Supervision system based on equipment operation and maintenance and supervision method thereof

Abstract

The invention discloses a supervision system based on equipment operation and maintenance and a supervision method thereof, the supervision system comprises a basic configuration center, a data acquisition center, a data alarm center, a data prediction center and an equipment classification center, the basic configuration center is used for performing classification configuration according to data information of each basic flow in the equipment operation and maintenance system; the data acquisition center is used for acquiring a supervision process data set in the operation and maintenance system according to the configuration of the basic configuration center; the data alarm center is used for comparing the supervision process data set with the configuration of the basic configuration center and giving an alarm when the configuration is met; the data prediction center is used for predicting whether the equipment fails or not and predicting the service life of the equipment according to the supervision process data set; and the equipment classification center is used for classifying the equipment according to the supervision process data set. The method has the characteristics of quickly integrating data, reducing the operation cost and improving the judgment accuracy.

Description

Technical Field

[0001] The present invention relates to an equipment operation and maintenance supervision system, in particular to a supervision system based on equipment operation and maintenance and a supervision method thereof. Background Art

[0002] The equipment operation and maintenance system usually includes modules such as equipment management, fault management, maintenance management, spare parts management and inspection management, involving equipment asset management process, equipment repair management process, equipment maintenance process, inspection management process, etc. The supervision system is responsible for monitoring, managing and controlling various processes in the equipment operation and maintenance system through data collection, processing and analysis, so as to ensure the compliance, safety and efficiency of the operation and maintenance process, improve the efficiency and accuracy of supervision, and reduce costs.

[0003] Existing equipment operation and maintenance supervision systems generally implement supervision through big data; real-time data generated by different process data sources in the equipment operation and maintenance system is collected through data collection tools, and then stored and managed through local data servers or cloud storage technology. However, large-scale data storage requires a lot of storage resources and technical support, as well as efficient indexing, retrieval mechanisms and professional analysis technologies to quickly access and analyze data, which is costly. In addition, since the data formats and structures of different devices and systems in the equipment operation and maintenance system may be different, the supervision system is complicated to integrate data, and the update frequency and accuracy of different data sources are also inconsistent, which will affect the accuracy of the supervision system's data judgment. Summary of the invention

[0004] The purpose of the present invention is to provide a monitoring system based on equipment operation and maintenance and a monitoring method thereof. The present invention has the characteristics of rapidly integrating data, reducing operating costs, and improving judgment accuracy.

[0005] The technical solution of the present invention: a supervision system based on equipment operation and maintenance, including a basic configuration center, a data acquisition center, a data alarm center, a data prediction center and an equipment classification center, wherein the basic configuration center is used to perform classification configuration according to the data information of each basic process in the equipment operation and maintenance system; the data acquisition center is used to obtain the supervision process data set in the operation and maintenance system according to the configuration of the basic configuration center; the data alarm center is used to compare the supervision process data set with the configuration of the basic configuration center, and to alarm for situations that meet the configuration; the data prediction center is used to predict whether the equipment will fail and the equipment life according to the supervision process data set; the equipment classification center is used to classify the equipment according to the supervision process data set.

[0006] In the aforementioned supervision system based on equipment operation and maintenance, the contents classified and configured in the basic configuration center are transactions, events, supervision points and supervision conditions; the transactions represent the basic processes in the equipment operation and maintenance system, the events represent the specific documents under each transaction, the supervision points represent the supervision content of each transaction, and the supervision conditions represent the abnormal situations of each supervision point.

[0007] In the aforementioned supervision system based on equipment operation and maintenance, the data acquisition center includes a data collection unit and a data processing unit. The data collection unit is used to collect user data matrix and supervision data matrix according to the supervision point to obtain the supervision system data matrix; the data processing unit is used to remove missing values ​​and duplicate values ​​in the supervision system data matrix to obtain a supervision process data set.

[0008] In the aforementioned monitoring system based on equipment operation and maintenance, the data prediction center includes an equipment failure prediction unit and an equipment life prediction unit. The equipment failure prediction unit is used to predict whether the equipment fails or not according to the monitoring process data set by using the prediction formula f(x)=sign(w·x+b), where w is a model parameter representing a weight vector of a feature, b is a bias term of the model, and x is an input feature vector;

[0009] The equipment life prediction unit is used to adopt the regression model y=β0+β1x1+β2x2+…+β n x n , where y is the dependent variable, x1, x2, …, x n are independent variables, β0,β1,…,β n It is the regression coefficient, which indicates the degree of influence of the independent variable on the dependent variable and predicts the equipment life.

[0010] In the aforementioned monitoring system based on equipment operation and maintenance, the equipment classification center is used to use a clustering model based on the monitoring process data set. Among them, Si is the i-th cluster, μi is the mean of the i-th cluster, and the equipment in the equipment operation and maintenance system is classified.

[0011] The above-mentioned supervision method based on the supervision system of equipment operation and maintenance includes the following steps:

[0012] S1. Basic configuration: Classify and configure the data information of each basic process in the equipment operation and maintenance system in the supervision system;

[0013] S2. Data acquisition: According to the configuration of the basic configuration center, obtain the supervision process data set in the operation and maintenance system;

[0014] S3, Data alarm: Compare the supervision process data set with the configuration of the basic configuration center, and issue an alarm for situations that meet the configuration;

[0015] S4. Data prediction: predict equipment failure and equipment life based on the regulatory process data set;

[0016] S5. Equipment classification: Classify the equipment according to the regulatory process data set.

[0017] In the aforementioned supervision method, in step S1, the content of the basic configuration definition is: defining the basic process in the equipment operation and maintenance system as a transaction: defining the specific documents of the transaction as an event, defining the supervision content of each transaction as a supervision point, and defining the abnormal situation of each supervision point as a supervision condition.

[0018] In the aforementioned supervision method, in step S2, data acquisition is specifically as follows: according to the supervision point, the user data set matrix and the supervision data set matrix are multiplied to obtain the supervision system data matrix; missing values ​​and duplicate data in the supervision system data matrix are removed to obtain the supervision process data set.

[0019] In the aforementioned supervision method, in step S4, data prediction specifically includes: 1) Equipment failure prediction: set the prediction formula f(x) = sign(w·x+b), where w is the model parameter, representing the weight vector of the feature, b is the bias term of the model, and x is the input feature vector; each data in the supervision process data set is used as the input feature vector x, and input into the prediction formula to obtain the final state of the event, indicating whether the equipment failure has occurred or not; 2) Equipment life prediction: establish a regression model y = β0+β1x1+β2x2+…+β n x n , where y is the dependent variable, x1, x2, …, x n are independent variables, β0,β1,…,β n It is the regression coefficient, which indicates the influence of the independent variable on the dependent variable. The data in the regulatory process data set are used as independent variables and input into the regression model to predict the life of the equipment.

[0020] In the aforementioned supervision method, in step S5, the equipment classification is specifically as follows: establishing a cluster analysis model Where Si is the i-th cluster, μi is the mean of the i-th cluster; the event equipment data in the supervision process data set is input into the cluster analysis model to classify the equipment and identify equipment groups with similar repair modes or characteristics.

[0021] Compared with the prior art, the present invention has the following beneficial effects:

[0022] The present invention defines four basic configurations of transactions, events, supervision points, and supervision conditions, and carries a variety of algorithms for data collection and processing, so as to monitor the operation of conventional processes in the equipment operation and maintenance system in real time, delete inaccurate, missing, and duplicate data, and obtain data more accurately and focusedly. It can more quickly discover the timeliness and rationality problems existing in the equipment operation and maintenance system, such as whether the processing of each process node in the equipment operation and maintenance system is timely, whether the processing is reasonable, and whether the costs involved are in compliance with the regulations; thereby improving the efficiency, reliability, and safety of equipment management and operation, reducing labor time, and optimizing resource allocation, reducing operating costs, and enhancing the overall performance of equipment.

[0023] The system is applicable to various types of buildings, including office buildings, shopping malls, hospitals, schools, etc. It is also applicable to equipment maintenance companies, property management companies and other institutions to achieve standardization, rationalization and supervision of operation and maintenance processes. DETAILED DESCRIPTION

[0024] The present invention will be further described below in conjunction with the embodiments, but they are not intended to limit the present invention.

[0025] Example:

[0026] A supervision system based on equipment operation and maintenance includes a basic configuration center, a data acquisition center, a data alarm center, a data prediction center and an equipment classification center. The basic configuration center is used to perform classification configuration according to the data information of each basic process in the equipment operation and maintenance system; the data acquisition center is used to obtain the supervision process data set in the operation and maintenance system according to the configuration of the basic configuration center; the data alarm center is used to compare the supervision process data set with the configuration of the basic configuration center and to alarm for situations that meet the configuration; the data prediction center is used to predict whether the equipment will fail and the equipment life according to the supervision process data set; the equipment classification center is used to classify the equipment according to the supervision process data set.

[0027] The classified configuration contents in the basic configuration center are transactions, events, supervision points and supervision conditions. The transactions represent the basic processes in the equipment operation and maintenance system, the events represent the specific documents under each transaction, the supervision points represent the supervision content of each transaction, and the supervision conditions represent the abnormal conditions of each supervision point.

[0028] The data acquisition center includes a data collection unit and a data processing unit. The data collection unit is used to collect user data matrix and supervision data matrix according to supervision points to obtain supervision system data matrix. The data processing unit is used to remove missing values ​​and duplicate values ​​in the supervision system data matrix to obtain supervision process data set.

[0029] The data prediction center includes an equipment failure prediction unit and an equipment life prediction unit. The equipment failure prediction unit is used to predict whether the equipment fails or not according to the equipment operation and maintenance system supervision process data set, using the prediction formula f(x)=sign(w·x+b), where w is a model parameter representing a weight vector of the feature, b is a bias term of the model, and x is an input feature vector;

[0030] The equipment life prediction unit is used to adopt the regression model y=β0+β1x1+β2x2+…+β n x n , where y is the dependent variable, x1, x2, …, x n are independent variables, β0,β1,…,β n It is the regression coefficient, which indicates the degree of influence of the independent variable on the dependent variable and predicts the equipment life.

[0031] The equipment classification center is used to collect data from the regulatory process and use a clustering model Among them, Si is the i-th cluster, μi is the mean of the i-th cluster, and the equipment in the equipment operation and maintenance system is classified.

[0032] A supervision method based on a supervision system for equipment operation and maintenance comprises the following steps:

[0033] S1. Basic configuration: In the supervision system, transactions, events, supervision points and supervision conditions are defined according to the data information of each basic process in the equipment operation and maintenance system;

[0034] Since there are multiple basic processes in the equipment operation and maintenance system, such as equipment asset storage process, repair work order processing process, and maintenance work order processing process, these basic processes are defined as transaction T i , transaction T i From T1, T2, T3...T n composition;

[0035] Define the specific document of each transaction as event E ii , event E ii By E 11 、E 12 、E 21 ……E nm In this structure, n represents a transaction, m represents a specific document, and n and m are positive integers.

[0036] The supervision content of each transaction is defined as the supervision point P ii , supervision point P ii By P 11 , P 12 , P 21 ...P nmwhere n represents the transaction, m represents the specific supervision content, and n and m are positive integers; or the key data of all transactions are defined as the supervision point P i , supervision point P i From P1, P2, P3...P n n is a positive integer. The supervision point can be the judgment of specific field value data or the analysis of data value in complex dimensions.

[0037] The abnormal situation of each supervision point is defined as the supervision condition.

[0038] For example, define the repair work order processing process as transaction T1;

[0039] Define the repair work order numbered BXGD_19900101010101 as event E 11 ;

[0040] Define the acceptance time limit of the repair work order as the supervision point P 11 ;

[0041] Supervision point P 11 The supervision conditions are: the processing time of the acceptance link of the repair work order (ts) - the submission time of the repair work order (bs)>24.

[0042] S2. Data acquisition:

[0043] a. Data collection: According to the supervision point, multiply the user data set matrix and the supervision data set matrix to obtain the supervision system data matrix;

[0044] b. Data exclusion: remove missing values ​​and duplicate data in the regulatory system data matrix to obtain the regulatory process data set;

[0045] S3, supervision condition alarm:

[0046] Compare the supervision process data set with the supervision conditions, and if the supervision conditions are met, an alarm is issued.

[0047] S4. Equipment failure prediction:

[0048] a. Set the prediction formula f(x) = sign(w x + b), where w is the model parameter, representing the weight vector of the feature, b is the bias term (intercept) of the model, and x is the input feature vector;

[0049] b. The data of each event, such as the equipment name, fault description, and repair time, in the transaction in the supervision process data set are respectively used as input feature vectors x and input into the prediction formula to obtain the final state of the event, such as whether the equipment is successfully repaired, the time spent on repair, etc., indicating whether the equipment failure has occurred or not;

[0050] S5 and equipment life prediction:

[0051] a. Establish regression model y=β0+β1x1+β2x2+…+β n x n , where y is the dependent variable (target variable), x1, x2, …, x n are independent variables (features), β0,β1,…,β n is the regression coefficient, which indicates the degree of influence of the independent variable on the dependent variable;

[0052] b. Use the data of events in the regulatory process data set as independent variables and input them into the regression model to predict the life of the equipment.

[0053] For example, for the warranty work order processing process transaction T1, event E 1i The equipment name, repair time, and fault description features are used as independent variables, and the equipment life is used as the dependent variable. A regression model is established and trained to predict the equipment life, optimize the maintenance plan or budget, improve the equipment utilization efficiency and reliability, and enhance the capacity of the supervision system.

[0054] S6. Classification of transaction operation modes:

[0055] a. Establish cluster analysis model Where S i is the i-th cluster, μ i is the mean of the ith cluster;

[0056] b. Input the event equipment data in the supervision process data set into the cluster analysis model, classify the equipment, and identify equipment groups with similar repair modes or characteristics, so as to understand the operation mode of the equipment, optimize the maintenance strategy, or predict possible failure modes in the future. For example, based on the clustering results, personalized maintenance plans can be formulated for different equipment groups to improve the reliability and efficiency of the equipment.

Claims

1. A monitoring system based on equipment operation and maintenance, characterized by: It includes a basic configuration center, a data acquisition center, a data alarm center, a data prediction center and an equipment classification center. The basic configuration center is used to perform classification configuration according to the data information of each basic process in the equipment operation and maintenance system; the data acquisition center is used to obtain the supervision process data set in the operation and maintenance system according to the configuration of the basic configuration center; the data alarm center is used to compare the supervision process data set with the configuration of the basic configuration center and to alarm for situations that meet the configuration; the data prediction center is used to predict whether the equipment will fail and the equipment life according to the supervision process data set; the equipment classification center is used to classify the equipment according to the supervision process data set.

2. A monitoring system based on equipment operation and maintenance according to claim 1, characterized in that: The classified configuration contents in the basic configuration center are transactions, events, supervision points and supervision conditions; the transactions represent the basic processes in the equipment operation and maintenance system, the events represent the specific documents under each transaction, the supervision points represent the supervision content of each transaction, and the supervision conditions represent the abnormal situations of each supervision point.

3. A monitoring system based on equipment operation and maintenance according to claim 2, characterized in that: The data acquisition center includes a data collection unit and a data processing unit. The data collection unit is used to collect user data matrix and supervision data matrix according to supervision points to obtain supervision system data matrix; the data processing unit is used to remove missing values ​​and duplicate values ​​in the supervision system data matrix to obtain supervision process data set.

4. The monitoring system based on equipment operation and maintenance according to claim 1 is characterized in that: The data prediction center includes an equipment failure prediction unit and an equipment life prediction unit. The equipment failure prediction unit is used to predict whether the equipment fails or not according to the supervision process data set by using the prediction formula f(x)=sign(w·x+b), where w is a model parameter representing a weight vector of a feature, b is a bias term of the model, and x is an input feature vector; The equipment life prediction unit is used to adopt the regression model y=β0+β1x1+β2x2+…+β n x n , where y is the dependent variable, x1, x2, …, x n is the independent variable, β0, β1,…, βn are regression coefficients, which indicate the influence of the independent variable on the dependent variable and predict the equipment life.

5. The monitoring system based on equipment operation and maintenance according to claim 1 is characterized in that: The equipment classification center is used to collect data from the regulatory process and adopt a clustering model Where S i is the i-th cluster, μ i is the mean of the ith cluster, which classifies the devices in the equipment operation and maintenance system.

6. The method for supervising a supervision system according to any one of claims 1 to 5, characterized in that: The following steps are involved: S1. Basic configuration: Classify and configure the data information of each basic process in the equipment operation and maintenance system in the supervision system; S2. Data acquisition: According to the configuration of the basic configuration center, obtain the supervision process data set in the operation and maintenance system; S3, Data alarm: Compare the supervision process data set with the configuration of the basic configuration center, and issue an alarm for situations that meet the configuration; S4. Data prediction: predict equipment failure and equipment life based on the regulatory process data set; S5. Equipment classification: Classify the equipment according to the regulatory process data set.

7. The method for supervising a supervision system according to claim 6, characterized in that: In step S1, the basic configuration is defined as follows: defining the basic process in the equipment operation and maintenance system as a transaction: defining the specific documents of the transaction as an event, defining the supervision content of each transaction as a supervision point, and defining the abnormal situation of each supervision point as a supervision condition.

8. The method for supervising a supervision system according to claim 6, characterized in that: In step S2, data acquisition is specifically as follows: according to the supervision point, the user data set matrix and the supervision data set matrix are multiplied to obtain the supervision system data matrix; missing values ​​and duplicate data in the supervision system data matrix are removed to obtain the supervision process data set.

9. The method for supervising a supervision system according to claim 6, characterized in that: In step S4, data prediction specifically includes: 1) Equipment failure prediction: set the prediction formula f(x) = sign(w·x+b), where w is the model parameter, representing the weight vector of the feature, b is the bias term of the model, and x is the input feature vector; each data in the supervision process data set is used as the input feature vector x, and input into the prediction formula to obtain the final state of the event, indicating whether the equipment failure has occurred or not; 2) Equipment life prediction: establish a regression model y = β0+β1x1+β2x2+…+β n x n , where y is the dependent variable, x1, x2, …, x n are independent variables, β0,β1,…,β n It is the regression coefficient, which indicates the influence of the independent variable on the dependent variable. The data in the regulatory process data set are used as independent variables and input into the regression model to predict the life of the equipment.

10. The monitoring method of the monitoring system according to claim 6, characterized in that: In step S5, the device classification is specifically as follows: establishing a cluster analysis model Where Si is the i-th cluster, μi is the mean of the i-th cluster; the event equipment data in the supervision process data set is input into the cluster analysis model to classify the equipment and identify equipment groups with similar repair modes or characteristics.

Images