Research On Core Technology Of Intelligent Equipment Internet Of Things Monitoring Service Platform

National intelligent manufacturing system regards intelligent equipment,intelligent services,industrial software,and industrial Internet as the core content of the manufacturing industry transformation.The intelligent equipment Io T monitoring service platform aims to develop industrial software for intelligent equipment in combination with industrial Io T technology,and provides intelligent equipment with full life cycle intelligent services.Among them,the prediction of the remaining useful life of equipment is the core technology in the field of intelligent services.As the joint of the industry,the health status of the bearing is of great significance to the stable operation of the industrial equipment.This paper conducts research on the remaining useful life prediction of intelligent equipment with rolling bearings as the research object,and develops a set of intelligent equipment Internet of Things monitoring service software platform.The main contents of this article include:(1)Develop a set of intelligent equipment Io T monitoring service software platform.For the practicality of intelligent services,according to the relevant standards of fault prediction and health management technology,the main responsibilities of the software platform are divided into five parts: data acquisition,signal processing,condition monitoring,health assessment,remaining life prediction.Software design and development are carried out for the first four parts,and combine with data acquisition equipment to form a system platform integrating software and hardware.In addition,experimental research is carried out on the fifth part.Taking rolling bearing as the key research object,a rotating machinery vibration test bench is designed to provide data material for deeper experimental research.(2)Degradation feature extraction and selection in the full life cycle of rolling bearings.For the problem that a single feature has weak ability to characterize bearing degradation,the whole life cycle vibration signal of rolling bearing is analyzed,and its performance degradation process is described.The time-domain,frequency-domain and time-frequency domain features of the bearing are extracted based on the vibration data,and the sensitivity of the time-frequency domain features extracted based on the EEMD algorithm to the operating conditions of the bearing is analyzed.The extracted features are selected using trend,monotonicity and robustness as feature evaluation indicators.In addition,the multi-domain feature set that can reflect bearing degradation is screened.(3)Research and experimental verification of the remaining useful life prediction algorithm of rolling bearing.Aiming at the prediction accuracy of bearing remaining life,firstly,two models of LSTM and TCN are used to predict the remaining life of rolling bearing taking multi-domain feature set as input,and compared with related academic achievement,it verifies the overall effectiveness of the two prediction models built by combining the multi-domain features extracted in this article.Comparing the evaluation indicators of the two models,TCN has achieved better prediction accuracy,which indicates that TCN can more effectively extract the features contained in the data and is suitable for rolling bearing life prediction.Aiming at the defects of insufficient ability to characterize non-stationary vibration signals with time domain features and frequency domain features,using the time-frequency features extracted by EEMD combined with TCN to conduct life prediction experiments,the prediction accuracy is better than that of using the multi-domain feature set,which shows that the features extracted based on EEMD are sensitive to bearing service characteristics and operating conditions.Moreover,it shows the validity of the EEMD-TCN model proposed in this article for predicting the remaining useful life of rolling bearing.(4)Intelligent equipment Io T monitoring service software platform test and verification.The developed software is deployed in the bearing vibration monitoring instrument of the enterprise.And the vibration monitoring experiment is carried out using deep groove ball bearings.Each module of the developed software platform is verified separately,and the validity and practicability of the software platform are demonstrated by analyzing the experimental results.