Search

CN-121979941-A - Industrial equipment intelligent monitoring method and system based on multi-source data fusion

CN121979941ACN 121979941 ACN121979941 ACN 121979941ACN-121979941-A

Abstract

The invention provides an intelligent monitoring method and system for industrial equipment based on multi-source data fusion, which comprise the steps of obtaining multi-source heterogeneous data of an industrial site, carrying out real-time cleaning and anomaly detection processing on the multi-source heterogeneous data to obtain standardized processing data, generating a data access strategy for a user role based on a role access control model in combination with field attributes of the standardized processing data, screening target data fields according to the data access strategy, analyzing the target data fields through a visual adaptation algorithm, matching the analyzed target data fields with a preset target chart template to generate self-adaptive visual data, and pushing the self-adaptive visual data to a Web display terminal through a remote transmission mechanism based on an edge cloud collaborative framework.

Inventors

  • KANG HAILIN
  • Tang dejia

Assignees

  • 大连海通系统集成有限公司

Dates

Publication Date
20260505
Application Date
20260123

Claims (10)

  1. 1. An intelligent monitoring method for industrial equipment based on multi-source data fusion is characterized by comprising the following steps: acquiring multi-source heterogeneous data of an industrial field; Carrying out real-time cleaning and abnormality detection processing on the multi-source heterogeneous data to obtain standardized processing data; Based on a preset role access control model, generating a data access strategy for a user role acquired in advance by combining field attributes of the standardized processing data; Screening target data fields from the standardized processing data according to the data access strategy, analyzing the target data fields through a visual adaptation algorithm, matching the analyzed target data fields with a preset target chart template, and generating self-adaptive visual data; Pushing the self-adaptive visual data to a Web display terminal through a remote transmission mechanism based on an edge cloud cooperative architecture; the visual adaptation algorithm is built through automatic field identification and force-directed layout optimization, and the remote transmission mechanism is cooperatively realized through edge node caching and WebSocket long connection.
  2. 2. The method of claim 1, wherein the multi-source heterogeneous data comprises one or more of environmental parameter data, security parameter data, and device operational parameter data; The environmental parameter data comprises temperature and humidity data and water immersion state data; the safety parameter data includes oxygen concentration data; The equipment operation parameter data comprise oxygenerator current data and oxygenerator vibration data; The multi-source heterogeneous data is collected through an MQTT/HTTP protocol and supports TCP/IP, 4G and 5G, wiFi multi-network transmission modes.
  3. 3. The method of claim 2, wherein performing real-time cleaning and anomaly detection processing on the multi-source heterogeneous data to obtain normalized processed data comprises: Inputting the multi-source heterogeneous data into a preset streaming computing framework, and filtering noise information in the data to obtain preliminary cleaning data; Performing cross-protocol compatible processing on the preliminary cleaning data based on an OPC-UA protocol to obtain preliminary cleaning data after unified modeling; Inputting the unified modeled preliminary cleaning data into a preset machine learning model, and performing fault prediction on the equipment running state in the industrial field to obtain abnormal data; Performing standardized conversion on normal data in the unified modeled preliminary cleaning data, and generating standardized processing data containing normal standardized data and abnormal labeling data by combining the identification information of the abnormal data; The streaming computing framework is APACHE FLINK framework, and the machine learning model is LSTM model.
  4. 4. The method of claim 1, wherein the generating a data access policy for a pre-acquired user role based on a preset role access control model in combination with field attributes of the standardized processing data comprises: Defining a user role hierarchy based on a preset role access control model, wherein the user role hierarchy comprises an administrator role, an operation and maintenance role and a visitor role; configuring a corresponding data visible range and operation authority for each user role level, and generating a role authority configuration table; Inquiring a role authority configuration table of a user role based on a pre-acquired current logged-in user role, and determining a data visible range and an operation authority corresponding to the current logged-in user role by combining field attributes of the standardized processing data; And generating a data access strategy according to the data visible range and the operation authority.
  5. 5. The method of claim 1, wherein the parsing the target data field by the visualization adaptation algorithm matches the parsed target data field with a preset target chart template, and generating the adaptive visualization data comprises: Analyzing the name and the data type of the target data field through an automatic field identification function in the visual adaptation algorithm, and establishing a mapping relation between the field and the data type; Matching the field with a data type mapping relation to a corresponding target chart template based on a preset field chart type mapping rule; Dynamically arranging the matched target chart templates in a layout mode through a force guide layout optimization function in the visual adaptation algorithm; Filling the content of a corresponding target data field in the standardized processing data into a corresponding target chart template to generate self-adaptive visual data; The target chart template comprises one or more of a histogram template, a thermodynamic diagram template, a graph template and a dashboard template.
  6. 6. The method of claim 1, wherein pushing the adaptive visualization data to a Web presentation terminal via a remote transport mechanism based on an edge cloud co-architecture comprises: An edge server is deployed in a local machine room of the industrial field, and the standardized processing data of high-frequency access is cached and stored to generate edge cache data; Judging whether original data corresponding to the self-adaptive visual data exist in the edge cache data or not: if the self-adaptive visual data exists, acquiring corresponding edge cache data from the edge server, and generating push data by combining the chart configuration information of the self-adaptive visual data; If the self-adaptive visual data does not exist, corresponding original data is obtained from the cloud database, and push data is generated by combining chart configuration information of the self-adaptive visual data; And transmitting the push data to the Web display terminal through the WebSocket long connection in the remote transmission mechanism.
  7. 7. The method of claim 6, wherein the cloud database employs a hybrid architecture of a time series database and a relational database; The time sequence database is InfluxDB; the relational database is MySQL; And the Web display terminal renders the push data through a WebGL-based 3D large screen rendering engine.
  8. 8. An industrial equipment intelligent monitoring system based on multisource data fusion, which is characterized by comprising: the data acquisition module is used for acquiring multi-source heterogeneous data of an industrial field; the data processing module is used for carrying out real-time cleaning and abnormality detection processing on the multi-source heterogeneous data to obtain standardized processing data; The permission control module is used for generating a data access strategy for a user role acquired in advance based on a preset role access control model and in combination with the field attribute of the standardized processing data; The visual adaptation module is used for screening target data fields from the standardized processing data according to the data access strategy, analyzing the target data fields through a visual adaptation algorithm, matching the analyzed target data fields with a preset target chart template and generating self-adaptive visual data; the display pushing module is used for pushing the self-adaptive visual data to the Web display terminal through a remote transmission mechanism based on the edge cloud collaborative architecture; the visual adaptation algorithm is built through automatic field identification and force-directed layout optimization, and the remote transmission mechanism is cooperatively realized through edge node caching and WebSocket long connection.
  9. 9. The system of claim 8, wherein the multi-source heterogeneous data comprises one or more of environmental parameter data, security parameter data, and device operational parameter data; The environmental parameter data comprises temperature and humidity data and water immersion state data; the safety parameter data includes oxygen concentration data; The equipment operation parameter data comprise oxygenerator current data and oxygenerator vibration data; The multi-source heterogeneous data is collected through an MQTT/HTTP protocol and supports TCP/IP, 4G and 5G, wiFi multi-network transmission modes.
  10. 10. The system of claim 9, wherein the data processing module comprises: the data cleaning sub-module is used for inputting the multi-source heterogeneous data into a preset stream computing frame, and filtering noise information in the data to obtain primary cleaning data; the protocol compatibility sub-module is used for carrying out cross-protocol compatibility processing on the preliminary cleaning data based on an OPC-UA protocol to obtain preliminary cleaning data after unified modeling; The abnormality detection sub-module is used for inputting the unified modeled preliminary cleaning data into a preset machine learning model, and carrying out fault prediction on the equipment running state in the industrial field to obtain abnormal data; The standardized sub-module is used for carrying out standardized conversion on normal data in the primary cleaning data after unified modeling, and generating standardized processing data comprising normal standardized data and abnormal labeling data by combining the identification information of the abnormal data; The streaming computing framework is APACHE FLINK framework, and the machine learning model is LSTM model.

Description

Industrial equipment intelligent monitoring method and system based on multi-source data fusion Technical Field The invention relates to the technical field of industrial Internet of things and data visualization, in particular to an intelligent industrial equipment monitoring method and system based on multi-source data fusion. Background At present, with the rapid development of industrial Internet of things and data visualization technology, intelligent monitoring of industrial equipment has become a core support for guaranteeing efficient and safe operation of industrial production, and is widely applied to scenes such as industrial automation, energy management and the like. The types of parameters to be monitored in the industrial field are various, the parameters cover multiple dimensions of environment, safety, equipment operation and the like, and strict requirements are provided for multi-source data integration, dynamic authority adaptation, visual accurate transmission and remote real-time response of a monitoring system. However, the existing industrial monitoring system still has various technical bottlenecks, the actual application requirements are difficult to meet, the problem of functional singleness in the existing monitoring is generally solved, most of the existing monitoring only supports single type parameter monitoring, the unified fusion and processing capacity of multi-source heterogeneous data such as environment parameters, safety parameters and equipment operation parameters is lacked, the monitoring dimension is incomplete, the data value mining is insufficient, an authority management mechanism is stiff, a user interface and data display content are fixed, the visible range and operation authority of data cannot be dynamically adjusted according to the actual requirements of different user roles such as an administrator, an operation and maintenance member and a visitor, the data safety risk exists, the adaptation of different use scenes is difficult, the visualization efficiency is low, a chart is generated and depends on a preset template, the attribute and type of a data field cannot be automatically analyzed, the adaptive display form is matched, the multiple charts are fixed in layout, visual confusion is easily caused due to overload of the data, the information transmission efficiency is affected, the remote access is obviously delayed, the data transmission and rendering mechanism is not optimized in pertinently, the high-frequency access data excessively depends on a central server, the load is excessively high, the large-screen display card is caused, the data lag is difficult to meet the actual requirements of different, the data update, the problem of the real-time is difficult to be solved, the requirements of the real-time is not required to be improved, the monitoring system is required to be rapidly be improved, and the real-time is required to be rapidly is required to be changed, and the monitoring system is required to be flexibly be improved, and is required to be changed, and is required to be rapidly to be developed. These problems are mutually overlapped, so that the flexibility, accuracy, instantaneity and safety of the existing industrial monitoring system are insufficient, the application effect of the existing industrial monitoring system in complex industrial scenes is severely restricted, and the existing industrial monitoring system becomes a core technical challenge to be solved in the field of intelligent monitoring of current industrial equipment. Disclosure of Invention In order to solve the technical problems of insufficient multi-source heterogeneous data fusion capability, stiff authority management, poor visualization adaptability and remote access delay of the traditional industrial monitoring system, the invention provides an intelligent industrial equipment monitoring method based on multi-source data fusion, which comprises the following steps: acquiring multi-source heterogeneous data of an industrial field; Carrying out real-time cleaning and abnormality detection processing on the multi-source heterogeneous data to obtain standardized processing data; Based on a preset role access control model, generating a data access strategy for a user role acquired in advance by combining field attributes of the standardized processing data; Screening target data fields from the standardized processing data according to the data access strategy, analyzing the target data fields through a visual adaptation algorithm, matching the analyzed target data fields with a preset target chart template, and generating self-adaptive visual data; Pushing the self-adaptive visual data to a Web display terminal through a remote transmission mechanism based on an edge cloud cooperative architecture; the visual adaptation algorithm is built through automatic field identification and force-directed layout optimization, and the remote transmission mechanism is co