Machine room equipment remote operation and maintenance management method and system based on Internet of Things technology

Abstract

The invention relates to the technical field of the Internet of Things, and particularly discloses a machine room equipment remote operation and maintenance management method and system based on the Internet of Things technology, and the method specifically comprises the steps: collecting the operation state data of machine room equipment in real time through an Internet of Things sensor network, and obtaining the mean value and standard deviation of the historical operation data of the equipment; transmitting the data to a cloud server through a 5G network; the cloud server performs data cleaning, format conversion and storage, and establishes a unified equipment operation database; performing principal component analysis on the standardized data, extracting main feature vectors, and calculating the contribution rate of each operation parameter; calculating a total weight according to the contribution rate of each operation parameter, mapping the total weight to an original parameter, and respectively calculating a weight corresponding to each operation parameter; calculating a device health degree score based on the device operation data; selecting a logistic regression model, establishing a fault prediction model and calculating a fault probability; and comparing the fault probability with a preset threshold value, and determining whether fault early warning is triggered or not.

Description

Technical Field

[0001] The present invention relates to the technical field of Internet of Things, and in particular to a remote operation and maintenance management method and system for computer room equipment based on the Internet of Things technology. Background Art

[0002] With the rapid development of information technology, the number and types of equipment in computer rooms are increasing. The traditional operation and maintenance management mode of computer room equipment can no longer meet the needs of high efficiency, precision and intelligence. The existing operation and maintenance management methods of computer room equipment mainly have the following problems: First, the information island phenomenon is serious. Computer room equipment usually comes from different manufacturers, and the data format is not unified. It is difficult to achieve data interconnection and sharing, resulting in low efficiency of operation and maintenance management. Secondly, the fault warning capability is insufficient. The traditional operation and maintenance mode mainly relies on manual inspection and post-processing. It lacks real-time monitoring of the equipment operation status and fault warning capabilities, making it difficult to prevent problems before they occur. In addition, the operation and maintenance efficiency is low. The operation and maintenance personnel need to frequently travel back and forth to the computer room site, which is heavy in workload and low in efficiency. It is difficult to meet the operation and maintenance needs of large-scale computer room equipment. There is also a lack of intelligent analysis. The massive equipment operation data lacks effective analysis and utilization, making it difficult to mine the data value and provide support for operation and maintenance decisions. Finally, the remote operation and maintenance capabilities are limited. The existing technology does not support remote operation and maintenance enough, especially in complex network environments, making it difficult to achieve efficient remote monitoring and control. Therefore, there is an urgent need for a high-efficiency operation and maintenance management system that can realize real-time collection and transmission of equipment operation data, break information silos, achieve unified data management, and improve fault warning capabilities. Summary of the invention

[0003] (1) Technical issues to be resolved The purpose of the present invention is to provide a remote operation and maintenance management method and system for computer room equipment based on Internet of Things technology. Through the Internet of Things sensor network and 5G network, the operating status data of the equipment is collected in real time, and efficiently transmitted to the cloud server to solve the real-time problem of data collection and transmission; through data cleaning, format conversion and storage, a unified equipment operation database is established to solve the problems of inconsistent data formats and information islands. Based on principal component analysis and logistic regression models, a fault prediction model is constructed to achieve accurate prediction and early warning of equipment failures, solving the problem of insufficient fault early warning capabilities in traditional operation and maintenance modes; through remote monitoring, automated operation and maintenance and intelligent analysis, the frequency of manual inspections is reduced, the operation and maintenance efficiency is improved, and the problem of low operation and maintenance efficiency is solved.

[0004] (2) Technical solution To achieve the above object, the present invention provides a remote operation and maintenance management method for computer room equipment based on Internet of Things technology, the method comprising the following steps: S1, using the IoT sensor network to collect real-time operating status data of equipment in the computer room, including temperature, CPU utilization, memory usage, and hard disk read and write rates; obtain the mean and standard deviation of equipment operation history data; and transmit the collected data and equipment operation history data to the cloud server through the 5G network; S2, receiving the operation status data of the equipment, the cloud server performs data cleaning, format conversion and storage, and establishes a unified equipment operation database; data cleaning includes removing outliers and filling missing values, and format conversion includes unifying the data of different equipment into a standardized format; principal component analysis is performed on the standardized data to extract the main eigenvectors and calculate the contribution rate of each operation parameter; S3, calculate the total weight according to the contribution rate of each operating parameter, map the total weight to the original parameter, and calculate the weight corresponding to each operating parameter respectively; calculate the equipment health score based on the equipment operation data; select the logistic regression model, establish a fault prediction model and calculate the fault probability; compare the fault probability with the preset threshold to confirm whether to trigger the fault warning and notify the operation and maintenance personnel.

[0005] Furthermore, the cloud server performs data cleaning, format conversion and storage on the operating status data of the receiving device to establish a unified device operation database; data cleaning includes removing outliers and filling missing values, and format conversion includes unifying data of different devices into a standardized format. The method includes: First, the equipment operation data is cleaned to remove outliers and fill missing values ​​with the mean; Secondly, the equipment operation data is standardized to obtain standardized data , and stored in the cloud server, the standardized processing formula is as follows: ; in: For the The current value of the operating parameter, For the The mean of the operating parameters; For the The standard deviation of the operating parameters.

[0006] Furthermore, the method of performing principal component analysis on the standardized data, extracting the main eigenvectors, and calculating the contribution rate of each operating parameter includes: S31, standardized historical data, the standardized data matrix is , whose dimensions are , represents m equipment operating parameters, each operating parameter has n samples, and the standardization formula is: ; in, It is The sample operating parameter values, It is The mean of the operating parameters, It is The standard deviation of the operating parameters; S32, calculate the covariance matrix C, the formula is: ; in, is the standardized data matrix, yes The transposed matrix of , C is a A symmetric matrix, which represents the correlation between the operating parameters; S33, calculate eigenvalues ​​and eigenvectors: The eigenvalues ​​and eigenvectors satisfy: ; in, is the xth eigenvalue, is the xth eigenvector; perform eigenvalue decomposition on the covariance matrix C and obtain the eigenvalue and the corresponding eigenvector ; S34, sorting eigenvalues ​​and eigenvectors: Sort the eigenvalues ​​from large to small and rearrange the corresponding eigenvectors; the sorted eigenvalues ​​and eigenvectors are: ; ; S35, calculate the contribution rate: Each eigenvalue Contribution rate It indicates the proportion of the variance explained by the principal component to the total variance. The calculation formula is: ; Contribution rate The larger it is, the more information the principal component contains; S36, select the principal components: Select the first s principal components so that the cumulative contribution rate reaches the set threshold, and the cumulative contribution rate The calculation formula is: .

[0007] Furthermore, the method of calculating the total weight according to the contribution rate of each operating parameter, mapping the total weight to the original parameter, and respectively calculating the weight corresponding to each operating parameter includes: According to the contribution rate , calculate the total weight , the weight calculation formula is as follows: ; The total weight Mapped to the original parameters, calculate the weights corresponding to temperature T, CPU utilization U, memory usage M, and hard disk read and write rate D respectively: Weight of temperature T: ; in, Indicates The first value of the eigenvector, Indicates The first value of the eigenvector, Indicates The first value of the eigenvector; Weight of CPU utilization U: ; in, Indicates The second value of the eigenvector, Indicates The second value of the eigenvector, Indicates The second value of the eigenvector; The weight of memory usage M: ; in, Indicates The third value of the eigenvector, Indicates The third value of the eigenvector, Indicates The third value of the eigenvector; The weight of the hard disk read and write rate D: ; in, Indicates The fourth value of the eigenvector, Indicates The fourth value of the eigenvector, Indicates The fourth value of the eigenvector.

[0008] Furthermore, the method for calculating the equipment health score based on the equipment operation data includes: Calculate the equipment health score based on equipment operation data , the formula is as follows: ; in, For the The current value of an operating parameter; and Respectively The minimum and maximum values ​​of the operating parameters; For the The weights of the operating parameters satisfy ; Health score The value range is [0,1]. The closer the value is to 1, the healthier the device is.

[0009] Furthermore, the method of selecting a logistic regression model, establishing a fault prediction model and calculating the fault probability; comparing the fault probability with a preset threshold, and confirming whether to trigger a fault warning and notify the operation and maintenance personnel includes: A fault prediction model is established through a logistic regression model of fault probability prediction. The formula of the logistic regression model is: ; in: is the weight vector, , is the bias term, is a linear combination of eigenvectors; To extract feature vectors; Calculating the probability of failure : .

[0010] Based on the same inventive concept, on the other hand, the present invention also provides a remote operation and maintenance management system for computer room equipment based on Internet of Things technology, the system comprising: The data collection and transmission module is used to use the IoT sensor network to collect the operating status data of the equipment in the computer room in real time, including temperature, CPU utilization, memory usage, and hard disk read and write rates; obtain the mean and standard deviation of the equipment operation history data; and transmit the collected data and equipment operation history data to the cloud server through the 5G network; The data processing and analysis module is used to receive the operation status data of the equipment. The cloud server performs data cleaning, format conversion and storage to establish a unified equipment operation database. Data cleaning includes removing outliers and filling missing values. Format conversion includes unifying the data of different equipment into a standardized format. The standardized data is subjected to principal component analysis to extract the main eigenvectors and calculate the contribution rate of each operation parameter. The system prediction and decision-making module is used to calculate the total weight according to the contribution rate of each operating parameter, map the total weight to the original parameter, and calculate the weight corresponding to each operating parameter respectively; calculate the equipment health score based on the equipment operation data; select the logistic regression model, establish the fault prediction model and calculate the fault probability; compare the fault probability with the preset threshold to confirm whether to trigger the fault warning and notify the operation and maintenance personnel.

[0011] (3) Beneficial effects Compared with the prior art, the present invention has the following beneficial effects: 1. Realize real-time collection and transmission of equipment operation data.

[0012] 2. Break down information silos and achieve unified data management.

[0013] 3. Improve fault warning capabilities. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 This is a flowchart of a remote operation and maintenance management method for computer room equipment based on Internet of Things technology according to Embodiment 1 of the present invention; Figure 2 This is a module block diagram of the remote operation and maintenance management system for computer room equipment based on Internet of Things technology in Example 2 of the present invention. DETAILED DESCRIPTION

[0015] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0016] Before giving examples, it is necessary to explain the application scenarios conceived by the present invention. The present invention is applied to remote operation and maintenance management of computer room equipment based on the Internet of Things technology.

[0017] Example 1: Figure 1 As shown, this embodiment provides a remote operation and maintenance management method for computer room equipment based on Internet of Things technology, and the method includes the following steps: S1, using the IoT sensor network to collect real-time operating status data of equipment in the computer room, including temperature, CPU utilization, memory usage, and hard disk read and write rates; obtain the mean and standard deviation of equipment operation history data; and transmit the collected data and equipment operation history data to the cloud server through the 5G network; S2, receiving the operation status data of the equipment, the cloud server performs data cleaning, format conversion and storage, and establishes a unified equipment operation database; data cleaning includes removing outliers and filling missing values, and format conversion includes unifying the data of different equipment into a standardized format; principal component analysis is performed on the standardized data to extract the main eigenvectors and calculate the contribution rate of each operation parameter; S3, calculate the total weight according to the contribution rate of each operating parameter, map the total weight to the original parameter, and calculate the weight corresponding to each operating parameter respectively; calculate the equipment health score based on the equipment operation data; select the logistic regression model, establish a fault prediction model and calculate the fault probability; compare the fault probability with the preset threshold to confirm whether to trigger the fault warning and notify the operation and maintenance personnel.

[0018] For example, there is a server in a computer room, and its operating status needs to be remotely monitored and managed. The main operating parameters of the server include: temperature (T, unit: °C), CPU utilization (U, unit: %), memory usage (M, unit: %), hard disk read and write rate (D, unit: MB / s); The collected data are as follows: the temperature sensor collects the server temperature T=45℃, the CPU utilization sensor collects U=80%, the memory utilization sensor collects M=70%, and the hard disk read and write rate sensor collects D=120MB / s; the data is uploaded to the cloud server through the MQTT protocol; The mean and standard deviation of the equipment operation history data are obtained as follows: temperature: ;CPU utilization: ;Memory usage: ; Hard disk read and write speed: ; Cloud server related setting parameters: Failure probability threshold .

[0019] Furthermore, the cloud server performs data cleaning, format conversion and storage on the operating status data of the receiving device to establish a unified device operation database; data cleaning includes removing outliers and filling missing values, and format conversion includes unifying data of different devices into a standardized format. The method includes: First, the equipment operation data is cleaned to remove outliers and fill missing values ​​with the mean; Secondly, the equipment operation data is standardized to obtain standardized data , and stored in the cloud server, the standardized processing formula is as follows: ; in: For the The current value of the operating parameter, For the The mean of the operating parameters; For the The standard deviation of the operating parameters.

[0020] Exemplarily, first, the equipment operation data is cleaned, and no abnormal values ​​or missing values ​​are found; Secondly, the equipment operation data is standardized according to the standardized processing formula: .

[0021] Furthermore, the method of performing principal component analysis on the standardized data, extracting the main eigenvectors, and calculating the contribution rate of each operating parameter includes: S31, standardized historical data, the standardized data matrix is , whose dimensions are , represents m equipment operating parameters, each operating parameter has n samples, and the standardization formula is: ; in, It is The sample operating parameter values, It is The mean of the operating parameters, It is The standard deviation of the operating parameters; S32, calculate the covariance matrix C, the formula is: ; in, is the standardized data matrix, yes The transposed matrix of , C is a A symmetric matrix, which represents the correlation between the operating parameters; S33, calculate eigenvalues ​​and eigenvectors: The eigenvalues ​​and eigenvectors satisfy: ; in, is the xth eigenvalue, is the xth eigenvector; perform eigenvalue decomposition on the covariance matrix C and obtain the eigenvalue and the corresponding eigenvector ; S34, sorting eigenvalues ​​and eigenvectors: Sort the eigenvalues ​​from large to small and rearrange the corresponding eigenvectors; the sorted eigenvalues ​​and eigenvectors are: ; ; S35, calculate the contribution rate: Each eigenvalue Contribution rate It indicates the proportion of the variance explained by the principal component to the total variance. The calculation formula is: ; Contribution rate The larger it is, the more information the principal component contains; S36, select the principal components: Select the first s principal components so that the cumulative contribution rate reaches the set threshold, and the cumulative contribution rate The calculation formula is: .

[0022] Exemplarily, the calculation process is as follows: S31, standardized historical data, the device has 4 sets of operating parameters (temperature T, CPU utilization U, memory usage M, hard disk read and write rate D), that is, m=4, each set of parameters has 5 samples, that is, n=5, the standardized data matrix as follows: ; Among them, each column represents a parameter (T′, U′, M′, D′); each row represents a sample; S32, calculate the covariance matrix C: ; ; ; ; S33, calculate eigenvalues ​​and eigenvectors: Depend on ; By solving the characteristic equation we get: ,in is the identity matrix; The eigenvalues ​​are calculated numerically by the power iteration method: ; ; ; ; For each eigenvalue , solve the equation , and obtain the corresponding feature vector ; for , substituting into the solution equation to obtain , solve the solution equation and get: ; Solve in sequence ; S34, sort eigenvalues ​​and eigenvectors: Sort the eigenvalues ​​from large to small, and rearrange the corresponding eigenvectors to obtain: ; ; ; ; ; ; ; ; S35, calculate the contribution rate, m=4: Calculate the total variance: ; Calculate the contribution rate of each operating parameter: ; S35, select the principal components: Set the threshold to 90%, s=3, that is, select the first three principal components, and the cumulative contribution rate for: ; Cumulative contribution rate It is 93.3%, which is greater than the set threshold; Select the first three principal components, and the corresponding eigenvectors are: ; ; .

[0023] Furthermore, the method of calculating the total weight according to the contribution rate of each operating parameter, mapping the total weight to the original parameter, and respectively calculating the weight corresponding to each operating parameter includes: According to the contribution rate , calculate the total weight , the weight calculation formula is as follows: ; The total weight Mapped to the original parameters, calculate the weights corresponding to temperature T, CPU utilization U, memory usage M, and hard disk read and write rate D respectively: Weight of temperature T: ; in, Indicates The first value of the eigenvector, Indicates The first value of the eigenvector, Indicates The first value of the eigenvector; Weight of CPU utilization U: ; in, Indicates The second value of the eigenvector, Indicates The second value of the eigenvector, Indicates The second value of the eigenvector; The weight of memory usage M: ; in, Indicates The third value of the eigenvector, Indicates The third value of the eigenvector, Indicates The third value of the eigenvector; The weight of the hard disk read and write rate D: ; in, Indicates The fourth value of the eigenvector, Indicates The fourth value of the eigenvector, Indicates The fourth value of the eigenvector.

[0024] For example, the total weight is calculated : ; Mapped to the original parameters, the total weight Mapping to the original parameters (temperature T, CPU utilization U, memory usage M, hard disk read and write rate D), the specific steps are as follows: Weight of temperature T: ; Weight of CPU utilization U: ; The weight of memory usage M: ; The weight of the hard disk read and write rate D: .

[0025] Furthermore, the method for calculating the equipment health score based on the equipment operation data includes: Calculate the equipment health score based on equipment operation data , the formula is as follows: ; in, For the The current value of an operating parameter; and Respectively The minimum and maximum values ​​of the operating parameters; For the The weights of the operating parameters satisfy ; Health score The value range is [0,1]. The closer the value is to 1, the healthier the device is.

[0026] For example, the health score is calculated according to the equipment health score formula: : ; The health score is =0.6067 (full score 1.0).

[0027] Furthermore, the method of selecting a logistic regression model, establishing a fault prediction model and calculating the fault probability; comparing the fault probability with a preset threshold, and confirming whether to trigger a fault warning and notify the operation and maintenance personnel includes: A fault prediction model is established through a logistic regression model of fault probability prediction. The formula of the logistic regression model is: ; in: is the weight vector, , is the bias term, is a linear combination of eigenvectors; To extract feature vectors; Calculating the probability of failure : .

[0028] Exemplarily, the model parameters are set as follows: weight vector ; Bias term ; Extract feature vector ; Compute linear combinations : ; ; Calculating the probability of failure , substitute the linear combination result into the logistic regression model formula: ; Compare with the preset threshold to determine whether a fault warning needs to be issued: preset fault probability threshold ,because , triggering fault warning and notifying operation and maintenance personnel.

[0029] Embodiment 2: Based on the same inventive concept, Figure 2 As shown, this embodiment also provides a remote operation and maintenance management system for computer room equipment based on Internet of Things technology, and the system includes: The data collection and transmission module is used to use the IoT sensor network to collect the operating status data of the equipment in the computer room in real time, including temperature, CPU utilization, memory usage, and hard disk read and write rates; obtain the mean and standard deviation of the equipment operation history data; and transmit the collected data and equipment operation history data to the cloud server through the 5G network; The data processing and analysis module is used to receive the operation status data of the equipment. The cloud server performs data cleaning, format conversion and storage to establish a unified equipment operation database. Data cleaning includes removing outliers and filling missing values. Format conversion includes unifying the data of different equipment into a standardized format. The standardized data is subjected to principal component analysis to extract the main eigenvectors and calculate the contribution rate of each operation parameter. The system prediction and decision-making module is used to calculate the total weight according to the contribution rate of each operating parameter, map the total weight to the original parameter, and calculate the weight corresponding to each operating parameter respectively; calculate the equipment health score based on the equipment operation data; select the logistic regression model, establish the fault prediction model and calculate the fault probability; compare the fault probability with the preset threshold to confirm whether to trigger the fault warning and notify the operation and maintenance personnel.

[0030] It should be noted that, regarding the system in the above embodiment, the specific manner in which each module performs operations has been described in detail in the embodiment of the method, and will not be elaborated here.

[0031] Finally, it should be noted that: Although the present invention has been described in detail with reference to the aforementioned embodiments, it is still possible for those skilled in the art to modify the technical solutions described in the aforementioned embodiments, or to make equivalent substitutions for some of the technical features therein. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A remote operation and maintenance management method for computer room equipment based on Internet of Things technology, characterized in that: The method comprises: S1, using the IoT sensor network to collect real-time operating status data of equipment in the computer room, including temperature, CPU utilization, memory usage, and hard disk read and write rates; obtain the mean and standard deviation of equipment operation history data; and transmit the collected data and equipment operation history data to the cloud server through the 5G network; S2, receiving the operation status data of the equipment, the cloud server performs data cleaning, format conversion and storage, and establishes a unified equipment operation database; data cleaning includes removing outliers and filling missing values, and format conversion includes unifying the data of different equipment into a standardized format; principal component analysis is performed on the standardized data to extract the main eigenvectors and calculate the contribution rate of each operation parameter; S3, calculate the total weight according to the contribution rate of each operating parameter, map the total weight to the original parameter, and calculate the weight corresponding to each operating parameter respectively; calculate the equipment health score based on the equipment operation data; select the logistic regression model, establish a fault prediction model and calculate the fault probability; compare the fault probability with the preset threshold to confirm whether to trigger the fault warning and notify the operation and maintenance personnel.

2. The remote operation and maintenance management method for computer room equipment based on Internet of Things technology according to claim 1 is characterized in that: The receiving device's operating status data, the cloud server performs data cleaning, format conversion and storage, and establishes a unified device operation database; data cleaning includes removing outliers and filling missing values, and format conversion includes unifying data from different devices into a standardized format. The method includes: First, the equipment operation data is cleaned to remove outliers and fill missing values ​​with the mean; Secondly, the equipment operation data is standardized to obtain standardized data , and stored in the cloud server, the standardized processing formula is as follows: ； in: For the The current value of the operating parameter, For the The mean of the operating parameters; For the The standard deviation of the operating parameters.

3. The remote operation and maintenance management method for computer room equipment based on Internet of Things technology according to claim 2 is characterized in that: The method of performing principal component analysis on the standardized data, extracting the main eigenvectors, and calculating the contribution rate of each operating parameter includes: S31, standardized historical data, the standardized data matrix is , whose dimensions are , represents m equipment operating parameters, each operating parameter has n samples, and the standardization formula is: ； in, It is The sample operating parameter values, It is The mean of the operating parameters, It is The standard deviation of the operating parameters; S32, calculate the covariance matrix C, the formula is: ； in, is the standardized data matrix, yes The transposed matrix of , C is a A symmetric matrix, which represents the correlation between the operating parameters; S33, calculate eigenvalues ​​and eigenvectors: The eigenvalues ​​and eigenvectors satisfy: ； in, is the xth eigenvalue, is the xth eigenvector; perform eigenvalue decomposition on the covariance matrix C and obtain the eigenvalue and the corresponding eigenvector ; S34, sorting eigenvalues ​​and eigenvectors: Sort the eigenvalues ​​from large to small and rearrange the corresponding eigenvectors; the sorted eigenvalues ​​and eigenvectors are: ； ； S35, calculate the contribution rate: Each eigenvalue Contribution rate It indicates the proportion of the variance explained by the principal component to the total variance. The calculation formula is: ； Contribution rate The larger it is, the more information the principal component contains; S36, select the principal components: Select the first s principal components so that the cumulative contribution rate reaches the set threshold, and the cumulative contribution rate The calculation formula is: 。 4. The remote operation and maintenance management method for computer room equipment based on Internet of Things technology according to claim 2 is characterized in that: The method of calculating the total weight according to the contribution rate of each operating parameter, mapping the total weight to the original parameter, and respectively calculating the weight corresponding to each operating parameter includes: According to the contribution rate , calculate the total weight , the weight calculation formula is as follows: ； The total weight Mapped to the original parameters, calculate the weights corresponding to temperature T, CPU utilization U, memory usage M, and hard disk read and write rate D respectively: Weight of temperature T: ； in, Indicates The first value of the eigenvector, Indicates The first value of the eigenvector, Indicates The first value of the eigenvector; Weight of CPU utilization U: ； in, Indicates The second value of the eigenvector, Indicates The second value of the eigenvector, Indicates The second value of the eigenvector; The weight of memory usage M: ； in, Indicates The third value of the eigenvector, Indicates The third value of the eigenvector, Indicates The third value of the eigenvector; The weight of the hard disk read and write rate D: ； in, Indicates The fourth value of the eigenvector, Indicates The fourth value of the eigenvector, Indicates The fourth value of the eigenvector.

5. The remote operation and maintenance management method for computer room equipment based on Internet of Things technology according to claim 3 is characterized in that: The method for calculating the equipment health score based on the equipment operation data includes: Calculate the equipment health score based on equipment operation data , the formula is as follows: ； in, For the The current value of an operating parameter; and Respectively The minimum and maximum values ​​of the operating parameters; For the The weights of the operating parameters satisfy ; Health score The value range is [0,1]. The closer the value is to 1, the healthier the device is.

6. The remote operation and maintenance management method for computer room equipment based on Internet of Things technology according to claim 4 is characterized in that: The described method selects a logistic regression model, establishes a fault prediction model and calculates the fault probability; The method of comparing the failure probability with the preset threshold to determine whether to trigger a failure warning and notify the operation and maintenance personnel includes: A fault prediction model is established through a logistic regression model of fault probability prediction. The formula of the logistic regression model is: ； in: is the weight vector, , is the bias term, is a linear combination of eigenvectors; To extract feature vectors; Calculating the probability of failure : 。 7. The remote operation and maintenance management system for computer room equipment based on Internet of Things technology is characterized by: The system comprises: The data collection and transmission module is used to use the IoT sensor network to collect the operating status data of the equipment in the computer room in real time, including temperature, CPU utilization, memory usage, and hard disk read and write rates; obtain the mean and standard deviation of the equipment operation history data; and transmit the collected data and equipment operation history data to the cloud server through the 5G network; The data processing and analysis module is used to receive the operation status data of the equipment. The cloud server performs data cleaning, format conversion and storage to establish a unified equipment operation database. Data cleaning includes removing outliers and filling missing values. Format conversion includes unifying the data of different equipment into a standardized format. The standardized data is subjected to principal component analysis to extract the main eigenvectors and calculate the contribution rate of each operation parameter. The system prediction and decision-making module is used to calculate the total weight according to the contribution rate of each operating parameter, map the total weight to the original parameter, and calculate the weight corresponding to each operating parameter respectively; calculate the equipment health score based on the equipment operation data; select the logistic regression model, establish the fault prediction model and calculate the fault probability; compare the fault probability with the preset threshold to confirm whether to trigger the fault warning and notify the operation and maintenance personnel.

Images