Vision-based abnormal state monitoring and fault diagnosis method for digital workshop MES system

Abstract

The invention discloses a vision-based abnormal state monitoring and fault diagnosis method for a digital workshop MES system, and relates to a machine vision-based technology. A plurality of sensorsare arranged in a digital production workshop to monitor an industrial field, and a video collector is used for collecting a bottom industrial control field video in real time. The method comprises two modes: 1, an online analysis mode: the video acquisition device acquiring on-site videos in real time, extracting video data information at the current moment through the video analysis and processing module, integrating production data, monitoring data and video data information, and sending the information to the abnormity monitoring and fault diagnosis module for abnormity monitoring and abnormal event detection; and 2, an offline analysis model: according to the abnormal occurrence time period, querying a corresponding video, extracting video data information of the time period through avideo analysis and processing module, integrating and sending production data, monitoring data and video data information of the time period to an abnormal monitoring and fault diagnosis module, performing fault diagnosis, and analyzing fault reasons.

Description

technical field [0001] The invention relates to an industrial control technology for monitoring and diagnosing abnormal states based on machine vision technology in an MES system, specifically a method for monitoring abnormal states and diagnosing faults based on vision in a digital workshop MES system. Background technique [0002] In the intelligent production process of modern enterprises, Manufacturing Execution System (MES) plays an important role. It is a workshop-oriented management information system between the upper-level plan management system and the lower-level industrial control system. It can provide on-site management subdivision, on-site data collection, electronic kanban management, warehouse material storage, and production task allocation for enterprise production. , Warehouse management, Responsibility traceability, Performance statistical evaluation, Statistical analysis and comprehensive analysis and other functions have greatly improved the management...

AI Technical Summary

Problems solved by technology

One is that the MES system collects the underlying industrial control data, which is mainly the production process status and information data, which is normal data in the production process, and it is difficult to directly reflect the fault or abnormality of the production process, which is not conducive to the detection and analysis of faults or abnormalities

Second, because the MES system mainly serves production, in the underlying industrial control, the sensor layout for fault monitoring or abnormal diagnosis is insufficient, and the extraction, processing and analysis of relevant data are not sufficient, which affects the detection and analysis of faults or abnormalities

In addition, the data mentioned above are the results obtained from on-site detection or measurement through sensors and instruments in the process of industrial control, and the description of the on-site situation is not direct and accurate enough.

Images