Research On Equipment Fault Intelligent Prediction System Based On IIoT And Big Data Analytics Technologies

During the operation of large-scale precision industrial equipment,various state parameters are complicated.When the abnormal phenomenon of the equipment is not obvious and the development process is slow,these gradual and weak abnormal information is often ignored.Operators,including ordinary equipment monitoring systems,are difficult to directly detect equipment abnormalities at the first time,and cannot diagnose,predict,and deal with equipment abnormalities that may cause accidents in a timely manner.Since the faults aren’t predicted and addressed in time during the initial stage,it is easy to lead to subsequent major equipment accidents,inflicted significant enterprise economic losses,and it can even lead to major physical safety accidents.Therefore,the design and implementation of the IIo T(Industrial Internet of Things)and Big Data platforms,the development of intelligent industrial equipment fault prediction system,real-time monitoring of equipment status and equipment potential failure prediction,to improve the reliability,stability,security of industrial equipment,have important significance.It is also one of the important contents of developing intelligent manufacturing and building a manufacturing power.In view of the problems that still need to be improved in precision and stability of various fault prediction methods,this paper designs and implements the intelligent fault prediction system of equipment based on IIo T and Big Data analysis technology.First,use the IIo T technology to collect various data from front-end sensors,and aggregate them into a Big Data platform for sorting,storage and analysis.Secondly,a DFPNN(Device Fault Prediction Neural Network)network model is proposed using deep learning techniques,and the device fault prediction algorithm is implemented.A trend prediction model based on LSTM(Long-Short Term Memory)and a fault diagnosis model based on Res Net(Residual Network)are combined.The state prediction model is used to predict the state change trend of the equipment in the future period,and the fault diagnosis model is used to predict the possible failures and types of failures of the equipment in the future period.The experimental results indicate that the intelligent fault prediction system of equipment based on IIo T and Big Data analysis technology proposed in this paper has high precisions and faster response times.It can effectively assist device operators in the maintenance and management of device,guarantee the correct operations of device,and provide significant technological support to improve the reliability and stability of the device.The specific research content of this paper is as follows:(1)Build an IIo T and Big Data platform.Aiming at the characteristics of industrial equipment operating data,with open source components EMQ X,TDengine and Grafana as the core,an IIo T and Big Data equipment status awareness platform was custom-built,with data acquisition,cleaning,storage,query,visualization and other functions to visualize the current environment and system conditions.Using Node.js to simulate the IIo T environment data acquisition scenario to verify the reliability and stability of the IIo T and Big Data platform built in this paper.(2)A DFPNN fault prediction network model is proposed,and an intelligent device fault prediction algorithm is implemented.The prediction algorithm consists of two parts: the device running state trend prediction algorithm based on LSTM network and the device fault diagnosis classification algorithm based on Res Net network.The public dataset(Paderborn Dataset)is used to train,verify and test the network model,determine the model parameters,and verify the feasibility and correctness of the fault prediction algorithm proposed in this paper.(3)With the GDS6230 scribing machine as the research object,a specific equipment condition monitoring and fault prediction system is designed and constructed by using the DFPNN fault prediction model.The results indicate that the intelligent device fault prediction system proposed in this paper is stability and reliable,it allows for effective monitoring of the equipment,assist the equipment operator to predict the equipment failure.