CN114219177A - Computer room environment regulation and control method and device, electronic equipment and storage medium - Google Patents

Computer room environment regulation and control method and device, electronic equipment and storage medium Download PDF

Info

Publication number
CN114219177A
CN114219177A CN202111638846.4A CN202111638846A CN114219177A CN 114219177 A CN114219177 A CN 114219177A CN 202111638846 A CN202111638846 A CN 202111638846A CN 114219177 A CN114219177 A CN 114219177A
Authority
CN
China
Prior art keywords
environment
machine room
neural network
principal component
factor data
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
CN202111638846.4A
Other languages
Chinese (zh)
Inventor
邵海涛
卢道和
罗锶
张晓通
曾可
黄耿冬
郭江涛
鲁东东
冯期明
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.)
WeBank Co Ltd
Original Assignee
WeBank Co 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 WeBank Co Ltd filed Critical WeBank Co Ltd
Priority to CN202111638846.4A priority Critical patent/CN114219177A/en
Publication of CN114219177A publication Critical patent/CN114219177A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
    • 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
    • G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
    • G06F18/2135—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods based on approximation criteria, e.g. principal component analysis
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • G06N3/084—Backpropagation, e.g. using gradient descent

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • General Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Artificial Intelligence (AREA)
  • Evolutionary Computation (AREA)
  • Computational Linguistics (AREA)
  • Economics (AREA)
  • Strategic Management (AREA)
  • Health & Medical Sciences (AREA)
  • Biomedical Technology (AREA)
  • Biophysics (AREA)
  • Human Resources & Organizations (AREA)
  • General Health & Medical Sciences (AREA)
  • Molecular Biology (AREA)
  • Computing Systems (AREA)
  • Mathematical Physics (AREA)
  • Software Systems (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Development Economics (AREA)
  • Evolutionary Biology (AREA)
  • Game Theory and Decision Science (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Marketing (AREA)
  • Operations Research (AREA)
  • Quality & Reliability (AREA)
  • Tourism & Hospitality (AREA)
  • General Business, Economics & Management (AREA)
  • Air Conditioning Control Device (AREA)

Abstract

The application provides a method and a device for regulating and controlling a machine room environment, electronic equipment and a storage medium. The method comprises the following steps: the method comprises the steps of constructing a neural network prediction model, obtaining environment factor data of a machine room, generating environment adjusting information of the machine room according to the environment factor data and the neural network prediction model, and regulating and controlling the environment of the machine room according to the environment adjusting information. Because the environment factor data of the machine room is processed by utilizing the neural network prediction model to generate the environment regulation information of the machine room, the environment of the machine room is regulated and controlled according to the environment regulation information, the prejudgment of the regulation information of the machine room by utilizing the environment factor data is realized, the problem that in the prior art, the temperature and humidity data in the machine room of a data center are collected at fixed points through a temperature and humidity probe, and the problem of hysteresis exists in a mode of carrying out instruction or artificial regulation and control by comparing temperature and humidity safety threshold values is solved, and the accuracy of machine room regulation and control is improved.

Description

Computer room environment regulation and control method and device, electronic equipment and storage medium
Technical Field
The application relates to the field of science and technology finance, in particular to a method and a device for regulating and controlling a machine room environment, electronic equipment and a storage medium.
Background
With the development of computer technology, more and more technologies are applied in the financial field, the traditional financial industry is gradually changing to financial technology (Finteh), and the ordered classification technology is no exception, but due to the requirements of the financial industry on safety and real-time performance, higher requirements are also put forward on the technology. With the rapid development of communication and network technologies, the scale and power density of Data centers (Internet Data centers) are increasing, and a large number of Internet Technology (IT) devices are concentrated in a Data Center room. In order to enable the data center machine room to meet the technical requirements of constant temperature and constant humidity, a refrigeration and humidity control system is generally required to be arranged to control the temperature and humidity of the data center machine room.
In the prior art, in order to control the temperature and humidity of a data center machine room, temperature and humidity data of each local point inside the data center machine room are collected by a temperature and humidity monitoring probe and reported to a central monitoring system, and a worker manually regulates and controls the cooling capacity of an air conditioner and the power output of a humidifier according to the temperature and humidity data and a set temperature and humidity safety threshold and working experience, so that the data center machine room can meet the environmental requirements of constant temperature and constant humidity.
However, in the prior art, the humiture data in the data center machine room is collected through the humiture probe fixed point, and hysteresis exists in the mode of carrying out instruction or artificial regulation by comparing the humiture safety threshold, so that the accuracy of machine room regulation is low.
Disclosure of Invention
The application provides a machine room environment regulation and control method and device, electronic equipment and a storage medium, which are used for solving the technical problem that the machine room environment regulation and control in the prior art is low in accuracy.
In a first aspect, the present application provides a method for regulating a machine room environment, comprising:
constructing a neural network prediction model; acquiring environmental factor data of a machine room; generating environment adjusting information of the machine room according to the environment factor data and the neural network prediction model; and regulating and controlling the environment of the machine room according to the environment regulation information.
In the embodiment of the application, the environmental factor data of the machine room are processed by utilizing the neural network prediction model, the environmental regulation information of the machine room is generated, and then the environment of the machine room is regulated and controlled according to the environmental regulation information, the utilization of the environmental factor data is realized, the pre-judgment of the machine room regulation information is realized, the problem that in the prior art, the temperature and humidity data in the machine room of a data center are collected at fixed points through temperature and humidity probes, the problem of hysteresis exists in the mode of carrying out instruction or artificial regulation and control by comparing temperature and humidity safety threshold values is solved, and the accuracy of machine room regulation and control is improved.
In a possible implementation manner, the method for regulating and controlling a machine room environment provided in an embodiment of the present application further includes, before constructing the neural network prediction model:
and analyzing the historical environmental factor data of the machine room according to a principal component analysis method to determine the principal component factors in the environmental factors.
The neural network prediction model construction method comprises the following steps: and constructing a neural network prediction model according to the principal component factors.
In the embodiment of the application, the historical environment factor data is analyzed by adopting a principal component analysis method, the principal component factors in the environment factors are determined, and then the neural network model is constructed by utilizing the principal component factors, so that the accuracy and granularity of the neural network prediction model are improved.
In a possible implementation manner, the method for regulating and controlling a machine room environment provided in the embodiment of the present application constructs a neural network prediction model according to a principal component factor, including:
constructing a neural network structure of a neural network prediction model; generating a training sample pair, wherein the training sample pair comprises a principal component factor data sample and an environment regulation information sample; and training the neural network structure by using the training sample pair to generate a neural network prediction model.
In a possible implementation manner, in the method for regulating and controlling a computer room environment provided by the embodiment of the present application, the neural network structure includes an input layer, a hidden layer, and an output layer,
the input layer is used for inputting the principal component factor data samples; the input of the concealment layer is a first weighted sum of the principal component factor data samples, and the output of the concealment layer is a first transfer function value of the first weighted sum; the input to the output layer is a second weighted sum of the first transfer function values and the output of the output layer is a second transfer function value of the second weighted sum.
In a possible implementation manner, the method for regulating and controlling a machine room environment provided in an embodiment of the present application, which trains a neural network structure by using a training sample pair, includes:
determining a first weighting adjustment quantity and a second weighting adjustment quantity of a neural network structure, wherein the first weighting adjustment quantity is the weighting adjustment quantity from an input layer to a hidden layer, and the second weighting adjustment quantity is the weighting adjustment quantity from the hidden layer to an output layer; inputting the principal component factor data sample into a neural network structure to generate predicted environment regulation information; determining the weighted correction quantity of the first weighted adjustment quantity and the second weighted adjustment quantity by utilizing the predicted environment adjustment information and the environment adjustment information sample; and updating the first weighted adjustment quantity and the second weighted adjustment quantity of the neural network structure by using the weighted correction quantity, and then training the training sample pair by using the updated neural network structure until the error between the predicted environment adjustment information and the environment adjustment information sample meets the preset condition, and finishing the training.
In a possible implementation manner, the method for regulating and controlling a machine room environment provided in an embodiment of the present application generates a training sample pair, including:
determining a principal component factor data sample and an environmental conditioning information sample; and carrying out normalization processing on the principal component factor data sample and the environmental conditioning information sample to generate a training sample pair.
In the embodiment of the application, after the principal component factor data samples and the environmental regulation information samples are determined, normalization processing is carried out on the principal component factor data samples and the environmental regulation information samples, so that data abnormity of different principal component factor data samples due to dimension difference is prevented, and certain principal component factor data samples are prevented from being masked due to small numerical values.
In a possible implementation manner, the method for regulating and controlling an environment of a machine room, provided by the embodiment of the present application, analyzes historical environmental factor data of the machine room according to a principal component analysis method, and determines a principal component factor in the environmental factors, including:
carrying out standardization processing on the historical environmental factor data to generate a standardized data matrix of the historical environmental factor data; determining a correlation coefficient matrix of the normalized data matrix; determining the contribution rate and the accumulated contribution rate of each environmental factor by using a correlation coefficient matrix; and determining the principal component factor according to the contribution rate and the accumulated contribution rate.
In the embodiment of the application, the reliability of the principal component factors is improved by analyzing and processing the historical environmental factor data and determining the principal component factors according to the contribution rate and the accumulated contribution rate of each environmental factor.
In a possible implementation manner, the method for regulating and controlling an environment of a machine room, provided by the embodiment of the present application, determines a principal component factor according to a contribution rate and an accumulated contribution rate, and includes:
determining the least quantity of the environmental factors as a target quantity when the accumulated contribution rate reaches a preset contribution rate; and determining the environmental factors of the target quantity as principal component factors according to the magnitude sequence of the contribution rates.
The computer room environment control device, the electronic device, the computer readable storage medium, and the computer program product provided in the embodiments of the present application are described below, and the content and effect of the computer room environment control device, the electronic device, the computer readable storage medium, and the computer program product may refer to the computer room environment control method provided in the embodiments of the present application, and are not described again.
In a second aspect, the present application provides a machine room environment conditioning device, comprising:
and the building module is used for building the neural network prediction model.
And the acquisition module is used for acquiring the environmental factor data of the machine room.
And the processing module is used for generating the environment regulation information of the machine room according to the environment factor data and the neural network prediction model.
And the regulation and control module is used for regulating and controlling the environment of the machine room according to the environment regulation information.
In a possible implementation manner, the apparatus for regulating and controlling an environment of a machine room provided in an embodiment of the present application further includes:
and the determining module is used for analyzing the historical environmental factor data of the machine room according to a principal component analysis method and determining the principal component factors in the environmental factors.
A building block, specifically configured to: and constructing a neural network prediction model according to the principal component factors.
In a possible implementation manner, the machine room environment control device provided in the embodiment of the present application constructs a module, which is specifically configured to:
constructing a neural network structure of a neural network prediction model; generating a training sample pair, wherein the training sample pair comprises a principal component factor data sample and an environment regulation information sample; and training the neural network structure by using the training sample pair to generate a neural network prediction model.
In a possible implementation manner, the computer room environment regulation and control device provided by the embodiment of the application, the neural network structure includes an input layer, a hidden layer and an output layer,
the input layer is used for inputting the principal component factor data samples; the input of the concealment layer is a first weighted sum of the principal component factor data samples, and the output of the concealment layer is a first transfer function value of the first weighted sum; the input to the output layer is a second weighted sum of the first transfer function values and the output of the output layer is a second transfer function value of the second weighted sum.
In a possible implementation manner, the machine room environment control device provided in the embodiment of the present application constructs a module, which is specifically configured to:
determining a first weighting adjustment quantity and a second weighting adjustment quantity of a neural network structure, wherein the first weighting adjustment quantity is the weighting adjustment quantity from an input layer to a hidden layer, and the second weighting adjustment quantity is the weighting adjustment quantity from the hidden layer to an output layer; inputting the principal component factor data sample into a neural network structure to generate predicted environment regulation information; determining the weighted correction quantity of the first weighted adjustment quantity and the second weighted adjustment quantity by utilizing the predicted environment adjustment information and the environment adjustment information sample; and updating the first weighted adjustment quantity and the second weighted adjustment quantity of the neural network structure by using the weighted correction quantity, and then training the training sample pair by using the updated neural network structure until the error between the predicted environment adjustment information and the environment adjustment information sample meets the preset condition, and finishing the training.
In a possible implementation manner, the machine room environment control device provided in the embodiment of the present application constructs a module, which is specifically configured to:
determining a principal component factor data sample and an environmental conditioning information sample; and carrying out normalization processing on the principal component factor data sample and the environmental conditioning information sample to generate a training sample pair.
In a possible implementation manner, the determining module of the machine room environment regulation and control device provided in the embodiment of the present application is specifically configured to:
carrying out standardization processing on the historical environmental factor data to generate a standardized data matrix of the historical environmental factor data; determining a correlation coefficient matrix of the normalized data matrix; determining the contribution rate and the accumulated contribution rate of each environmental factor by using a correlation coefficient matrix; and determining the principal component factor according to the contribution rate and the accumulated contribution rate.
In a possible implementation manner, the determining module of the machine room environment regulation and control device provided in the embodiment of the present application is specifically configured to:
determining the least quantity of the environmental factors as a target quantity when the accumulated contribution rate reaches a preset contribution rate; and determining the environmental factors of the target quantity as principal component factors according to the magnitude sequence of the contribution rates.
In a third aspect, an embodiment of the present application provides an electronic device, including:
a processor, and a memory communicatively coupled to the processor;
the memory stores computer-executable instructions;
the processor executes the computer execution instructions stored in the memory to implement the computer room environment regulation and control method provided by the first aspect or the implementable manner of the first aspect.
In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, where computer-executable instructions are stored in the computer-readable storage medium, and when the computer-executable instructions are executed by a processor, the computer-executable instructions are used to implement the computer room environment regulation and control method provided in the first aspect or the implementable manner of the first aspect.
In a fifth aspect, an embodiment of the present application provides a computer program product, which includes computer executable instructions, and the computer executable instructions are executed by a processor to implement the computer room environment regulation and control method provided in the first aspect or the implementable manner of the first aspect.
According to the method, the device, the electronic equipment and the storage medium for regulating and controlling the environment of the machine room, the neural network prediction model is built, the environment factor data of the machine room are obtained, then the environment regulation information of the machine room is generated according to the environment factor data and the neural network prediction model, and finally the environment of the machine room is regulated and controlled according to the environment regulation information. Because the environment factor data of the machine room is processed by utilizing the neural network prediction model to generate the environment regulation information of the machine room, the environment of the machine room is regulated and controlled according to the environment regulation information, the prejudgment of the regulation information of the machine room by utilizing the environment factor data is realized, the problem that in the prior art, the temperature and humidity data in the machine room of a data center are collected at fixed points through a temperature and humidity probe, and the problem of hysteresis exists in a mode of carrying out instruction or artificial regulation and control by comparing temperature and humidity safety threshold values is solved, and the accuracy of machine room regulation and control is improved.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and together with the description, serve to explain the principles of the application.
FIG. 1 is an exemplary application scenario architecture diagram provided by an embodiment of the present application;
fig. 2 is a schematic flow chart of a machine room environment regulation and control method provided in an embodiment of the present application;
fig. 3 is a schematic flow chart of a machine room environment regulation and control method according to another embodiment of the present application;
FIG. 4 is a schematic structural diagram of a neural network structure provided in an embodiment of the present application;
FIG. 5 is a schematic diagram of a neuron according to an embodiment of the present application;
fig. 6 is a schematic structural diagram of a machine room environment control device according to an embodiment of the present application;
fig. 7 is a schematic structural diagram of a machine room environment control device according to another embodiment of the present application;
fig. 8 is a schematic structural diagram of an electronic device provided in an embodiment of the present application.
With the above figures, there are shown specific embodiments of the present application, which will be described in more detail below. These drawings and written description are not intended to limit the scope of the inventive concepts in any manner, but rather to illustrate the inventive concepts to those skilled in the art by reference to specific embodiments.
Detailed Description
Reference will now be made in detail to the exemplary embodiments, examples of which are illustrated in the accompanying drawings. When the following description refers to the accompanying drawings, like numbers in different drawings represent the same or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present application. Rather, they are merely examples of apparatus and methods consistent with certain aspects of the present application, as detailed in the appended claims.
The terms "first" and "second" in the description and claims of the present application and the above-described drawings are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used is interchangeable under appropriate circumstances such that the embodiments of the application described herein are, for example, capable of operation in sequences other than those illustrated or otherwise described herein. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed, but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.
With the development of computer technology, more and more technologies are applied in the financial field, the traditional financial industry is gradually changing to financial technology, and the ordered classification technology is not an exception, but higher requirements are also put forward on the technologies due to the requirements of the financial industry on safety and real-time performance. Along with the rapid development of communication and network technologies, the scale and power density of a data center are continuously increased, in the prior art, in order to control the temperature and humidity of a data center machine room, temperature and humidity data of each local point in the data center machine room are collected by using a temperature and humidity monitoring probe generally, and are reported to a central monitoring system, and a worker manually regulates and controls the cooling capacity of an air conditioner and the power output of a humidifier according to the temperature and humidity data and according to a set temperature and humidity safety threshold and working experience.
In order to solve the technical problem, the invention idea of the method and the device for regulating and controlling the environment of the machine room, the electronic device and the storage medium provided by the embodiment of the application is that the environment factor data of the machine room is processed by using the neural network prediction model to generate the environment regulation information of the machine room, and then the environment of the machine room is regulated and controlled according to the environment regulation information, so that the prejudgment on the regulation information of the machine room by using the environment factor data is realized, and the accuracy of the regulation and control of the machine room is improved. Furthermore, the embodiment of the application analyzes the historical environment factor data by using a principal component analysis method, determines the principal component factors in the environment factors, and further constructs a neural network model by using the principal component factors, so that the accuracy and the granularity of the neural network prediction model are improved.
An exemplary application scenario of the embodiments of the present application is described below.
The machine room environment regulation and control method provided by the embodiment of the application can be executed through the machine room environment regulation and control device provided by the embodiment of the application, the machine room environment regulation and control device provided by the embodiment of the application can be integrated on terminal equipment or a server, or the machine room environment regulation and control device can be the terminal equipment or the server.
The method, the device, the electronic equipment and the storage medium for regulating and controlling the computer room environment provided by the embodiment of the application can be applied to a computer room of a data center, wherein the data center is a place for carrying out concentration, integration, sharing and analysis on data resources and is a business system of an enterprise. The data center machine room can be divided into a main machine room, an auxiliary area, an office area and the like according to functions, and the infrastructure of the data center machine room mainly comprises: the system comprises a power supply and distribution system, an air conditioner and ventilation system, a lightning protection and grounding system, a fire alarm and automatic fire extinguishing system, a weak current system, IT equipment and the like.
In a possible implementation manner, fig. 1 is an exemplary application scenario architecture diagram provided in an embodiment of the present application, and as shown in fig. 1, the architecture mainly includes: terminal equipment, server, air conditioner and humidifier. For example, according to the embodiment of the application, a neural network prediction model can be built through terminal equipment, the terminal equipment can obtain environment factor data of a machine room through a server, then the terminal equipment inputs the environment factor data into the neural network prediction model, the neural network prediction model is processed to output environment adjustment information of the machine room, the environment adjustment information can be control information of an air conditioner and/or control information of a humidifier, and the terminal equipment can send the environment adjustment information to the air conditioner and/or the humidifier to further control the air conditioner and/or the humidifier to work, so that the environment of the machine room is regulated and controlled. The embodiment of the present application is only an example of this implementation scenario, and is not limited to this scenario.
The following describes the technical solutions of the present application and how to solve the above technical problems with specific embodiments. The following several specific embodiments may be combined with each other, and details of the same or similar concepts or processes may not be repeated in some embodiments. Embodiments of the present application will be described below with reference to the accompanying drawings.
Fig. 2 is a schematic flow chart of a method for controlling a machine room environment according to an embodiment of the present application, where the method may be executed by a machine room environment controlling device, and the device may be implemented in a software and/or hardware manner, and the method for controlling a machine room environment is described below by taking a terminal device as an execution subject as an example. As shown in fig. 2, the method for regulating and controlling a machine room environment provided in the embodiment of the present application may include:
step S101: and constructing a neural network prediction model.
A causal relationship exists between environment factor data and environment regulation information in a data center machine room, but the environment factor data and the environment regulation information have a strong nonlinear relationship and are difficult to describe through some analytical expressions, a neural network is a highly self-adaptive nonlinear power system, massive original data are extracted from a historical database through a BP (Back propagation) neural network prediction model to perform modeling learning, and a highly nonlinear mapping relationship between input and output can be obtained.
Therefore, a nonlinear relationship between the environmental factor data of the machine room and the environmental regulation information can be established by using the neural network prediction model, wherein the environmental regulation information can be information such as air conditioner cooling capacity requirement, humidifier power and the like.
Therefore, the nonlinear relation between the data input of the environmental factors of the machine room and the output of the air conditioner cold quantity demand under the constant temperature demand of the machine room can be realized by constructing the neural network prediction model, and/or the nonlinear relation between the data input of the environmental factors of the machine room and the power output of the humidifier under the constant humidity demand of the machine room can be realized.
Step S102: and acquiring environmental factor data of the machine room.
The environmental factors affecting the environment inside the machine room may be various, for example, the environmental factor data may include one or more of the following, for example: the system comprises an IT load in the machine room, an external environment temperature, an external environment humidity, an external environment wind speed, an external environment air pressure, an external environment illumination intensity, a machine room internal space volume, machine room floors, machine room orientation, machine room longitude and latitude and the like. The environmental factor data may have a certain influence on the change of the temperature and humidity inside the machine room.
The method includes the steps that environmental factor data of a machine room are obtained through a server, for example, environmental factor data of the current external environment such as temperature, humidity, weather, wind speed, air pressure and illumination intensity are obtained from the server in real time, or environmental factor data of the current external environment such as temperature, humidity, weather, wind speed, air pressure and illumination intensity are obtained at intervals of preset time; the fixed parameters of the machine room can be obtained locally, for example, the size of the internal space of the machine room, the floor of the machine room, the orientation of the machine room and other environmental factor data.
Step S103: and generating environment regulation information of the machine room according to the environment factor data and the neural network prediction model.
After the environmental factor data is obtained, the environmental factor data is input into a neural network prediction model, and the neural network prediction model outputs environmental regulation information of the computer room through a series of calculations. In a possible implementation manner, the environment regulation information may be information such as air conditioner cooling capacity requirement, humidifier power and the like, and is used for respectively regulating and controlling the air conditioner and the humidifier in the machine room, so that the requirements of the machine room for constant temperature and constant humidity are met. The embodiment of the application does not limit the specific requirements of the machine room for constant temperature and humidity, and in a possible implementation manner, the temperature requirement inside the machine room is 22 +/-2 ℃, and the humidity requirement of the machine room is 45% +/-10%.
Step S104: and regulating and controlling the environment of the machine room according to the environment regulation information.
The environment regulation information of the machine room is generated through the neural network prediction model, in a possible implementation manner, the terminal device can control the control object device in a manner of sending an environment regulation and control instruction to the control object device, so as to further realize regulation and control of the machine room environment, for example, the environment regulation and control information is humidifier power, the terminal device can send a humidifier regulation and control instruction to the humidifier, and the humidifier regulation and control instruction is used for indicating the humidifier to regulate power to the humidifier power. The embodiments of the present application are merely examples, and are not limited thereto.
In the embodiment of the application, the environment factor data of the machine room is processed by utilizing the neural network prediction model, the environment regulation information of the machine room is generated, the environment of the machine room is regulated and controlled according to the environment regulation information, the utilization of the environment factor data is realized, the prejudgment of the machine room regulation information is realized, the problem of hysteresis in the existing scheme is solved, and the problem of abnormal temperature and humidity of the machine room caused by experience and misoperation of workers is avoided by reducing human intervention.
In order to improve the accuracy of the neural network prediction model, in a possible implementation manner, fig. 3 is a schematic flow diagram of a machine room environment regulation and control method provided in another embodiment of the present application, as shown in fig. 3, in step S101: before the neural network prediction model is constructed, the method may further include:
step S201: and analyzing the historical environmental factor data of the machine room according to a principal component analysis method to determine the principal component factors in the environmental factors.
The principal component analysis method aims to convert a plurality of environment factors into a few principal component factors by using the idea of dimension reduction, wherein each principal component factor can reflect most information of the original environment factor, and the contained information is not repeated.
The embodiment of the application does not limit the specific process of analyzing the historical environmental factor data of the machine room according to the principal component analysis method, the number of the principal component factors in the environmental factors and the like.
In a possible implementation manner, the method for regulating and controlling an environment of a machine room, provided by the embodiment of the present application, analyzes historical environmental factor data of the machine room according to a principal component analysis method, and determines a principal component factor in the environmental factors, including:
carrying out standardization processing on the historical environmental factor data to generate a standardized data matrix of the historical environmental factor data; determining a correlation coefficient matrix of the normalized data matrix; determining the contribution rate and the accumulated contribution rate of each environmental factor by using a correlation coefficient matrix; and determining the principal component factor according to the contribution rate and the accumulated contribution rate.
For convenience of understanding, the embodiments of the present application exemplify that the historical environment factor data includes the following seven types of data: the size of the internal space of the machine room, the illumination intensity of the external environment of the machine room, the air pressure of the external environment of the machine room, the wind speed of the external environment of the machine room, the humidity of the external environment of the machine room, the temperature of the external environment of the machine room and the IT load in the machine room. Here, each environmental factor includes 5 sample data as an example.
Table one is an exemplary historical environmental factor data table provided by the embodiments of the present application
Figure BDA0003440888820000101
Figure BDA0003440888820000111
The table is an exemplary historical environmental factor data table provided by the embodiment of the application. The historical environmental factor data is normalized to generate a normalized data matrix of the historical environmental factor data, wherein the normalization process can be realized according to the formula (1):
Figure BDA0003440888820000112
wherein, x'ijNormalized data, x, representing the historical data of the jth sample corresponding to the ith environmental factorijRepresenting the historical data of the jth sample corresponding to the ith environmental factor,
Figure BDA0003440888820000113
sample mean, σ, of historical data representing the ith environmental factoriAnd (3) standard deviation of historical data representing the ith environmental factor.
The normalized data matrix of the historical environmental factor data is denoted by Z, and is denoted by Z ═ Z1, Z2..., Z7, and the normalized data matrix Z data is recorded in table two.
Table ii is an exemplary normalized historical environmental factor data table provided in this embodiment of the present application
0.9562 0.9562 -0.239 0.7303 -1.7889 -0.6244 -0.7303
-0.239 -0.239 -0.239 -1.0954 0.4472 1.6881 -0.7303
-0.239 -0.239 0.9562 -1.0954 0.4472 -0.1619 1.0954
-1.4343 -1.4343 -0.239 0.7303 0.4472 -0.0462 1.0954
0.9562 0.9562 -0.239 0.7303 0.4472 -0.8556 -0.7303
After the determination of the normalized data matrix, a correlation coefficient matrix of the normalized data matrix is determined, wherein from the normalized data matrix Z, a correlation coefficient matrix R is calculated (R ═ Rij)p×nThis can be achieved according to equation (2):
Figure BDA0003440888820000114
wherein p denotes the number of columns of the correlation coefficient matrix, n denotes the number of rows of the correlation coefficient matrix, k denotes the data of the kth column of the correlation coefficient matrix, xkiData representing the ith row of the k column of the correlation coefficient matrix,
Figure BDA0003440888820000115
sample mean, σ, representing the ith environmental factoriDenotes the standard deviation, x, of the ith environmental factorkjData representing the jth column and jth row of the correlation coefficient matrix,
Figure BDA0003440888820000116
sample mean, σ, representing the jth environmental factorjRepresents the standard deviation of the jth environmental factor.
After determining the correlation coefficient matrix R, the contribution rate e of each environmental factor is determined by using the correlation coefficient matrix RiAnd cumulative contribution rate Em. The contribution rate e of each environmental factor can be calculated by calculating the characteristic value of R and then using the characteristic value of RiAnd cumulative contribution rate Em。
Illustratively, the feature root λ is calculated from the feature equation | R- λ I | ═ 0iAnd arrange it from large to small: lambda [ alpha ]1≥λ2≥…≥λpWherein I represents an identity matrix, and then determining a contribution rate e of each environmental factor using formula (3) and formula (4), respectivelyiAnd cumulative contribution rate E of the first m environmental factorsm。
Figure BDA0003440888820000121
Figure BDA0003440888820000122
Table three is the contribution rate and the cumulative contribution rate table provided in the first embodiment of the present application
Figure BDA0003440888820000123
The principal component factor numbers in table three represent the above seven environmental factors, respectively, and the eigenvalues, contribution rates, and cumulative contribution rates of the principal component factors are recorded in table three. After determining the contribution rate and the cumulative contribution rate, a principal component factor is determined based on the contribution rate and the cumulative contribution rate. The larger the contribution rate is, the more information the corresponding principal component factor contains, and we refer to the one with the largest contribution rate as the first component, the second largest as the second principal component, and so on. The embodiment of the present application does not limit the specific implementation manner of determining the principal component factor according to the contribution rate and the cumulative contribution rate.
In a possible implementation manner, the method for regulating and controlling an environment of a machine room, provided by the embodiment of the present application, determines a principal component factor according to a contribution rate and an accumulated contribution rate, and includes:
determining the least quantity of the environmental factors as a target quantity when the accumulated contribution rate reaches a preset contribution rate; and determining the environmental factors of the target quantity as principal component factors according to the magnitude sequence of the contribution rates.
For example, 1, 2, 3, 4, 5, and 6 in table three can be used as principal component factors of a computer room to replace the original seven environmental factors, so as to achieve the purpose of reducing the dimensionality of the influencing factors. For example, the method and the device for screening the IT load in the machine room, the temperature of the external environment of the machine room, the humidity of the external environment of the machine room, the wind speed of the external environment of the machine room, the air pressure of the external environment of the machine room and the illumination intensity of the external environment of the machine room are used as independent variable main factors influencing the output of the cold quantity of the constant-temperature air conditioner in the machine room.
In the embodiment of the application, the reliability of the principal component factors is improved by analyzing and processing the historical environmental factor data and determining the principal component factors according to the contribution rate and the accumulated contribution rate of each environmental factor. The historical environmental factor data is analyzed by adopting a principal component analysis method, the principal component factors in the environmental factors are determined, and then the neural network model is constructed by utilizing the principal component factors, so that the accuracy and granularity of the neural network prediction model are improved.
Then, step S101 of the method for controlling a machine room environment provided in the embodiment of the present application may be implemented by constructing a neural network prediction model, and step S202.
Step S202: and constructing a neural network prediction model according to the principal component factors.
After the principal component factor is determined, a neural network prediction model may be constructed according to the principal component factor in the embodiment of the present application, and in a possible implementation manner, the method for regulating and controlling a machine room environment, which is provided by the embodiment of the present application, constructs the neural network prediction model according to the principal component factor, and includes:
constructing a neural network structure of a neural network prediction model; generating a training sample pair, wherein the training sample pair comprises a principal component factor data sample and an environment regulation information sample; and training the neural network structure by using the training sample pair to generate a neural network prediction model.
The neural network structure can adopt an Error Back-Propagation (Error Back-Propagation) neural network structure, and the neural network structure has a multilayer perceptron structure. In a possible implementation manner, fig. 4 is a schematic structural diagram of a neural network structure provided in an embodiment of the present application, and as shown in fig. 4, in the method for regulating and controlling a machine room environment provided in the embodiment of the present application, the neural network structure includes an input layer, a hidden layer, and an output layer.
The input layer is used for inputting the principal component factor data samples; the input of the concealment layer is a first weighted sum of the principal component factor data samples, and the output of the concealment layer is a first transfer function value of the first weighted sum; the input to the output layer is a second weighted sum of the first transfer function values and the output of the output layer is a second transfer function value of the second weighted sum.
As shown in fig. 4, the input layer, the hidden layer and the output layer include a plurality of neurons, and for convenience of introduction, in one possible implementation, fig. 5 is a schematic structural diagram of a neuron provided by an embodiment of the present application, and each neuron has a plurality of input independent variable factors X as shown in fig. 50，X1，…，XnWeighting a plurality of independent variable factors respectively through channels to generate weighted values of the independent variable factors, and inputting the weighted values into the neuron, wherein the independent variable factor X0，X1，…，XnAre respectively Wi0，Wi1，…，Win。
After input to the neuron, the weighted values of the independent variable factors are added and subjected to function conversion to form dependent variable output information YiThen, the dependent variable outputs information YiAnd then the weighted signal is transmitted to other neurons after being weighted by an output channel, and each neuron acts as an adder and an information converter. The weighted value addition formula of the independent variable factor is as follows:
Figure BDA0003440888820000141
Yi＝f(Si) (6)
the function f is taken as sigmoid function:
Figure BDA0003440888820000142
wherein, XkThe k-th argument factor, W, representing the inputikRepresents XkWeighting from input layer to hidden layer node, SiRepresenting the weighted sum of the ith neuron, YiRepresenting the output of the ith neuron.
In the embodiment of the present application, each neuron is described with reference to fig. 5, and a neural network structure provided in the embodiment of the present application includes a plurality of neurons, as shown in fig. 4, in the embodiment of the present application, a principal component factor data sample is input to an input layer, then, a first weighted sum is obtained by performing weighted summation on the principal component factor data sample, and the first weighted sum is input to a hidden layer.
Wherein the first weighted sum of the principal component factor data samples may be calculated by equation (8):
Figure BDA0003440888820000143
wherein the content of the first and second substances,
Figure BDA0003440888820000144
a first weighted sum is represented that is,
Figure BDA0003440888820000145
k-th data sample, W, representing the i-th environmental factorihTo represent
Figure BDA0003440888820000146
Weighting from the input layer to the hidden layer.
After the first weighted sum of the principal component factor data samples is input to the concealment layer, the concealment layer calculates a first transfer function value for the first weighted sum, which can be implemented by equation (9).
Figure BDA0003440888820000147
Wherein the content of the first and second substances,
Figure BDA0003440888820000148
the first transfer function value is represented by,
Figure BDA0003440888820000149
is composed of
Figure BDA00034408888200001410
The transfer function of the embodiment of the present application is shown in equation (7).
The first transfer function value is weighted by the channel to obtain a second weighted sum of the first transfer function value, the second weighted sum is input to the output layer, and the output layer determines a second transfer function value of the second weighted sum.
Wherein the second weighted sum of the first transfer function values may be calculated using equation (10):
Figure BDA0003440888820000151
the second transfer function value of the second weighted sum may be calculated using equation (11):
Figure BDA0003440888820000152
wherein, WhjFor the weighting from the hidden layer to the output layer,
Figure BDA0003440888820000153
the first transfer function value is represented by,
Figure BDA0003440888820000154
which represents the second weighted sum of the weights,
Figure BDA0003440888820000155
which represents the value of the second transfer function,
Figure BDA0003440888820000156
to represent
Figure BDA0003440888820000157
The transfer function of the embodiment of the present application is shown in equation (7).
After the neural network structure of the neural network prediction model is constructed, training sample pairs are generated, wherein the training sample pairs comprise main component factor data samples and environment regulation information samples.
In a possible implementation manner, the principal component factor data samples and the environmental conditioning information samples may be historical principal component factor data and historical environmental conditioning information data of a computer room, and the number of training sample pairs is not limited in the embodiment of the present application.
In a possible implementation manner, the method for regulating and controlling a machine room environment provided in an embodiment of the present application generates a training sample pair, including: determining a principal component factor data sample and an environmental conditioning information sample; and carrying out normalization processing on the principal component factor data sample and the environmental conditioning information sample to generate a training sample pair.
After the principal component factor data samples and the environmental conditioning information samples are determined, the training sample pairs may be normalized to the [0, 1] interval in one possible embodiment. Since the Sigmold function applied in the Matlab calculation program changes very slowly in the [0, 0.1] and [0.9, 1] intervals and easily falls into the saturation region of the neural network, in another possible implementation, the embodiment of the present application normalizes the training sample pair to the [0.1, 0.9] interval.
The problem of data abnormity can be well solved by normalizing the input data of the input layer, and the output data of the output layer needs to be subjected to inverse normalization processing in order to ensure the reliability of the output data. The specific implementation manner is not described in detail in the embodiment of the present application.
In the embodiment of the application, after the principal component factor data samples and the environmental regulation information samples are determined, normalization processing is carried out on the principal component factor data samples and the environmental regulation information samples, so that data abnormity of different principal component factor data samples due to dimension difference is prevented, and certain principal component factor data samples are prevented from being masked due to small numerical values.
After the neural network structure is constructed and the training sample pairs are generated, the neural network structure is trained by utilizing the training sample pairs to generate a neural network prediction model.
In a possible implementation manner, the method for regulating and controlling a machine room environment provided in an embodiment of the present application, which trains a neural network structure by using a training sample pair, includes:
determining a first weighting adjustment quantity and a second weighting adjustment quantity of a neural network structure, wherein the first weighting adjustment quantity is the weighting adjustment quantity from an input layer to a hidden layer, and the second weighting adjustment quantity is the weighting adjustment quantity from the hidden layer to an output layer; inputting the principal component factor data sample into a neural network structure to generate predicted environment regulation information; determining the weighted correction quantity of the first weighted adjustment quantity and the second weighted adjustment quantity by utilizing the predicted environment adjustment information and the environment adjustment information sample; and updating the first weighted adjustment quantity and the second weighted adjustment quantity of the neural network structure by using the weighted correction quantity, and then training the training sample pair by using the updated neural network structure until the error between the predicted environment adjustment information and the environment adjustment information sample meets the preset condition, and finishing the training.
In a possible implementation manner, the first weighted adjustment quantity of the neural network structure and the second weighted adjustment quantity of the neural network structure may be implemented by randomly setting a fraction, then selecting training sample pairs one by one, calculating weighted correction quantities Δ W of the first weighted adjustment quantity and the second weighted adjustment quantity from the output layer to the input layer according to the predicted environment adjustment information and the environment adjustment information samples, respectively adding the weighted correction quantities Δ W of the first weighted adjustment quantity and the second weighted adjustment quantity to the first weighted adjustment quantity and the second weighted adjustment quantity, respectively, updating the neural network structure, then training the training sample pairs by using the updated neural network structure until an error between the predicted environment adjustment information and the environment adjustment information samples meets a preset condition, and ending the training.
Wherein the weighted correction amount of the first weighted adjustment amount can be calculated by equation (12):
Figure BDA0003440888820000171
the weighted correction amount of the second weighted adjustment amount can be calculated by equation (13):
Figure BDA0003440888820000172
wherein: eta is a training rate coefficient, and the embodiment of the application has the advantage of etaThe value of volume is not limiting and in one possible embodiment, η is 0.01. E is an error function, WihTo represent
Figure BDA0003440888820000173
The weighting from the input layer to the hidden layer,
Figure BDA0003440888820000174
which is indicative of a desired output, is,
Figure BDA0003440888820000175
which represents the value of the second transfer function,
Figure BDA0003440888820000176
which represents the second weighted sum of the weights,
Figure BDA0003440888820000177
to represent
Figure BDA0003440888820000178
A transfer function of WhiRepresenting the residual from the concealment layer to the output layer,
Figure BDA0003440888820000179
a first weighted sum is represented that is,
Figure BDA00034408888200001710
is composed of
Figure BDA00034408888200001711
The transfer function of (a) is selected,
Figure BDA00034408888200001712
k-th data sample, W, representing the h-th environmental factorhjFor the weighting from the hidden layer to the output layer,
Figure BDA00034408888200001713
the first transfer function value is indicated.
In one possible embodiment, the error function may be calculated by equation (14):
Figure BDA00034408888200001714
wherein the content of the first and second substances,
Figure BDA00034408888200001715
which is indicative of a desired output, is,
Figure BDA00034408888200001716
representing the second transfer function value.
In the embodiment of the application, the historical environment factor data is analyzed by adopting a principal component analysis method, the principal component factors in the environment factors are determined, and then the neural network model is constructed by utilizing the principal component factors, so that the accuracy and granularity of the neural network prediction model are improved.
The following are embodiments of the apparatus of the present application that may be used to perform embodiments of the method of the present application. For details which are not disclosed in the embodiments of the apparatus of the present application, reference is made to the embodiments of the method of the present application.
Fig. 6 is a schematic structural diagram of a machine room environment control device provided in an embodiment of the present application, where the device may be implemented in a software and/or hardware manner, for example, may be implemented by a server, as shown in fig. 6, where the machine room environment control device provided in the embodiment of the present application may include: a construction module 31, an acquisition module 32, a processing module 33 and a regulation module 34.
And the building module 31 is used for building a neural network prediction model.
And the obtaining module 32 is configured to obtain the environmental factor data of the machine room.
And the processing module 33 is configured to generate environment adjustment information of the machine room according to the environment factor data and the neural network prediction model.
And the regulation and control module 34 is used for regulating and controlling the environment of the machine room according to the environment regulation information.
The apparatus of this embodiment may perform the method embodiment shown in fig. 2, and the technical principle and technical effect are similar to those of the above embodiment, which are not described herein again.
On the basis of the embodiment shown in fig. 6, further, fig. 7 is a schematic structural diagram of a machine room environment regulation and control device provided in another embodiment of the present application, where the device may be implemented in a software and/or hardware manner, for example, may be implemented by a terminal device, as shown in fig. 7, the machine room environment regulation and control device provided in the embodiment of the present application may further include a determining module 35.
And the determining module 35 is configured to analyze the historical environmental factor data of the computer room according to a principal component analysis method, and determine a principal component factor in the environmental factors.
In a possible implementation manner, the determining module 35 of the machine room environment regulation and control device provided in the embodiment of the present application is specifically configured to:
carrying out standardization processing on the historical environmental factor data to generate a standardized data matrix of the historical environmental factor data; determining a correlation coefficient matrix of the normalized data matrix; determining the contribution rate and the accumulated contribution rate of each environmental factor by using a correlation coefficient matrix; and determining the principal component factor according to the contribution rate and the accumulated contribution rate.
In a possible implementation manner, the determining module 35 of the machine room environment regulation and control device provided in the embodiment of the present application is specifically configured to:
determining the least quantity of the environmental factors as a target quantity when the accumulated contribution rate reaches a preset contribution rate; and determining the environmental factors of the target quantity as principal component factors according to the magnitude sequence of the contribution rates.
The building module 31 in the embodiment of the present application is specifically configured to: and constructing a neural network prediction model according to the principal component factors.
In a possible implementation manner, the machine room environment control apparatus provided in the embodiment of the present application, the building module 31, is specifically configured to:
constructing a neural network structure of a neural network prediction model; generating a training sample pair, wherein the training sample pair comprises a principal component factor data sample and an environment regulation information sample; and training the neural network structure by using the training sample pair to generate a neural network prediction model.
In a possible implementation manner, the computer room environment regulation and control device provided by the embodiment of the application, the neural network structure includes an input layer, a hidden layer and an output layer,
the input layer is used for inputting the principal component factor data samples; the input of the concealment layer is a first weighted sum of the principal component factor data samples, and the output of the concealment layer is a first transfer function value of the first weighted sum; the input to the output layer is a second weighted sum of the first transfer function values and the output of the output layer is a second transfer function value of the second weighted sum.
In a possible implementation manner, the machine room environment control apparatus provided in the embodiment of the present application, the building module 31, is specifically configured to:
determining a first weighting adjustment quantity and a second weighting adjustment quantity of a neural network structure, wherein the first weighting adjustment quantity is the weighting adjustment quantity from an input layer to a hidden layer, and the second weighting adjustment quantity is the weighting adjustment quantity from the hidden layer to an output layer; inputting the principal component factor data sample into a neural network structure to generate predicted environment regulation information; determining the weighted correction quantity of the first weighted adjustment quantity and the second weighted adjustment quantity by utilizing the predicted environment adjustment information and the environment adjustment information sample; and updating the first weighted adjustment quantity and the second weighted adjustment quantity of the neural network structure by using the weighted correction quantity, and then training the training sample pair by using the updated neural network structure until the error between the predicted environment adjustment information and the environment adjustment information sample meets the preset condition, and finishing the training.
In a possible implementation manner, the machine room environment control apparatus provided in the embodiment of the present application, the building module 31, is specifically configured to:
determining a principal component factor data sample and an environmental conditioning information sample; and carrying out normalization processing on the principal component factor data sample and the environmental conditioning information sample to generate a training sample pair.
The apparatus of this embodiment may perform the method embodiment shown in fig. 3, and the technical principle and technical effect are similar to those of the above embodiment, which are not described herein again.
The device embodiments provided in the present application are merely schematic, and the module division in fig. 6 and fig. 7 is only one logic function division, and there may be other division ways in actual implementation. For example, multiple modules may be combined or may be integrated into another system. The coupling of the various modules to each other may be through interfaces that are typically electrical communication interfaces, but mechanical or other forms of interfaces are not excluded. Thus, modules described as separate components may or may not be physically separate, may be located in one place, or may be distributed in different locations on the same or different devices.
Fig. 8 is a schematic structural diagram of an electronic device provided in an embodiment of the present application, where the electronic device may be a server, and as shown in fig. 8, the electronic device includes:
a receiver 40, a transmitter 41, a processor 42 and a memory 43 and computer programs; wherein the receiver 40 and the transmitter 41 implement data transmission with other devices, and a computer program is stored in the storage 43 and configured to be executed by the processor 42, and the computer program includes instructions for executing the above-mentioned room environment regulation and control method, the contents and effects of which refer to the method embodiment.
In addition, embodiments of the present application further provide a computer-readable storage medium, in which computer-executable instructions are stored, and when at least one processor of the user equipment executes the computer-executable instructions, the user equipment performs the above-mentioned various possible methods.
Computer-readable media includes both computer storage media and communication media including any medium that facilitates transfer of a computer program from one place to another. A storage media may be any available media that can be accessed by a general purpose or special purpose computer. An exemplary storage medium is coupled to the processor such the processor can read information from, and write information to, the storage medium. Of course, the storage medium may also be integral to the processor. The processor and the storage medium may reside in an ASIC. Additionally, the ASIC may reside in user equipment. Of course, the processor and the storage medium may reside as discrete components in a communication device.
Those of ordinary skill in the art will understand that: all or a portion of the steps of implementing the above-described method embodiments may be performed by hardware associated with program instructions. The program may be stored in a computer-readable storage medium. When executed, the program performs steps comprising the method embodiments described above; and the aforementioned storage medium includes: various media that can store program codes, such as ROM, RAM, magnetic or optical disks.
The embodiment of the present application further provides a computer program product, which includes computer instructions, and the computer instructions, when executed by a processor, implement the steps in the method for regulating and controlling a machine room environment in the foregoing embodiment.
Other embodiments of the present application will be apparent to those skilled in the art from consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of the invention following, in general, the principles of the application and including such departures from the present disclosure as come within known or customary practice within the art to which the invention pertains. It is intended that the specification and examples be considered as exemplary only, with a true scope and spirit of the application being indicated by the following claims.
It will be understood that the present application is not limited to the precise arrangements described above and shown in the drawings and that various modifications and changes may be made without departing from the scope thereof. The scope of the application is limited only by the appended claims.

Claims (13)

1. A machine room environment regulation method is characterized by comprising the following steps:
constructing a neural network prediction model;
acquiring environmental factor data of the machine room;
generating environment regulation information of the machine room according to the environment factor data and the neural network prediction model;
and regulating and controlling the environment of the machine room according to the environment regulation information.
2. The method of claim 1, further comprising, prior to said constructing a neural network predictive model:
analyzing historical environmental factor data of the machine room according to a principal component analysis method to determine principal component factors in the environmental factors;
the building of the neural network prediction model comprises the following steps:
and constructing a neural network prediction model according to the principal component factors.
3. The method of claim 2, wherein constructing a neural network prediction model based on the principal component factors comprises:
constructing a neural network structure of the neural network prediction model;
generating a training sample pair, wherein the training sample pair comprises a principal component factor data sample and an environment regulation information sample;
and training the neural network structure by using the training sample pairs to generate the neural network prediction model.
4. The method of claim 3, wherein the neural network structure comprises an input layer, a hidden layer, and an output layer,
the input layer is used for inputting the principal component factor data samples;
the input of the concealment layer is a first weighted sum of the principal component factor data samples, and the output of the concealment layer is a first transfer function value of the first weighted sum;
the input of the output layer is a second weighted sum of the first transfer function values and the output of the output layer is a second transfer function value of the second weighted sum.
5. The method of claim 4, wherein training the neural network structure using the training sample pairs comprises:
determining a first weighting adjustment amount and a second weighting adjustment amount of the neural network structure, wherein the first weighting adjustment amount is the weighting adjustment amount from the input layer to the hidden layer, and the second weighting adjustment amount is the weighting adjustment amount from the hidden layer to the output layer;
inputting the principal component factor data samples into the neural network structure to generate predicted environmental conditioning information;
determining weighted correction amounts of the first weighted adjustment amount and the second weighted adjustment amount by using the predicted environment adjustment information and the environment adjustment information sample;
and updating the first weighted adjustment quantity and the second weighted adjustment quantity of the neural network structure by using the weighted correction quantity, and then training the training sample pair by using the updated neural network structure until the error between the predicted environment adjustment information and the environment adjustment information sample meets a preset condition, and finishing the training.
6. The method of claim 3, wherein generating training sample pairs comprises:
determining the principal component factor data samples and the environmental conditioning information samples;
and carrying out normalization processing on the principal component factor data sample and the environment regulation information sample to generate the training sample pair.
7. The method according to any one of claims 2 to 6, wherein the analyzing the historical environmental factor data of the computer room according to a principal component analysis method to determine the principal component factor in the environmental factors comprises:
carrying out standardization processing on the historical environmental factor data to generate a standardized data matrix of the historical environmental factor data;
determining a correlation coefficient matrix of the normalized data matrix;
determining a contribution rate and a cumulative contribution rate of each environmental factor by using the correlation coefficient matrix;
and determining the principal component factor according to the contribution rate and the accumulated contribution rate.
8. The method of claim 7, wherein determining the principal component factor based on the contribution rate and the cumulative contribution rate comprises:
determining the least quantity of the environmental factors as a target quantity when the accumulated contribution rate reaches a preset contribution rate;
and determining the environmental factors of the target quantity as the principal component factors according to the magnitude sequence of the contribution rates.
9. A machine room environment conditioning device, comprising:
the building module is used for building a neural network prediction model;
the acquisition module is used for acquiring the environmental factor data of the machine room;
the processing module is used for generating environment adjusting information of the machine room according to the environment factor data and the neural network prediction model;
and the regulation and control module is used for regulating and controlling the environment of the machine room according to the environment regulation information.
10. The apparatus of claim 9, further comprising:
the determining module is used for analyzing the historical environmental factor data of the machine room according to a principal component analysis method and determining a principal component factor in the environmental factors;
the building module is specifically configured to:
and constructing a neural network prediction model according to the principal component factors.
11. An electronic device, comprising: a processor, and a memory communicatively coupled to the processor;
the memory stores computer-executable instructions;
the processor executes computer-executable instructions stored by the memory to implement the method of any of claims 1-8.
12. A computer-readable storage medium having computer-executable instructions stored therein, which when executed by a processor, are configured to implement the method of any one of claims 1-8.
13. A computer program product comprising computer executable instructions which, when executed by a processor, implement the method of any one of claims 1 to 8.
CN202111638846.4A 2021-12-28 2021-12-28 Computer room environment regulation and control method and device, electronic equipment and storage medium Pending CN114219177A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202111638846.4A CN114219177A (en) 2021-12-28 2021-12-28 Computer room environment regulation and control method and device, electronic equipment and storage medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202111638846.4A CN114219177A (en) 2021-12-28 2021-12-28 Computer room environment regulation and control method and device, electronic equipment and storage medium

Publications (1)

Publication Number Publication Date
CN114219177A true CN114219177A (en) 2022-03-22

Family

ID=80706841

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202111638846.4A Pending CN114219177A (en) 2021-12-28 2021-12-28 Computer room environment regulation and control method and device, electronic equipment and storage medium

Country Status (1)

Country Link
CN (1) CN114219177A (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115167590A (en) * 2022-09-08 2022-10-11 浙江省邮电工程建设有限公司 Intelligent temperature and humidity control method for communication machine room based on terminal of Internet of things
CN116193819A (en) * 2023-01-19 2023-05-30 中国长江三峡集团有限公司 Energy-saving control method, system and device for data center machine room and electronic equipment

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115167590A (en) * 2022-09-08 2022-10-11 浙江省邮电工程建设有限公司 Intelligent temperature and humidity control method for communication machine room based on terminal of Internet of things
CN115167590B (en) * 2022-09-08 2023-02-14 浙江省邮电工程建设有限公司 Intelligent temperature and humidity control method for communication machine room based on terminal of Internet of things
CN116193819A (en) * 2023-01-19 2023-05-30 中国长江三峡集团有限公司 Energy-saving control method, system and device for data center machine room and electronic equipment
CN116193819B (en) * 2023-01-19 2024-02-02 中国长江三峡集团有限公司 Energy-saving control method, system and device for data center machine room and electronic equipment

Similar Documents

Publication Publication Date Title
CN109659933B (en) Electric energy quality prediction method for power distribution network with distributed power supply based on deep learning model
CN111539515B (en) Complex equipment maintenance decision method based on fault prediction
CN106874581B (en) Building air conditioner energy consumption prediction method based on BP neural network model
Benardos et al. Optimizing feedforward artificial neural network architecture
Idri et al. Design of radial basis function neural networks for software effort estimation
CN107392368B (en) Meteorological forecast-based office building dynamic heat load combined prediction method
CN113837356B (en) Intelligent sewage treatment prediction method based on fused neural network
CN114219177A (en) Computer room environment regulation and control method and device, electronic equipment and storage medium
CN110826774B (en) Bus load prediction method and device, computer equipment and storage medium
CN110571792A (en) Analysis and evaluation method and system for operation state of power grid regulation and control system
CN109523155B (en) Power grid risk assessment method of Monte Carlo and least square support vector machine
CN111461463A (en) Short-term load prediction method, system and equipment based on TCN-BP
Abdufattokhov et al. Data-driven finite horizon control based on gaussian processes and its application to building climate control
CN111680712A (en) Transformer oil temperature prediction method, device and system based on similar moments in the day
CN114004397A (en) Multi-factor influence considered regional energy consumption situation prediction method and device
CN111339157B (en) Method, system and equipment for calculating and predicting daily operation efficiency of power distribution network
Zong et al. Embedded software fault prediction based on back propagation neural network
CN106067075B (en) Building energy load prediction model building and load prediction method and device
Govender et al. Short-term load forecasting using artificial neural networks and multiple linear regression
CN114266593A (en) Power consumption prediction method based on KPCA and linear regression
Nemchenko et al. Neuro-Fuzzy Model of Development Forecasting and Effective Agrarian Sector Transformations of Ukraine.
Tereshchuk et al. Automated system of fuzzy identification of expert’s competence for assessing the quality of pedagogical phenomena and processes
RU2813682C1 (en) Device for calculating degree of confidence in reconnaissance tool
Ratchagit et al. The m-delay Autoregressive Model with Application
Amalnik et al. Cash flow prediction using artificial neural network and GA-EDA optimization

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication