CN114065602A - Temperature prediction method, model training method and related equipment - Google Patents

Temperature prediction method, model training method and related equipment Download PDF

Info

Publication number
CN114065602A
CN114065602A CN202010795254.2A CN202010795254A CN114065602A CN 114065602 A CN114065602 A CN 114065602A CN 202010795254 A CN202010795254 A CN 202010795254A CN 114065602 A CN114065602 A CN 114065602A
Authority
CN
China
Prior art keywords
data
temperature
air
air inlet
cold pool
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202010795254.2A
Other languages
Chinese (zh)
Inventor
袁祥枫
曾宇
杜民
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
China Telecom Corp Ltd
Original Assignee
China Telecom Corp Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by China Telecom Corp Ltd filed Critical China Telecom Corp Ltd
Priority to CN202010795254.2A priority Critical patent/CN114065602A/en
Publication of CN114065602A publication Critical patent/CN114065602A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00—Computer-aided design [CAD]
    • G06F30/20—Design optimisation, verification or simulation
    • G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01D—MEASURING NOT SPECIALLY ADAPTED FOR A SPECIFIC VARIABLE; ARRANGEMENTS FOR MEASURING TWO OR MORE VARIABLES NOT COVERED IN A SINGLE OTHER SUBCLASS; TARIFF METERING APPARATUS; MEASURING OR TESTING NOT OTHERWISE PROVIDED FOR
    • G01D21/00—Measuring or testing not otherwise provided for
    • G01D21/02—Measuring two or more variables by means not covered by a single other subclass
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
    • G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N20/00—Machine learning
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
    • G06F2119/08—Thermal analysis or thermal optimisation

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Evolutionary Computation (AREA)
  • Data Mining & Analysis (AREA)
  • Software Systems (AREA)
  • Artificial Intelligence (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • General Engineering & Computer Science (AREA)
  • Medical Informatics (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Evolutionary Biology (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Computing Systems (AREA)
  • Mathematical Physics (AREA)
  • Computer Hardware Design (AREA)
  • Geometry (AREA)
  • Air Conditioning Control Device (AREA)

Abstract

The disclosure provides a temperature prediction method and device, a model training method and device, electronic equipment and a computer readable storage medium, and relates to the technical field of artificial intelligence. The temperature prediction method comprises the following steps: acquiring air conditioner air outlet data, server load data and cold pool air inlet air volume data; inputting air conditioner air outlet data into a pre-trained first machine learning model to obtain temperature and humidity prediction data of an air inlet of a cold pool; and inputting the temperature and humidity prediction data of the air inlet of the cold pool, the server load data and the air quantity data of the air inlet of the cold pool into a pre-trained second machine learning model to obtain the temperature prediction data of the air outlet of the cabinet. This prediction rack air outlet temperature that this disclosure can be accurate to real-time and convenient efficient acquire rack air outlet temperature.

Description

Temperature prediction method, model training method and related equipment
Technical Field
The present disclosure relates to the field of artificial intelligence technologies, and in particular, to a temperature prediction method and apparatus, a model training method and apparatus, an electronic device, and a computer-readable storage medium.
Background
In the machine room, there are usually racks with different heights, and the racks may have overhigh temperature during operation.
In order to detect whether the cabinet has an over-high temperature phenomenon in the working process, the temperature of the air outlet of the cabinet needs to be acquired. The traditional method for obtaining the temperature of the air outlet of the cabinet generally needs to deploy a temperature sensor or a patrol robot in a machine room.
Disclosure of Invention
The technical problem that this disclosure solved is, how to acquire rack air outlet temperature in real time and convenient efficient.
According to a first aspect of embodiments of the present disclosure, there is provided a temperature prediction method, including: acquiring air conditioner air outlet data, server load data and cold pool air inlet air volume data; inputting air conditioner air outlet data into a pre-trained first machine learning model to obtain temperature and humidity prediction data of an air inlet of a cold pool; and inputting the temperature and humidity prediction data of the air inlet of the cold pool, the server load data and the air quantity data of the air inlet of the cold pool into a pre-trained second machine learning model to obtain the temperature prediction data of the air outlet of the cabinet.
In some embodiments, the temperature prediction method further comprises: determining target air volume data of the air inlet of the cold pool according to the temperature and humidity prediction data of the air inlet of the cold pool, server load data, cabinet air outlet temperature target data and a pre-trained second machine learning model; and adjusting the air volume data of the air inlet of the cold pool into target air volume data of the air inlet of the cold pool so as to adjust the temperature prediction data of the air outlet of the cabinet into target temperature data of the air outlet of the cabinet.
In some embodiments, determining the target air volume data of the cold pool air inlet according to the temperature and humidity prediction data of the cold pool air inlet, the server load data, the cabinet air outlet temperature target data, and a pre-trained second machine learning model includes: traversing the air quantity test data of the air inlet of the cold pool by a preset step length; inputting the temperature and humidity prediction data of the cold pool air inlet, the server load data and the cold pool air inlet air quantity test data into a pre-trained second machine learning model to obtain cabinet air outlet temperature test data; and when the cabinet air outlet temperature test data is equal to the cabinet air outlet temperature target data, taking the cold pool air inlet test air volume data as cold pool air inlet target air volume data.
According to a second aspect of the embodiments of the present disclosure, there is provided a model training method, including: collecting air conditioner air outlet sample data and cold pool air inlet temperature and humidity sample data; training a first machine learning model by using air conditioner air outlet sample data and cold pool air inlet temperature and humidity sample data, so that the first machine learning model can obtain cold pool air inlet temperature and humidity prediction data according to the air conditioner air outlet data; collecting temperature and humidity sample data of an air inlet of a cold pool, server load sample data, air quantity sample data of the air inlet of the cold pool and temperature sample data of an air outlet of a cabinet; and training a second machine learning model by using the temperature and humidity sample data of the cold pool air inlet, the server load sample data, the air volume sample data of the cold pool air inlet and the cabinet air outlet temperature sample data, so that the second machine learning model can obtain cabinet air outlet temperature prediction data according to the temperature and humidity prediction data of the cold pool air inlet, the server load data and the air volume data of the cold pool air inlet.
According to a third aspect of the embodiments of the present disclosure, there is provided a temperature prediction apparatus including: a data acquisition module configured to: acquiring air conditioner air outlet data, server load data and cold pool air inlet air volume data; a first data prediction module configured to: inputting air conditioner air outlet data into a pre-trained first machine learning model to obtain temperature and humidity prediction data of an air inlet of a cold pool; and the second data prediction module is configured to input the temperature and humidity prediction data of the cold pool air inlet, the server load data and the air volume data of the cold pool air inlet into a pre-trained second machine learning model to obtain the temperature prediction data of the cabinet air outlet.
In some embodiments, the temperature prediction device further comprises: a data determination module configured to: determining target air volume data of the air inlet of the cold pool according to the temperature and humidity prediction data of the air inlet of the cold pool, server load data, cabinet air outlet temperature target data and a pre-trained second machine learning model; a data conditioning module configured to: and adjusting the air volume data of the air inlet of the cold pool into target air volume data of the air inlet of the cold pool so as to adjust the temperature prediction data of the air outlet of the cabinet into target temperature data of the air outlet of the cabinet.
In some embodiments, the data determination module is configured to: traversing the air quantity test data of the air inlet of the cold pool by a preset step length; inputting the temperature and humidity prediction data of the cold pool air inlet, the server load data and the cold pool air inlet air quantity test data into a pre-trained second machine learning model to obtain cabinet air outlet temperature test data; and when the cabinet air outlet temperature test data is equal to the cabinet air outlet temperature target data, taking the cold pool air inlet test air volume data as cold pool air inlet target air volume data.
According to a fourth aspect of the embodiments of the present disclosure, there is provided a model training apparatus including: a first model training module configured to: collecting air conditioner air outlet sample data and cold pool air inlet temperature and humidity sample data; training a first machine learning model by using air conditioner air outlet sample data and cold pool air inlet temperature and humidity sample data, so that the first machine learning model can obtain cold pool air inlet temperature and humidity prediction data according to the air conditioner air outlet data; a second model training module configured to: collecting temperature and humidity sample data of an air inlet of a cold pool, server load sample data, air quantity sample data of the air inlet of the cold pool and temperature sample data of an air outlet of a cabinet; and training a second machine learning model by using the temperature and humidity sample data of the cold pool air inlet, the server load sample data, the air volume sample data of the cold pool air inlet and the cabinet air outlet temperature sample data, so that the second machine learning model can obtain cabinet air outlet temperature prediction data according to the temperature and humidity prediction data of the cold pool air inlet, the server load data and the air volume data of the cold pool air inlet.
According to a fifth aspect of the embodiments of the present disclosure, there is provided an electronic apparatus including: a memory; and a processor coupled to the memory, the processor configured to perform the aforementioned temperature prediction method or the aforementioned model training method based on instructions stored in the memory.
According to a sixth aspect of embodiments of the present disclosure, there is provided a computer-readable storage medium, wherein the computer-readable storage medium stores computer instructions, which when executed by a processor, implement the aforementioned temperature prediction method or the aforementioned model training method.
This prediction rack air outlet temperature that this disclosure can be accurate to real-time and convenient efficient acquire rack air outlet temperature.
Other features of the present disclosure and advantages thereof will become apparent from the following detailed description of exemplary embodiments thereof, which proceeds with reference to the accompanying drawings.
Drawings
In order to more clearly illustrate the embodiments of the present disclosure or the technical solutions in the prior art, the drawings needed to be used in the description of the embodiments or the prior art will be briefly introduced below, it is obvious that the drawings in the following description are only some embodiments of the present disclosure, and for those skilled in the art, other drawings can be obtained according to the drawings without inventive exercise.
Figure 1 shows a schematic diagram of the refrigeration of a machine room.
FIG. 2 illustrates a flow diagram of a model training method of some embodiments of the present disclosure.
FIG. 3 illustrates a flow diagram of a temperature prediction method of some embodiments of the present disclosure.
FIG. 4 shows a schematic flow chart of a temperature prediction method according to further embodiments of the present disclosure.
Fig. 5 shows a schematic structural diagram of a temperature prediction device according to some embodiments of the present disclosure.
FIG. 6 illustrates a schematic structural diagram of a model training apparatus according to some embodiments of the present disclosure.
Fig. 7 shows a schematic structural diagram of an electronic device of some embodiments of the present disclosure.
Detailed Description
The technical solutions in the embodiments of the present disclosure will be clearly and completely described below with reference to the drawings in the embodiments of the present disclosure, and it is obvious that the described embodiments are only a part of the embodiments of the present disclosure, and not all of the embodiments. The following description of at least one exemplary embodiment is merely illustrative in nature and is in no way intended to limit the disclosure, its application, or uses. All other embodiments, which can be derived by a person skilled in the art from the embodiments disclosed herein without making any creative effort, shall fall within the protection scope of the present disclosure.
After research, the inventor finds that the traditional method for obtaining the temperature of the air outlet of the cabinet needs to modify a machine room. Because the air inlets and the air outlets of the machine cabinet with different heights have different temperatures, sensors are required to be installed at different heights of the machine cabinet, so that the cost is high and the operation and maintenance are difficult. Meanwhile, the large number of temperature measuring points can cause the inspection robot to inspect the machine room for a week for too long time, so that the real-time monitoring of the temperature of the air outlet of the cabinet is difficult to realize, and the phenomenon that whether the temperature of the cabinet is too high or not in the working process can not be found in time.
Figure 1 shows a schematic diagram of the refrigeration of a machine room. The inventor analyzes that the local over-high temperature in the machine room is caused by unreasonable air flow organization in the machine room and high-load operation of servers in the cabinet. Because the machine room environment generally does not change violently, the temperature and humidity of air conditioner air-out generally has a stable function mapping relation with the temperature and humidity of air inlet of each cold pool air inlet, and simultaneously, the temperature of each cabinet air outlet has a more stable function mapping relation with the temperature and humidity of cold pool air inlet, air output and server load. These functional mappings may be learned and expressed by artificial intelligence algorithms.
Some embodiments of the disclosed model training method are first described in conjunction with fig. 2.
FIG. 2 illustrates a flow diagram of a model training method of some embodiments of the present disclosure. As shown in fig. 2, the present embodiment includes steps S201 to S204.
In step S201, air conditioner air outlet sample data and cold pool air inlet temperature and humidity sample data are collected.
For example, air conditioner air outlet sample data can be collected from a moving loop system of a machine room, and the air conditioner air outlet sample data specifically comprises temperature data, humidity data and air outlet volume data; a sensor can be additionally arranged at the air inlet of the cold pool to acquire temperature data and humidity data of the air inlet of the cold pool. These data may be uploaded to an intelligent hotspot detection and elimination system.
In step S202, a first machine learning model is trained by using air conditioner air outlet sample data and cold pool air inlet temperature and humidity sample data.
The first machine learning model is trained to learn the following mapping function fpool。
(Tpool_in_m,Hpool_in_m)＝fpool(Tac1,Hac1,Fac1,Tac2,Hac2,Fac2,...,Tacn,Hacn,Facn)
Wherein, Tpool_in_mThe temperature of the air inlet of the cooling pond at the position of m, Hpool_in_mHumidity of air inlet of cooling pond at m position, TacnIs the outlet air temperature of the nth air conditioner, HacnIs the outlet air humidity of the nth air conditioner, FacnThe air outlet quantity of the nth air conditioner. The trained first machine learning model can obtain cold pool air inlet temperature and humidity prediction data according to air conditioner air outlet data.
In step S203, sample temperature and humidity data of the cold pool air inlet, sample data of the server load, sample data of the cold pool air inlet air volume, and sample data of the cabinet air outlet temperature are collected. These data may also be uploaded to an intelligent hotspot detection and elimination system.
For example, a sensor can be additionally arranged at the air inlet of the cold pool to acquire temperature and humidity sample data and air volume sample data of the air inlet of the cold pool; server load sample data can be collected from a computer room server management system, and the server load sample data can specifically comprise CPU utilization rate, disk input and output speed and the like; a sensor can be additionally arranged at the air outlet of the cabinet to acquire temperature sample data of the air outlet of the cabinet.
In step S204, a second machine learning model is trained by using the temperature and humidity sample data of the cold pool air inlet, the server load sample data, the air volume sample data of the cold pool air inlet, and the cabinet air outlet temperature sample data.
The second machine learning model is trained to learn the following mapping function fcabin。
Tcabin_m_h＝fcabin(Tpool_in_m,Hpool_in_m,Fpool_in_m,Rcpu_m1,Rio_m1,Rcpu_m2,Rio_m2...,Rcpu_mn,Rio_mn)
Wherein, Tcain_m_hTemperature, T, of the air outlet at h height of the cabinet at m positionpool_in_mThe inlet air temperature H of the air inlet of the m-position cold poolpool_in_mThe air inlet humidity of the air inlet of the cooling pond at the position of m, Fpool_in_mThe air inlet volume R of the air inlet of the cooling pond at the position of mcpu_mnThe CPU utilization rate of the server with the m-position cabinet at the n height is shown, and Rio _ n is the disk input/output read-write speed of the cabinet with the m position at the server with the n height. The trained second machine learning model can obtain cabinet air outlet temperature prediction data according to cold pool air inlet temperature and humidity prediction data, server load data and cold pool air inlet air volume data.
Those skilled in the art will understand that the first Machine learning model and the second Machine learning model may adopt Machine learning models capable of better learning nonlinear functions and performing logistic regression, and specifically may be distributed gradient enhancement library XGBoost, SVM (Support Vector Machine), neural network models, and the like.
Some embodiments of the disclosed temperature prediction method are described below in conjunction with fig. 3.
FIG. 3 illustrates a flow diagram of a temperature prediction method of some embodiments of the present disclosure. As shown in fig. 3, the present embodiment includes steps S301 to S303.
In step S301, air conditioner air outlet data, server load data, and cold pool air inlet volume data are obtained.
For example, air conditioner air outlet data, server load data and cold pool air inlet air volume data in real time can be updated to an intelligent hot spot detection and elimination system. And then, acquiring air conditioner air outlet data, server load data and cold pool air inlet air volume data in a real-time state from the intelligent hot spot detection and elimination system.
In step S302, air conditioner outlet air data is input into a pre-trained first machine learning model, and temperature and humidity prediction data of the cold pool air inlet is obtained.
In step S303, the temperature and humidity prediction data of the cold pool air inlet, the server load data, and the air volume data of the cold pool air inlet are input into a second machine learning model trained in advance, so as to obtain prediction data of the cabinet air outlet temperature.
The embodiment provides an intelligent temperature prediction method applied to a machine room, wherein a sensor is adopted to collect data and model in an early data collection stage, the sensor is not required to be adopted in a later temperature prediction stage, and a machine learning model is adopted to realize full-time and real-time prediction on the temperature of an air outlet of a cabinet, so that the temperature of the air outlet of the cabinet can be obtained conveniently and efficiently in real time, the cost is low, and the operation and maintenance are convenient.
The following are specifically mentioned: on one hand, in the embodiment, the temperature of the air outlet of the cabinet is not directly predicted through air conditioner air outlet data and server load data, but modeling on temperature and humidity data of the air inlet of the cold pool is added, so that heat loss in the middle process of airflow flow direction can be reflected; on the other hand, this embodiment has still considered the influence of the fluctuation of server load to rack air outlet temperature from the angle of generating heat to the influence of cold pond air intake humiture to rack air outlet temperature has been considered from the refrigeration angle. Therefore, the temperature of the air outlet of the cabinet can be accurately predicted by the embodiment.
The inventor further researches and discovers that after the traditional method for acquiring the temperature of the air outlet of the cabinet is completed, an accurate suggestion cannot be further provided to relieve the phenomenon that the temperature of the cabinet is too high in the working process. In fact, the machine room usually adopts a large-scale over-cooling method to avoid the phenomenon of over-high temperature of the cabinet in the working process, thereby causing huge energy loss. Therefore, the present disclosure also provides a corresponding embodiment to further solve the problem.
Further embodiments of the temperature prediction method of the present disclosure are described below in conjunction with fig. 4.
FIG. 4 shows a schematic flow chart of a temperature prediction method according to further embodiments of the present disclosure. As shown in fig. 4, the present embodiment includes steps S404 to S405.
In step S404, the target air volume data of the cold pool air inlet is determined according to the temperature and humidity prediction data of the cold pool air inlet, the server load data, the cabinet air outlet temperature target data, and the pre-trained second machine learning model.
Firstly, traversing the air quantity test data of the air inlet of the cold pool by a preset step length. For example, at a given Tpool_in_m、Hpool_in_m、Rcpu_m1、Rio_m1、Rcpu_m2、Rio_m2……Rcpu_mn、Rio_mnIn the case of these parameters, for Fpool_in_mAnd traversing. And then, inputting the temperature and humidity prediction data of the cold pool air inlet, the server load data and the traversed air quantity test data of the cold pool air inlet into a pre-trained second machine learning model in sequence to obtain the temperature test data of the cabinet air outlet. When the temperature test data of the cabinet air outlet is equal to the target data T of the temperature of the cabinet air outlettargetAnd taking the test air volume data of the air inlet of the cold pool as the target air volume data of the air inlet of the cold pool. That is, finding that the second machine learning model output is equal to TtargetTime corresponding input parameter Ftarget。
In step S405, the cold pool air inlet volume data is adjusted to the cold pool air inlet target volume data, so as to adjust the cabinet air outlet temperature prediction data to the cabinet air outlet temperature target data.
When the air quantity data of the air inlet of the cold pool is adjusted, the air quantity data of the air inlet of the cold pool is assumed to be FnowWill FtargetAnd the data is used as the target air quantity data of the air inlet of the cold pool. Then, F ═ Ftarget-FnowNamely the adjustment quantity of the air quantity data of the air inlet of the current cold pool.
According to the air conditioner cabinet air inlet air quantity data adjusting method and device, on the basis of detecting cabinet air outlet temperature prediction data, relatively accurate cold pool air inlet target air quantity data and cold pool air inlet air quantity data adjusting quantity are given, and machine room operation and maintenance personnel are not needed to adjust the air quantity of the cold pool air inlet according to work experience, so that in a machine room production environment, the phenomenon that the temperature of the cabinet is too high in the working process can be conveniently and efficiently relieved and even avoided.
Some embodiments of the temperature prediction devices of the present disclosure are described below in conjunction with fig. 5.
Fig. 5 shows a schematic structural diagram of a temperature prediction device according to some embodiments of the present disclosure. As shown in fig. 5, the temperature prediction device 50 in the present embodiment includes: a data acquisition module 501 configured to: acquiring air conditioner air outlet data, server load data and cold pool air inlet air volume data; a first data prediction module 502 configured to: inputting air conditioner air outlet data into a pre-trained first machine learning model to obtain temperature and humidity prediction data of an air inlet of a cold pool; the second data prediction module 503 is configured to input the temperature and humidity prediction data of the cold pool air inlet, the server load data, and the air volume data of the cold pool air inlet into a second machine learning model trained in advance, so as to obtain the temperature prediction data of the cabinet air outlet.
The embodiment provides an intelligent temperature prediction method applied to a machine room, wherein a sensor is adopted to collect data and model in an early data collection stage, the sensor is not required to be adopted in a later temperature prediction stage, and a machine learning model is adopted to realize full-time and real-time prediction on the temperature of an air outlet of a cabinet, so that the temperature of the air outlet of the cabinet can be obtained conveniently and efficiently in real time, the cost is low, and the operation and maintenance are convenient.
The following are specifically mentioned: on one hand, in the embodiment, the temperature of the air outlet of the cabinet is not directly predicted through air conditioner air outlet data and server load data, but modeling on temperature and humidity data of the air inlet of the cold pool is added, so that heat loss in the middle process of airflow flow direction can be reflected; on the other hand, this embodiment has still considered the influence of the fluctuation of server load to rack air outlet temperature from the angle of generating heat to the influence of cold pond air intake humiture to rack air outlet temperature has been considered from the refrigeration angle. Therefore, the temperature of the air outlet of the cabinet can be accurately predicted by the embodiment.
In some embodiments, the temperature prediction device further comprises: a data determination module configured to: determining target air volume data of the air inlet of the cold pool according to the temperature and humidity prediction data of the air inlet of the cold pool, server load data, cabinet air outlet temperature target data and a pre-trained second machine learning model; a data conditioning module configured to: and adjusting the air volume data of the air inlet of the cold pool into target air volume data of the air inlet of the cold pool so as to adjust the temperature prediction data of the air outlet of the cabinet into target temperature data of the air outlet of the cabinet.
In some embodiments, the data determination module is configured to: traversing the air quantity test data of the air inlet of the cold pool by a preset step length; inputting the temperature and humidity prediction data of the cold pool air inlet, the server load data and the cold pool air inlet air quantity test data into a pre-trained second machine learning model to obtain cabinet air outlet temperature test data; and when the cabinet air outlet temperature test data is equal to the cabinet air outlet temperature target data, taking the cold pool air inlet test air volume data as cold pool air inlet target air volume data.
According to the air conditioner cabinet air inlet air quantity data adjusting method and device, on the basis of detecting cabinet air outlet temperature prediction data, relatively accurate cold pool air inlet target air quantity data and cold pool air inlet air quantity data adjusting quantity are given, and machine room operation and maintenance personnel are not needed to adjust the air quantity of the cold pool air inlet according to work experience, so that in a machine room production environment, the phenomenon that the temperature of the cabinet is too high in the working process can be conveniently and efficiently relieved and even avoided.
Some embodiments of the disclosed model training apparatus are described below in conjunction with FIG. 6.
FIG. 6 illustrates a schematic structural diagram of a model training apparatus according to some embodiments of the present disclosure. As shown in fig. 6, the model training device 60 in the present embodiment includes:
a first model training module 601 configured to: collecting air conditioner air outlet sample data and cold pool air inlet temperature and humidity sample data; training a first machine learning model by using air conditioner air outlet sample data and cold pool air inlet temperature and humidity sample data, so that the first machine learning model can obtain cold pool air inlet temperature and humidity prediction data according to the air conditioner air outlet data; a second model training module 602 configured to: collecting temperature and humidity sample data of an air inlet of a cold pool, server load sample data, air quantity sample data of the air inlet of the cold pool and temperature sample data of an air outlet of a cabinet; and training a second machine learning model by using the temperature and humidity sample data of the cold pool air inlet, the server load sample data, the air volume sample data of the cold pool air inlet and the cabinet air outlet temperature sample data, so that the second machine learning model can obtain cabinet air outlet temperature prediction data according to the temperature and humidity prediction data of the cold pool air inlet, the server load data and the air volume data of the cold pool air inlet.
Some embodiments of the disclosed electronic device are described below in conjunction with fig. 7.
Fig. 7 shows a schematic structural diagram of an electronic device of some embodiments of the present disclosure. As shown in fig. 7, the electronic apparatus 70 of this embodiment includes: a memory 710 and a processor 720 coupled to the memory 710, the processor 720 being configured to perform the temperature prediction method or the model training method in any of the embodiments based on instructions stored in the memory 710.
Memory 710 may include, for example, system memory, fixed non-volatile storage media, and the like. The system memory stores, for example, an operating system, an application program, a Boot Loader (Boot Loader), and other programs.
The temperature prediction device 70 may further include an input-output interface 730, a network interface 740, a storage interface 750, and the like. These interfaces 730, 740, 750, as well as the memory 710 and the processor 720, may be connected, for example, by a bus 760. The input/output interface 730 provides a connection interface for input/output devices such as a display, a mouse, a keyboard, and a touch screen. The network interface 740 provides a connection interface for various networking devices. The storage interface 750 provides a connection interface for external storage devices such as an SD card and a usb disk.
The present disclosure also includes a computer readable storage medium having stored thereon computer instructions that, when executed by a processor, implement the temperature prediction method or the aforementioned model training method in any of the foregoing embodiments.
The present disclosure is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the disclosure. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
The above description is only exemplary of the present disclosure and is not intended to limit the present disclosure, so that any modification, equivalent replacement, or improvement made within the spirit and principle of the present disclosure should be included in the scope of the present disclosure.

Claims (10)

1. A method of temperature prediction, comprising:
acquiring air conditioner air outlet data, server load data and cold pool air inlet air volume data;
inputting air conditioner air outlet data into a pre-trained first machine learning model to obtain temperature and humidity prediction data of an air inlet of a cold pool;
and inputting the temperature and humidity prediction data of the air inlet of the cold pool, the server load data and the air quantity data of the air inlet of the cold pool into a pre-trained second machine learning model to obtain the temperature prediction data of the air outlet of the cabinet.
2. The temperature prediction method of claim 1, further comprising:
determining target air volume data of the air inlet of the cold pool according to the temperature and humidity prediction data of the air inlet of the cold pool, server load data, cabinet air outlet temperature target data and a pre-trained second machine learning model;
and adjusting the air volume data of the air inlet of the cold pool into target air volume data of the air inlet of the cold pool so as to adjust the temperature prediction data of the air outlet of the cabinet into target temperature data of the air outlet of the cabinet.
3. The temperature prediction method according to claim 2, wherein the determining the target air volume data of the cold pool air inlet according to the temperature and humidity prediction data of the cold pool air inlet, the server load data, the cabinet air outlet temperature target data, and a pre-trained second machine learning model comprises:
traversing the air quantity test data of the air inlet of the cold pool by a preset step length;
inputting the temperature and humidity prediction data of the cold pool air inlet, the server load data and the cold pool air inlet air quantity test data into a pre-trained second machine learning model to obtain cabinet air outlet temperature test data;
and when the cabinet air outlet temperature test data is equal to the cabinet air outlet temperature target data, taking the cold pool air inlet test air volume data as cold pool air inlet target air volume data.
4. A model training method, comprising:
collecting air conditioner air outlet sample data and cold pool air inlet temperature and humidity sample data;
training a first machine learning model by using air conditioner air outlet sample data and cold pool air inlet temperature and humidity sample data, so that the first machine learning model can obtain cold pool air inlet temperature and humidity prediction data according to the air conditioner air outlet data;
collecting temperature and humidity sample data of an air inlet of a cold pool, server load sample data, air quantity sample data of the air inlet of the cold pool and temperature sample data of an air outlet of a cabinet;
and training a second machine learning model by using the temperature and humidity sample data of the cold pool air inlet, the server load sample data, the air volume sample data of the cold pool air inlet and the cabinet air outlet temperature sample data, so that the second machine learning model can obtain cabinet air outlet temperature prediction data according to the temperature and humidity prediction data of the cold pool air inlet, the server load data and the air volume data of the cold pool air inlet.
5. A temperature prediction device, comprising:
a data acquisition module configured to: acquiring air conditioner air outlet data, server load data and cold pool air inlet air volume data;
a first data prediction module configured to: inputting air conditioner air outlet data into a pre-trained first machine learning model to obtain temperature and humidity prediction data of an air inlet of a cold pool;
and the second data prediction module is configured to input the temperature and humidity prediction data of the cold pool air inlet, the server load data and the air volume data of the cold pool air inlet into a pre-trained second machine learning model to obtain the temperature prediction data of the cabinet air outlet.
6. The temperature prediction device of claim 5, further comprising:
a data determination module configured to: determining target air volume data of the air inlet of the cold pool according to the temperature and humidity prediction data of the air inlet of the cold pool, server load data, cabinet air outlet temperature target data and a pre-trained second machine learning model;
a data conditioning module configured to: and adjusting the air volume data of the air inlet of the cold pool into target air volume data of the air inlet of the cold pool so as to adjust the temperature prediction data of the air outlet of the cabinet into target temperature data of the air outlet of the cabinet.
7. The temperature prediction device of claim 6, wherein the data determination module is configured to:
traversing the air quantity test data of the air inlet of the cold pool by a preset step length;
inputting the temperature and humidity prediction data of the cold pool air inlet, the server load data and the cold pool air inlet air quantity test data into a pre-trained second machine learning model to obtain cabinet air outlet temperature test data;
and when the cabinet air outlet temperature test data is equal to the cabinet air outlet temperature target data, taking the cold pool air inlet test air volume data as cold pool air inlet target air volume data.
8. A model training apparatus comprising:
a first model training module configured to: collecting air conditioner air outlet sample data and cold pool air inlet temperature and humidity sample data; training a first machine learning model by using air conditioner air outlet sample data and cold pool air inlet temperature and humidity sample data, so that the first machine learning model can obtain cold pool air inlet temperature and humidity prediction data according to the air conditioner air outlet data;
a second model training module configured to: collecting temperature and humidity sample data of an air inlet of a cold pool, server load sample data, air quantity sample data of the air inlet of the cold pool and temperature sample data of an air outlet of a cabinet; and training a second machine learning model by using the temperature and humidity sample data of the cold pool air inlet, the server load sample data, the air volume sample data of the cold pool air inlet and the cabinet air outlet temperature sample data, so that the second machine learning model can obtain cabinet air outlet temperature prediction data according to the temperature and humidity prediction data of the cold pool air inlet, the server load data and the air volume data of the cold pool air inlet.
9. An electronic device, comprising:
a memory; and
a processor coupled to the memory, the processor configured to perform the temperature prediction method of any of claims 1-3 or the model training method of claim 4 based on instructions stored in the memory.
10. A computer readable storage medium, wherein the computer readable storage medium stores computer instructions which, when executed by a processor, implement the temperature prediction method of any one of claims 1 to 3 or the model training method of claim 4.
CN202010795254.2A 2020-08-10 2020-08-10 Temperature prediction method, model training method and related equipment Pending CN114065602A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010795254.2A CN114065602A (en) 2020-08-10 2020-08-10 Temperature prediction method, model training method and related equipment

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010795254.2A CN114065602A (en) 2020-08-10 2020-08-10 Temperature prediction method, model training method and related equipment

Publications (1)

Publication Number Publication Date
CN114065602A true CN114065602A (en) 2022-02-18

Family

ID=80233015

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010795254.2A Pending CN114065602A (en) 2020-08-10 2020-08-10 Temperature prediction method, model training method and related equipment

Country Status (1)

Country Link
CN (1) CN114065602A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115685941A (en) * 2022-11-04 2023-02-03 中国电子工程设计院有限公司 Machine room operation regulation and control method and device based on cabinet hot spot temperature prediction

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115685941A (en) * 2022-11-04 2023-02-03 中国电子工程设计院有限公司 Machine room operation regulation and control method and device based on cabinet hot spot temperature prediction

Similar Documents

Publication Publication Date Title
CN109800066B (en) Energy-saving scheduling method and system for data center
CN110866528B (en) Model training method, energy consumption use efficiency prediction method, device and medium
CN102620378B (en) Method and system for data center energy saving controlling
JP5511698B2 (en) Air conditioner linkage control system, air conditioner linkage control method, and air conditioner linkage control program
CN108595301B (en) Server energy consumption prediction method and system based on machine learning
US9996092B2 (en) Determining a time for corrective action in a data center
CN107223195A (en) Variable air quantity for HVAC system is modeled
US20140303789A1 (en) Automated technique of measuring room air change rates in hvac system
CN116193819B (en) Energy-saving control method, system and device for data center machine room and electronic equipment
Van Le et al. Control of air free-cooled data centers in tropics via deep reinforcement learning
CN110852498A (en) Method for predicting data center energy consumption efficiency PUE based on GRU neural network
CN117413147A (en) Dynamic prediction control method, device and system for precise air conditioner
CN105357081A (en) Server test system, method, control system and method
CN114065602A (en) Temperature prediction method, model training method and related equipment
CN111814388A (en) CFD simulation verification method for lower air supply data center based on neural network
CN113821903A (en) Temperature control method and device, modular data center and storage medium
JP2017151617A (en) Simulation device, simulation method, and program
CN114312213B (en) Method, device, equipment and medium for predicting abnormal refrigerant capacity of vehicle
CN113574510A (en) Data center management system and method for calculating density efficiency measurements
CN113849052A (en) Machine room temperature prediction method and system based on artificial intelligence
JP2014214939A (en) Business arrangement optimization system and business arrangement optimization method
CN207457875U (en) One kind can program constant temperature and moisture test machine
WO2013176681A1 (en) Providing a model of impact of a cooling infrastructure
Ogawa et al. Cooling control restraining effects due to ICT equipment utilization of disturbance based on model predictive control for modular data center
JP2022028253A (en) Abnormality detection system, information processing apparatus, and information processing program

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination