifically, the user input unit 507 includes a touch panel 5071 and other input devices 5072. The touch panel 5071, also known as a touch screen, can collect the user's touch operation on or near it (such as the user's operation on the touch panel 5071 or near the touch panel 5071 using any suitable object or accessory such as a finger, stylus, etc.). The touch panel 5071 may include two parts: a touch detection device and a touch controller. Among them, the touch detection device detects the user's touch orientation, detects the signal brought by the touch operation, and transmits the signal to the touch controller; the touch controller receives the touch information from the touch detection device, converts it into the contact point coordinates, and then sends it to the processor 510, receives the command sent by the processor 510 and executes it. In addition, the touch panel 5071 can be implemented using multiple types such as resistive, capacitive, infrared and surface acoustic waves. In addition to the touch panel 5071, the user input unit 507 may also include other input devices 5072. Specifically, other input devices 5072 may include but are not limited to a physical keyboard, function keys (such as volume control keys, switch keys, etc.), a trackball, a mouse, and a joystick, which are not described in detail here.

[0226] Furthermore, the touch panel 5071 may be overlaid on the display panel 5061. When the touch panel 5071 detects a touch operation on or near it, it is transmitted to the processor 710 to determine the type of touch event. Then, the processor 510 provides corresponding visual output on the display panel 5061 according to the type of touch event. Figure 5 In the figure, the touch panel 5071 and the display panel 5061 are used as two independent components to realize the input and output functions of the electronic device. However, in some embodiments, the touch panel 5071 and the display panel 5061 can be integrated to realize the input and output functions of the electronic device, which is not limited here.

[0227] The interface unit 508 is an interface for connecting an external device to the electronic device 500. For example, the external device may include a wired or wireless headset port, an external power supply (or battery charger) port, a wired or wireless data port, a memory card port, a port for connecting a device with an identification module, an audio input / output (I / O) port, a video I / O port, a headphone port, etc. The interface unit 507 may be used to receive input (e.g., data information, power, etc.) from an external device and transmit the received input to one or more elements within the electronic device 500 or may be used to transmit data between the electronic device 500 and an external device.

[0228] The memory 509 can be used to store software programs and various data. The memory 509 can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, an application required for at least one function (such as a sound playback function, an image playback function, etc.), etc.; the data storage area can store data created according to the use of the mobile phone (such as audio data, a phone book, etc.), etc. In addition, the memory 509 can include a high-speed random access memory, and can also include a non-volatile memory, such as at least one disk storage device, a flash memory device, or other volatile solid-state storage devices.

[0229] The processor 510 is the control center of the electronic device. It uses various interfaces and lines to connect various parts of the entire electronic device. By running or executing software programs and / or modules stored in the memory 509 and calling data stored in the memory 509, it performs various functions of the electronic device and processes data, thereby monitoring the electronic device as a whole. The processor 510 may include one or more processing units; preferably, the processor 510 may integrate an application processor and a modem processor, wherein the application processor mainly processes the operating system, user interface, and application programs, and the modem processor mainly processes wireless communications. It is understandable that the above-mentioned modem processor may not be integrated into the processor 510.

[0230] The electronic device 500 may also include a power source 511 (such as a battery) for supplying power to various components. Preferably, the power source 511 may be logically connected to the processor 510 through a power management system, thereby implementing functions such as charging, discharging, and power consumption management through the power management system.

[0231] In addition, the electronic device 500 includes some functional modules not shown, which will not be described in detail here.

[0232] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the existence of other identical elements in the process, method, article or device including the element.

[0233] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, a magnetic disk, or an optical disk), and includes a number of instructions for enabling a terminal (which can be a mobile phone, a computer, a server, an air conditioner, or a network device, etc.) to execute the methods described in each embodiment of the present invention.

[0234] The embodiments of the present invention are described above in conjunction with the accompanying drawings, but the present invention is not limited to the above-mentioned specific implementation modes, which are merely illustrative rather than restrictive. Under the guidance of the present invention, ordinary technicians in this field can also make many forms without departing from the scope of protection of the present invention and the claims, all of which are within the protection of the present invention.

[0235] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed in the embodiments of the present invention can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0236] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0237] In the embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as 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 devices or units, which can be electrical, mechanical or other forms.

[0238] 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 network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0239] In addition, each functional unit in each embodiment of the present invention 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.

[0240] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including 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 the method described in each embodiment of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as USB flash drives, mobile hard disks, ROM, RAM, magnetic disks, or optical disks.

[0241] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. An air conditioning operation control method, characterized in that: The method comprises: Obtaining the importance of the target area and first historical collected data of the target area in the past first preset time period; wherein the first historical collected data includes a historical load rate of the air conditioner in the target area, a historical computing resource usage rate of the server, and a historical temperature difference between the temperature of the server and the ambient temperature in the target area; Inputting the first historical collected data into a multi-index prediction model to obtain a predicted load rate of the air conditioner, a predicted computing resource usage rate of the server, and a predicted temperature difference between the temperature of the server and the ambient temperature in a second preset time period following the first preset time period; the multi-index prediction model is trained using the second historical collected data of the target area; controlling the operation of the air conditioner according to the importance degree, the predicted load rate, the predicted computing resource usage rate, and the predicted temperature difference; The obtaining of the importance of the target area and the first historical data collected from the target area in the past first preset time period includes: Construct a wide table of server performance, wide table of temperature and humidity signals, wide table of air conditioning operation status, and wide table of static space resource services; The server performance wide table is used to collect the temperature of each regional server and the computing resource usage rate of the server; the temperature and humidity signal wide table is used to collect the ambient temperature of each area of ​​the computer room; the air conditioner operation wide table is used to collect the working current and compressor pressure of each regional air conditioner; the space resource service static wide table is used to collect the importance of each regional service; Obtain the importance of the target area from the spatial resource service static wide table; obtain the working current and compressor pressure of the air conditioner in the target area in the past first preset time period from the air conditioner operation status wide table, and determine the historical load rate of the air conditioner in the target area based on the working current and the compressor pressure; obtain the historical computing resource usage rate of the server in the target area in the past first preset time period from the server performance wide table; obtain the temperature of the server in the target area and the ambient temperature of the target area in the past first preset time period from the server performance wide table and the temperature and humidity signal wide table respectively, and calculate the historical temperature difference between the temperature of the server in the target area and the ambient temperature of the target area based on the temperature of the server and the ambient temperature of the target area; The multi-index prediction model includes a load rate prediction model, a computing resource usage rate prediction model and a temperature difference prediction model. The first historical collection data is input into the multi-index prediction model to predict the predicted load rate of the air conditioner, the predicted computing resource usage rate of the server, and the predicted temperature difference between the temperature of the server and the ambient temperature in a second preset time period following the first preset time period, including: Inputting the historical load rate into the load rate prediction model to obtain a predicted load rate of the air conditioner within the second preset time period; Inputting the historical computing resource usage rate into the computing resource usage rate prediction model to obtain the predicted computing resource usage rate of the server within the second preset time period; The historical temperature difference is input into the temperature difference prediction model to obtain a predicted temperature difference between the temperature of the server and the ambient temperature within the second preset time period.

2. The method according to claim 1, characterized in that The method further comprises: Acquire first training sample data, second training sample data, and third training sample data; wherein the first training sample data is the historical load rate of the air conditioner in the second historical collection data of the target area, the second training sample data is the historical computing resource usage rate of the server in the second historical collection data of the target area, and the third training sample data is the historical temperature difference between the temperature of the server in the second historical collection data of the target area and the ambient temperature in the target area; Training the load rate prediction model to be trained using the first training sample data to obtain a trained load rate prediction model; The computing resource usage prediction model to be trained is trained by using the second training sample data to obtain a trained computing resource usage prediction model; The temperature difference prediction model to be trained is trained by using the third training sample data to obtain a trained temperature difference prediction model.

3. The method according to any one of claims 1 to 2, characterized in that: The control of the operation of the air conditioner according to the importance degree, the predicted load rate, the predicted computing resource usage rate and the predicted temperature difference includes: determining an air conditioning control strategy according to the importance degree, the predicted load rate, the predicted computing resource usage rate, and the predicted temperature difference; The operation of the air conditioner is controlled by the air conditioner control strategy.

4. The method according to claim 3, characterized in that The importance levels are divided into four categories: extremely high, high, medium and low. The predicted load rate, the predicted computing resource usage rate and the predicted temperature difference are all divided into three categories: high, medium and low.

5. The method according to claim 4, characterized in that The control strategy at least includes: When the importance level is extremely high, the predicted load rate is high, and the predicted temperature difference is high, increasing the cooling capacity of the air conditioner; When the importance level is high, the predicted load rate is high, and the predicted temperature difference is high, the cooling capacity of the air conditioner is increased; when the importance level is high, the predicted load rate is high, and the predicted temperature difference is medium, the cooling capacity of the air conditioner is maintained; when the importance level is high, the predicted load rate is high, and the predicted temperature difference is low, the cooling capacity of the air conditioner is reduced; When the importance level is high, the predicted load rate is medium, and the predicted temperature difference is high, the cooling capacity of the air conditioner is maintained; when the importance level is high, the predicted load rate is medium, and the predicted temperature difference is medium or low, the cooling capacity of the air conditioner is reduced; When the importance level is medium, the predicted load rate is high, and the predicted temperature difference is high, the cooling capacity of the air conditioner is increased; when the importance level is medium, the predicted load rate is high, and the predicted temperature difference is medium or low, the cooling capacity of the air conditioner is reduced.

6. An air conditioning operation control device, characterized in that: The air conditioning operation control method according to any one of claims 1 to 5 comprises: A data acquisition module, used to acquire the importance of a target area and first historical collected data of the target area in the past first preset time period; wherein the first historical collected data includes a historical load rate of the air conditioner in the target area, a historical computing resource usage rate of the server, and a historical temperature difference between the temperature of the server and the ambient temperature in the target area; a result prediction module, for inputting the first historical collection data into a multi-index prediction model, and obtaining a predicted load rate of the air conditioner, a predicted computing resource usage rate of the server, and a predicted temperature difference between the temperature of the server and the ambient temperature in a second preset time period following the first preset time period; the multi-index prediction model is obtained by training the second historical collection data of the target area; An operation control module is used to control the operation of the air conditioner according to the importance degree, the predicted load rate, the predicted computing resource usage rate and the predicted temperature difference.

7. An electronic device, characterized in that: It includes a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other through the communication bus; The memory is used to store computer programs; The processor is used to implement the method according to any one of claims 1 to 5 when executing the program stored in the memory.

8. A computer-readable storage medium, characterized in that: Instructions are stored thereon, which, when executed by one or more processors, cause the processors to perform the method according to any one of claims 1 to 5.

Images