Power calculation device for thermal load prediction control of data machine room

Abstract

The invention relates to the technical field of intelligent control, in particular to a power calculation device for thermal load prediction control of a data machine room, and the device comprises an electric power and electric quantity collection module which is used for collecting electric quantity information of a cabinet of an air conditioner and transmitting the electric quantity information to an air conditioning unit communication module; the temperature acquisition module is used for acquiring air outlet data dispersed in each cabinet or machine position and transmitting the air outlet data to the air conditioning unit communication module; the air conditioning unit communication module is used for transmitting the electric quantity information and the air outlet data to the air conditioning processing module; and a trained load prediction model is embedded in the air conditioner processing module, and the air conditioner processing module is used for carrying out load change prediction on the electric quantity information and the air outlet data based on the trained load prediction model and adjusting operation parameters of end equipment of the air conditioner based on a prediction result. According to the device, the calculation load and air conditioner energy front optimization matching can be achieved, the high-load risk caused by response lag is avoided, efficient transmission and distribution and heat exchange of cooling capacity are achieved, and system energy is saved.

Description

Technical Field

[0001] The present invention relates to the field of intelligent control technology, and in particular to a computing device for predicting and controlling heat load in a data center in the field of intelligent control technology. Background Art

[0002] The power load is the primary reference variable for the heat load in a data center. When computing services change, the power load also changes. Due to the thermal inertia of the air conditioning system, changes in the cabinet outlet air temperature lag behind changes in the power load, and changes in the return air temperature of the air handling unit (AHU) or precision air conditioner lag behind the cabinet outlet air temperature. The air conditioning system's delayed response to the computing load causes some cabinets or computing servers with higher loads to experience risks such as high temperature or thermal shutdown. To avoid this risk, in high-load-density data centers, the supply air temperature is often set too low, and the air supply volume is too large, which reduces the energy efficiency of the data center's cooling system. Summary of the Invention

[0003] The purpose of the present invention is to provide an intelligent anticipatory control method for extreme stability, and the technical solutions adopted are as follows:

[0004] In a first aspect, an embodiment of the present invention provides a computing device for predicting and controlling heat load in a data center, the device comprising:

[0005] Conditioning processing module, air conditioning unit communication module, power collection module, temperature collection module; Among them:

[0006] The power and electricity collection module is used to collect the power information of the air-conditioning cabinet and transmit it to the air-conditioning unit communication module;

[0007] The temperature acquisition module is used to collect air outlet data distributed in each cabinet or unit position and transmit it to the air conditioning unit communication module;

[0008] The air conditioning unit communication module is used to transmit the power information and the air output data to the air conditioning processing module;

[0009] The air conditioning processing module is embedded with a trained load prediction model, which is used to predict load changes on the power information and the air outlet data based on the trained load prediction model, and adjust the operating parameters of the terminal equipment of the air conditioner based on the prediction results.

[0010] In some possible implementations, the power and electricity collection module is provided in a row cabinet and is further configured to collect power information of the entire row of cabinets in the row cabinet; collect the operating status of the entire row of cabinets, and construct a power load characteristic model of each cabinet in the entire row of cabinets based on the operating status and corresponding operating time;

[0011] The air conditioning processing module is further configured to use the trained load prediction model to perform load change prediction based on the power information, the air outlet data, and the power load characteristic model of each cabinet to obtain the prediction result.

[0012] In some possible implementations, the air conditioning processing module includes: a protocol analysis module, an automatic data screening module, and a machine learning and deep learning module; wherein:

[0013] The protocol analysis module has a data scanning function for determining the communication protocol of the received power information and the air output data;

[0014] The automatic data screening module is configured to determine the processing result, the power information, and the change information of the air output data; based on the power information and the change information of the air output data, determine the positional relationship between the power monitoring points corresponding to the air output data and the power information;

[0015] The machine learning and deep learning module is used to adopt the trained load forecasting model to perform load change forecasting based on the change information to obtain the prediction result.

[0016] In some possible implementations, the air conditioning processing module further includes:

[0017] The air conditioning unit is used to read the air supply temperature and fan frequency of the terminal device of the air conditioner, and call the air conditioning unit communication module to adjust the air supply temperature, and change the air supply volume and air supply pressure of the air conditioner based on the fan frequency.

[0018] In some possible implementations, the automatic data screening module is further used to determine position change information between the air outlet data and the monitoring power points corresponding to the power information; wherein the change information includes: the position change information.

[0019] In some possible implementations, the automatic data screening module is further configured to establish, based on the change information of the power information, a time model between the change information of the power information and the air flow data; and to establish, based on the change information of the air flow data, a temperature field model between the change information of the air flow data and the power information.

[0020] The machine learning and deep learning module is used to adopt the trained load prediction model to predict load changes based on the time model, the temperature field model and the power load characteristic model of each cabinet to obtain the prediction result.

[0021] In some possible implementations, the temperature acquisition module is also used to collect the air outlet temperatures distributed in various cabinets or machine positions, send them to the air-conditioning unit communication module in real time, and perform data analysis on the air outlet temperatures to obtain the air outlet data.

[0022] In some possible implementations, the machine learning and deep learning module is also used to construct the trained load prediction model based on the temperature field model, the time model, the power load characteristic model of each cabinet, the operating parameters of the air-conditioning unit in the air-conditioning processing module, and the operating parameters of the air conditioner.

[0023] In a second aspect, a computing method for predicting and controlling heat load in a data center is provided, the method comprising:

[0024] Collect power information and air output data of the air conditioner cabinet;

[0025] Filtering the power information and the air output data to obtain change information of the power information and the air output data;

[0026] Based on the trained load prediction model and the power information and the change information of the wind output data, load change prediction is performed to obtain a prediction result;

[0027] The operating parameters of the terminal device of the air conditioner are adjusted based on the prediction result.

[0028] In conjunction with the second aspect above, in some possible implementations, the method further includes:

[0029] Based on the change information of the power information, establishing a time model between the change information of the power information and the wind speed data;

[0030] Based on the change information of the air output data, a temperature field model is established between the change information of the air output data and the power information;

[0031] Constructing a power load characteristic model for each cabinet in the entire row of cabinets based on the operating status and operating time of the entire row of cabinets;

[0032] The trained load prediction model is constructed based on the temperature field model, the time model, the power load characteristic model of each cabinet, the operating parameters of the air conditioner and the operating parameters of the corresponding air conditioning unit.

[0033] According to a third aspect, a computer program product is provided. The computer program product includes a computer program code. When the computer program code is run on a computer, the computer is caused to execute the above method.

[0034] In a fourth aspect, a computer-readable storage medium is provided, which stores a computer program code. When the computer program code is run on a computer, the computer executes the above method.

[0035] The present invention has the following beneficial effects: a computing device for predicting and controlling the heat load in a data center comprises an air conditioning processing module, an air conditioning unit communication module, a power and electricity collection module, and a temperature collection module; wherein the power and electricity collection module collects power information of the air conditioning cabinet and transmits it to the air conditioning unit communication module; the temperature collection module collects air outlet data dispersed in each cabinet or position and transmits it to the air conditioning unit communication module, so that the air conditioning unit communication module transmits the power information and air outlet data to the air conditioning processing module, allowing the air conditioning processing model to analyze the current information and air outlet data. Finally, the trained load prediction model embedded in the air conditioning processing module predicts load changes for the power information and the air outlet data, and adjusts the operating parameters of the terminal equipment of the air conditioner based on the prediction results. In this way, the data center's power consumption information and air output data are collected and analyzed, and the heat load is accurately analyzed through the trained load prediction model of machine learning and deep learning. This enables an accurate operation model to be established for the data center's refrigeration and air-conditioning system, achieving pre-optimal matching of computing load and air-conditioning energy, avoiding the high load risk caused by response lag, achieving efficient transmission and distribution of cooling capacity and heat exchange, and saving system energy. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] In order to more clearly illustrate the technical solutions and advantages of the embodiments of the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the prior art descriptions. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0037] Figure 1 This is a schematic diagram of the implementation framework of a computing device for predicting and controlling heat load in a data center provided by the present invention;

[0038] Figure 2 This is a schematic diagram of the composition and structure of a computing device for predicting and controlling heat load in a data center provided by an embodiment of the present invention;

[0039] Figure 3 This is a schematic diagram of the implementation flow of a computing method for predicting and controlling heat load in a data center provided by an embodiment of the present invention;

[0040] Figure 4 This is another implementation flow diagram of a computing method for predicting and controlling heat load in a data center provided by an embodiment of the present invention;

[0041] Figure 5 It is a structural diagram of a computer device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0042] To further illustrate the technical means and effects employed by the present invention to achieve the intended purpose of the invention, the following, in conjunction with the accompanying drawings and preferred embodiments, describes in detail the specific implementation, structure, features, and effects of a computing device for predicting and controlling the heat load of a data center proposed by the present invention. In the following description, different references to "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics of one or more embodiments may be combined in any suitable manner.

[0043] In the description of the embodiments of the present invention, unless otherwise specified, " / " means or, for example, A / B can mean A or B: "and / or" in the text is only a description of the association relationship of associated objects, indicating that there can be three relationships, for example, A and / or B can mean: A exists alone, A and B exist at the same time, and B exists alone. In addition, in the description of the embodiments of the present invention, "multiple" refers to two or more than two.

[0044] In the following, the terms "first" and "second" are used for descriptive purposes only and should not be understood to imply or suggest relative importance or implicitly indicate the number of technical features indicated. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features.

[0045] Unless otherwise defined, all technical and scientific terms used in the embodiments of the present invention have the same meanings as commonly understood by one of ordinary skill in the art to which the present invention belongs.

[0046] In data centers, most of the energy consumed during computing is dissipated into the environment as heat, necessitating the development of efficient cooling systems to dissipate heat from information technology (IT) equipment. Furthermore, to ensure the stable operation of IT equipment, data centers maintain high ambient temperature and humidity requirements, often relying on air conditioners, chillers, and other equipment for cooling. This is the primary cause of high energy consumption in data center cooling systems. Data center energy-saving retrofits typically focus on improving cooling system efficiency and reducing cooling losses. Improving cooling system energy efficiency is crucial for data center energy conservation.

[0047] Due to differences in installed capacity, installed capacity, and operating status between cabinets, data servers generate significant heat. The heat load between cabinets can also vary significantly depending on the amount of computing power. To ensure stable operation of IT equipment, data center terminals typically utilize lower supply air temperatures and higher air volumes to maintain overall heat dissipation. Lower supply air temperatures reduce the energy efficiency of the cooling system and negatively impact relative humidity control. Higher air volumes also increase fan energy consumption in terminal air conditioners such as automatic handheld units (AHUs) or precision air conditioners.

[0048] As computing services change, the power load also changes. Due to the thermal inertia of the air conditioning system, changes in cabinet outlet air temperature lag behind changes in power load, and changes in return air temperature from AHUs or precision air conditioners lag behind cabinet outlet air temperature. This delayed response of the air conditioning system to computing load can lead to risks such as overheating or thermal shutdown in some heavily loaded cabinets or computing servers. To avoid this risk, in high-density data centers, supply air temperatures are often set too low and air volumes are too high, reducing the energy efficiency of the data center cooling system.

[0049] Based on this, the embodiment of the present invention provides a computing device for predicting and controlling the heat load of a data center. By establishing a cabinet power consumption and heat load model, it can accurately analyze the relationship between changes in power consumption and changes in the operation of the air-conditioning system, and optimize the operating parameters of the AHU or precision air-conditioning accordingly, and realize predictive pre-optimization adjustment of the air-conditioning system, which is of great significance to the energy-saving operation of the central refrigeration and air-conditioning in the data center. Figure 1 As shown, the row cabinet power collection module 11 sends the instantaneous power data set to the AI computing device 12 for heat load prediction and control of the data room; the cabinet outlet air temperature collection module 13 sends the outlet air temperature data set to the AI computing device 12 for heat load prediction and control of the data room; the AI computing device 12 for heat load prediction and control of the data room performs heat load prediction on the instantaneous power data set and the outlet air temperature data set through machine learning and deep learning, thereby adjusting the AHU supply air temperature and AHU fan frequency of the air conditioner according to the prediction results.

[0050] The following is a detailed description of a specific scheme of a computing device for predicting and controlling the heat load of a data center provided by the present invention in conjunction with the accompanying drawings. Figure 2 , which shows a schematic diagram of the implementation process of a computing device for predicting and controlling the heat load of a data center provided by an embodiment of the present invention. The device 200 includes: an air conditioning processing module 201, an air conditioning unit communication module 202, a power collection module 203, and a temperature collection module 204; wherein:

[0051] The power collection module 203 is used to collect power information of the air-conditioning cabinet and transmit it to the air-conditioning unit communication module;

[0052] Here, the power and electricity collection module can be a communication interface of a power and electricity collection device, which collects power information from the air conditioner cabinet through this interface and transmits it to the air conditioner unit communication module, which then transmits it to the air conditioner processing module. Power information includes voltage, current, instantaneous power, etc.

[0053] In some possible implementations, the power and electricity collection module is provided in the head cabinet, and is also used to collect power information of the entire row of cabinets in the head cabinet; and collect the operating status of the entire row of cabinets, and construct a power load characteristic model of each cabinet in the entire row of cabinets based on the operating status and the corresponding operating time. Among them, the operating status of the entire row of cabinets includes: the operating speed, air outlet temperature and frequency of the entire row of cabinets. By establishing a mapping relationship between the operating status of each cabinet and the corresponding operating time, the power load characteristic model of each cabinet is obtained. In this way, the operating status of the corresponding cabinet at each moment can be characterized by the power load characteristic model. After the power and electricity collection module constructs the power load characteristic model of each cabinet, the air conditioning processing module uses the trained load prediction model to predict load changes based on the power information, the air outlet data and the power load characteristic model of each cabinet, so that more accurate prediction results can be obtained.

[0054] The temperature collection module 204 is used to collect air outlet data distributed in various cabinets or positions and transmit the data to the air conditioning unit communication module.

[0055] Here, the temperature acquisition module can be implemented through the communication interface of the temperature acquisition device, and the air outlet data of each cabinet or unit position is collected through the communication interface of the temperature acquisition device to be transmitted to the temperature unit communication module.

[0056] In some possible implementations, the temperature acquisition module is further configured to collect the outlet temperatures of each cabinet or unit, transmit these temperatures in real time to the air conditioning unit communication module, and perform data analysis on the outlet temperatures to obtain the outlet air data. Specifically, the outlet temperatures of each cabinet or unit are collected via the communication interface of the temperature acquisition device. The outlet air temperatures of each area or unit are analyzed to obtain the outlet air data, thereby providing a basis for establishing a thermal model for the data cabinet.

[0057] The air conditioning unit communication module 202 is used to transmit the power information and the air output data to the air conditioning processing module;

[0058] The air conditioning unit communication module can be implemented through the precision air conditioning unit communication interface, which transmits power consumption information and air flow data to the air conditioning processing module. This precision air conditioning unit communication interface can read the supply air temperature and fan frequency at the air conditioning terminal, enabling data collection and analysis of central air conditioning terminal equipment. The supply air temperature can be adjusted through the communication interface, and the air volume and pressure can be changed by adjusting the fan frequency.

[0059] The air conditioning processing module 201 is embedded with a trained load prediction model, which is used to predict load changes on the power information and the air output data based on the trained load prediction model, and adjust the operating parameters of the terminal equipment of the air conditioner based on the prediction results.

[0060] In some possible implementations, the air conditioning processing module 201 includes: a protocol analysis module, an automatic data screening module, and a machine learning and deep learning module; wherein:

[0061] The protocol analysis module has a data scanning function, and is used to determine the communication protocol of the received power information and the air output data.

[0062] Here, the protocol analysis module has a data scanning function, which can automatically analyze the communication protocol of the access data and automatically correspond the data table with the analog quantity relationship.

[0063] The data automatic screening module is used to determine the processing results, the power information and the change information of the air outlet data; analyze the power information and the change information of the air outlet data, and determine the positional relationship between the monitoring power points corresponding to the air outlet data and the power information.

[0064] Here, the data automatic screening module has the function of logical analysis and judgment, analyzes the changes in power and air output data, and automatically corresponds the positional relationship and logical relationship between the temperature data and the corresponding monitoring power points.

[0065] In some possible implementations, the automatic data screening module is further configured to determine a position change relationship between the air flow data and the power monitoring point corresponding to the power information; wherein the change information includes: the position change information. The automatic data screening module is further configured to establish a time model between the power information change information and the air flow data based on the power information change information; and to establish a temperature field model between the air flow data change information and the power information based on the air flow data change information.

[0066] Here, the data automatic screening module analyzes the changes in power and air outlet data, and automatically matches the relative changes in the position between the temperature data and the corresponding monitoring power points, that is, obtains the position change relationship. By analyzing the power data and air outlet temperature data, the data automatic screening module can establish a time model between the change information of the power information and the air outlet data, and establish an accurate temperature field model of power (i.e., power data) and temperature changes (i.e., change information of the air outlet data). The time model and temperature field model are passed to the machine learning and deep learning modules.

[0067] The machine learning and deep learning module is used to adopt the trained load forecasting model to perform load change forecasting based on the change information to obtain the prediction result.

[0068] Here, the machine learning and deep learning modules incorporate AI algorithms for optimization, predicting load fluctuations and proactively adjusting operating parameters such as the AHU (Actual Handling Unit) and precision air conditioner supply air temperature and air volume, thereby achieving system energy savings. This model algorithm optimizes the supply air temperature and fan frequency of the AHU or precision air conditioner.

[0069] The machine learning and deep learning module is used to adopt the trained load prediction model to predict load changes based on the time model, the temperature field model and the power load characteristic model of each cabinet to obtain the prediction result.

[0070] Here, the machine learning and deep learning module trains a neural network (e.g., a convolutional neural network or a residual neural network) using the temperature field model, the time model, the power load characteristic model of each cabinet, the operating parameters of the air conditioning units in the air conditioning processing module, and the operating parameters of the air conditioners to obtain a trained load prediction model. The neural network is trained using the temperature field model, the time model, the power load characteristic model of each cabinet, the operating parameters of the air conditioning units in the air conditioning processing module, and the operating parameters of the air conditioners as training data to obtain a trained load prediction model.

[0071] After the trained load prediction model is obtained, the time model, the temperature field model and the power load characteristic model of each cabinet are input into the model to perform load change prediction, thereby obtaining a prediction result.

[0072] In some possible implementations, the air conditioning processing module also includes: an air conditioning unit, which is used to read the supply air temperature and fan frequency of the terminal device of the air conditioner, and call the air conditioning unit communication module to adjust the supply air temperature, and change the supply air volume and supply air pressure of the air conditioner based on the fan frequency.

[0073] Here, the AHU reads the air supply temperature and fan frequency at the air conditioning terminal through the precision air conditioning unit communication interface, enabling data collection and analysis of the central air conditioning terminal equipment. The AHU can also adjust the air supply temperature through the communication interface, and change the air supply volume and pressure by adjusting the fan frequency.

[0074] The embodiments of the present invention collect and analyze information such as the power supply of the data center, various parameters of air-conditioning operation, and equipment operating status. Through machine learning and deep learning, the heat load is accurately analyzed. An accurate operation model can be established for the data center refrigeration and air-conditioning system, achieving pre-optimal matching of computing load and air-conditioning energy, avoiding the high load risk caused by response lag, realizing efficient transmission and distribution of cooling capacity and heat exchange, and realizing system energy saving.

[0075] Furthermore, this embodiment of the present invention provides more data support for load distribution and space optimization of data center refrigeration and air conditioning systems. Through AI-based system analysis, it can better optimize cabinet layout, air conditioner selection, and improve the water balance of the air conditioning chilled water system, providing a basis for the efficient and stable operation of the refrigeration units. This embodiment of the present invention also provides data support for the efficient use of energy-saving technologies such as evaporative cooling and free cooling, further reducing data center energy consumption and improving the PUE of data centers.

[0076] The embodiment of the present invention provides a computing method for predicting and controlling the heat load of a data center. Figure 3 , which shows a schematic diagram of the implementation flow of a computing method for predicting and controlling the heat load of a data center provided by one embodiment of the present invention, the method comprising:

[0077] 301, collecting power information and air output data of the air conditioner cabinet.

[0078] Here, the power information and air output data of the air conditioner cabinet are collected through the power and electricity collection module.

[0079] 302 : Filter the power information and the air output data to obtain change information of the power information and the air output data.

[0080] Here, the power information and the air output data are screened and processed by the data automatic screening module to obtain change information of the power information and the air output data.

[0081] 303 : Based on the trained load prediction model and the power information and the change information of the wind output data, perform load change prediction to obtain a prediction result.

[0082] Here, a trained load prediction model is first constructed through machine learning and deep learning modules, and then the real-time power information and the change information of the wind output data are input into the model to predict the load change, thereby obtaining the prediction result.

[0083] In some possible implementations, the trained load forecasting model can be Figure 4 Follow the steps shown to build:

[0084] 401 : Based on the change information of the power information, establish a time model between the change information of the power information and the air output data.

[0085] 402 : Based on the change information of the air output data, establish a temperature field model between the change information of the air output data and the power information.

[0086] 403 : Based on the operating status and operating time of the entire row of cabinets, construct a power load characteristic model of each cabinet in the entire row of cabinets.

[0087] 404 , construct the trained load prediction model based on the temperature field model, the time model, the power load characteristic model of each cabinet, the operating parameters of the air conditioner, and the operating parameters of the corresponding air conditioning unit.

[0088] 304 : Adjust operating parameters of the terminal device of the air conditioner based on the prediction result.

[0089] Here, after obtaining the prediction results through the model algorithm, the air supply temperature and fan frequency of the air conditioning unit AHU or precision air conditioner can be optimized.

[0090] In an embodiment of the present invention, a power and electricity collection module collects power information from the air conditioner cabinets and transmits it to the air conditioner unit communication module. A temperature collection module collects air flow data distributed across each cabinet or unit and transmits it to the air conditioner unit communication module. This allows the air conditioner unit communication module to transmit the power and air flow data to the air conditioner processing module, enabling the air conditioner processing model to analyze the current information and air flow data. Finally, a trained load prediction model embedded in the air conditioner processing module predicts load changes based on the power and air flow data, and adjusts the operating parameters of the air conditioner's terminal equipment based on the prediction results. In this way, by collecting and analyzing information such as power and air flow data from the data center, and accurately analyzing the heat load using a trained load prediction model based on machine learning and deep learning, an accurate operational model can be established for the data center's refrigeration and air conditioning system, achieving optimal matching of computational load and air conditioning energy, avoiding the risk of high load caused by response lag, achieving efficient cooling and heat exchange, and saving system energy.

[0091] Optionally, the transmission medium can be a wired link (for example, but not limited to, coaxial cable, optical fiber and digital subscriber line (DSL)) or a wireless link (for example, but not limited to, wireless Fidelity (WIFI), Bluetooth and mobile device network). It should be noted that the system provided in the above embodiment is only illustrated by the division of the above functional modules. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the computer device can be divided into different functional modules to complete all or part of the functions described above. In addition, the method embodiments provided in the above embodiments belong to the same concept. The specific implementation process is detailed in the method embodiments, which will not be repeated here.

[0092] Figure 5 FIG. 1 is a schematic diagram of the structure of a computer device provided by an embodiment of the present invention. For example, Figure 5 As shown, the computer device 500 includes: a memory 501, a processor 502, and a computer program 503 stored in the memory 501 and running on the processor 502, wherein when the processor 502 executes the computer program 503, the computer device can execute any of the computing devices for predicting and controlling the heat load of a data center described above.

[0093] In addition, an embodiment of the present invention also protects a system, which may include a memory and a processor, wherein the memory stores an executable program code, and the processor is used to call and execute the executable program code to execute a computing device for predicting and controlling the heat load of a data room provided by an embodiment of the present invention. This embodiment can divide the system into functional modules according to the above-mentioned method example. For example, it can correspond to each functional module, or two or more functions can be integrated into one processing module. The above-mentioned integrated module can be implemented in the form of hardware. It should be noted that the division of modules in this embodiment is schematic, which is only a logical function division. There may be other division methods in actual implementation. It should be noted that all relevant contents of each step involved in the above-mentioned method embodiment can be referred to the functional description of the corresponding functional module, and will not be repeated here.

[0094] It should be understood that the system provided in this embodiment is used to execute the above-mentioned computing device for predicting and controlling the heat load of a data room, and therefore can achieve the same effect as the above-mentioned implementation method. In the case of an integrated unit, the system may include a processing module and a storage module. Specifically, when the system is applied to a device, the processing module can be used to control and manage the actions of the device. The storage module can be used to support the device in executing mutual program codes, etc. Specifically, the processing module can be a processor or a controller, which can implement or execute the various exemplary logic blocks, modules and circuits described in conjunction with the disclosure of the present invention. The processor can also be a combination that implements computing functions, such as a combination of one or more microprocessors, a combination of digital signal processing (DSP) and a microprocessor, etc., and the storage module can be a memory.

[0095] In addition, the system provided by the embodiments of the present invention may specifically be a chip, component, or module. The chip may include a connected processor and memory; the memory is used to store instructions. When the processor calls and executes the instructions, the chip can execute the computing device for predicting and controlling the heat load of a data center provided by the above embodiment. This embodiment also provides a computer-readable storage medium storing computer program code. When the computer program code is executed on a computer, the computer executes the above-mentioned method steps to implement the computing device for predicting and controlling the heat load of a data center provided by the above embodiment.

[0096] This embodiment also provides a computer program product. When the computer program product is executed on a computer, it causes the computer to execute the above-mentioned steps to implement a computing device for predicting and controlling the heat load of a data center provided in the above embodiment. The system, computer-readable storage medium, computer program product, or chip provided in this embodiment are all used to execute the corresponding method provided above. Therefore, the beneficial effects achieved by the system can refer to the beneficial effects of the corresponding method provided above and will not be repeated here. Through the description of the above embodiments, those skilled in the art will understand that for the sake of convenience and brevity, only the division of the above-mentioned functional modules is used as an example. In actual applications, the above-mentioned functions can be distributed among different functional modules as needed, that is, the internal structure of the system can be divided into different functional modules to complete all or part of the functions described above. In the embodiments provided by the present invention, it should be understood that the disclosed systems and methods can be implemented in other ways. For example, the system embodiments described above are merely illustrative. For example, the division of modules or units is only a logical functional division. In actual implementation, other division methods can be used. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not implemented. On the other hand, the mutual coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection through some interface, system or unit, which may be electrical, mechanical or other forms.

[0097] It should be noted that the above-mentioned order of the embodiments of the present invention is for description only and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multi-task processing and parallel processing are also possible or may be advantageous. The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other. Each embodiment focuses on the differences from other embodiments. The above content is only a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any technician familiar with the technical field can easily think of changes or replacements within the technical scope disclosed by the present invention, which should be covered within the scope of protection of the present invention.

Claims

1. A computing device for predicting and controlling heat load in a data center, characterized in that: The device includes: an air conditioning processing module, an air conditioning unit communication module, an electric power collection module, and a temperature collection module; wherein: The power and electricity collection module is used to collect the power information of the air-conditioning cabinet and transmit it to the air-conditioning unit communication module; The temperature acquisition module is used to collect air outlet data distributed in each cabinet or unit position and transmit it to the air conditioning unit communication module; The air conditioning unit communication module is used to transmit the power information and the air output data to the air conditioning processing module; The air conditioning processing module is embedded with a trained load prediction model, which is used to predict load changes on the power information and the air outlet data based on the trained load prediction model, and adjust the operating parameters of the terminal equipment of the air conditioner based on the prediction results.

2. A computing device for predicting and controlling heat load in a data center according to claim 1, characterized in that: The power and electricity collection module is provided in the row head cabinet and is further used to collect the power information of the entire row of cabinets in the row head cabinet; collect the operating status of the entire row of cabinets, and build a power load characteristic model of each cabinet in the entire row of cabinets based on the operating status and corresponding operating time; The air conditioning processing module is further configured to use the trained load prediction model to perform load change prediction based on the power information, the air outlet data, and the power load characteristic model of each cabinet to obtain the prediction result.

3. The computing device for predicting and controlling heat load in a data center according to claim 1, characterized in that: The air conditioning processing module includes: a protocol analysis module, an automatic data screening module, a machine learning and deep learning module; wherein: The protocol analysis module has a data scanning function for determining the communication protocol of the received power information and the air output data; The automatic data screening module is configured to determine the processing result, the power information, and the change information of the air output data; based on the power information and the change information of the air output data, determine the positional relationship between the power monitoring points corresponding to the air output data and the power information; The machine learning and deep learning module is used to adopt the trained load forecasting model to perform load change forecasting based on the change information to obtain the prediction result.

4. A computing device for predicting and controlling heat load in a data center according to claim 3, characterized in that: The air conditioning processing module further includes: The air conditioning unit is used to read the air supply temperature and fan frequency of the terminal device of the air conditioner, and call the air conditioning unit communication module to adjust the air supply temperature, and change the air supply volume and air supply pressure of the air conditioner based on the fan frequency.

5. The computing device for predicting and controlling heat load in a data center according to claim 3, characterized in that: The automatic data screening module is further used to determine the position change information between the air outlet data and the power monitoring point corresponding to the power information; wherein the change information includes: the position change information.

6. The computing device for predicting and controlling heat load in a data center according to claim 3, characterized in that: The automatic data screening module is further configured to establish a time model between the change information of the power information and the air flow data based on the change information of the power information; and to establish a temperature field model between the change information of the air flow data and the power information based on the change information of the air flow data; The machine learning and deep learning module is used to adopt the trained load prediction model to predict load changes based on the time model, the temperature field model and the power load characteristic model of each cabinet to obtain the prediction result.

7. The computing device for predicting and controlling heat load in a data center according to claim 1, characterized in that: The temperature acquisition module is also used to collect the air outlet temperatures distributed in various cabinets or machine positions, send them to the air conditioning unit communication module in real time, and perform data analysis on the air outlet temperatures to obtain the air outlet data.

8. The computing device for predicting and controlling heat load in a data center according to claim 3, characterized in that: The machine learning and deep learning module is also used to construct the trained load prediction model based on the temperature field model, the time model, the power load characteristic model of each cabinet, the operating parameters of the air-conditioning unit in the air-conditioning processing module and the operating parameters of the air conditioner.

9. A computing method for predicting and controlling heat load in a data center, characterized in that: The method comprises: Collect power information and air output data of the air conditioner cabinet; Filtering the power information and the air output data to obtain change information of the power information and the air output data; Based on the trained load prediction model and the power information and the change information of the wind output data, load change prediction is performed to obtain a prediction result; The operating parameters of the terminal device of the air conditioner are adjusted based on the prediction result.

10. A computing method for predicting and controlling heat load in a data center according to claim 9, characterized in that: The method further comprises: Based on the change information of the power information, establishing a time model between the change information of the power information and the wind speed data; Based on the change information of the air output data, a temperature field model is established between the change information of the air output data and the power information; Constructing a power load characteristic model for each cabinet in the entire row of cabinets based on the operating status and operating time of the entire row of cabinets; The trained load prediction model is constructed based on the temperature field model, the time model, the power load characteristic model of each cabinet, the operating parameters of the air conditioner and the operating parameters of the corresponding air conditioning unit.

Images