Research On Status Monitoring System Of Manufacturing Workshop Equipment Based On Deep Learning

This dissertation studies the equipment status monitoring system of manufacturing workshop based on deep learning,and divides it into two processes of data acquisition and status monitoring.Aimed at the characteristics of new and old equipment in the manufacturing workshop,wide range of categories,numerous brands and complex and variable monitoring requirements,it analyzes and studies the data collection and status monitoring analysis of workshop equipment,and designed the corresponding solution.Firstly,the OPC technology is analyzed and researched.The multi-threaded configurable client design is used to establish a data acquisition model based on OPC technology to realize multi-device and multi-tasking status data integration acquisition;Secondly,from the perspective of image processing,the dissertation analyzes the feasible data collection methods for the equipment data that OPC cannot collect,and designs the neural network text to identify the data acquisition,and establishes the data acquisition model combining interface segmentation,text location and recognition;Then,the concept and task of status monitoring are analyzed.Based on the research of common status prediction technology,combined with the characteristics of plant equipment status data,an equipment status data monitoring model based on condition based maintenance mode and long short-term memory(LSTM)timing analysis neural network fusion is proposed;Finally,the operation principle of the equipment status monitoring system in the manufacturing workshop is expounded.The data acquisition model based on OPC and deep learning is combined with the status data monitoring model.The overall architecture of the system is constructed and developed in parallel.The system effect was verified.Research on the manufacturing workshop equipment status monitoring system based on deep learning,which can help employees to obtain equipment status information in time,monitor and process these operational data,determine the current and next-time status of the equipment,and make preventive maintenance diagnosis decisions in order to reduce Spare parts,reduce costs,and increase equipment utilization.