Research On Equipment Failure Intelligent Warning Technology Based On Digital Twin

With the rapid development of information technology,the traditional manufacturing industry begins to transform to intelligent manufacturing,and the intelligent level of its industrial equipment is constantly improving.With the upgrading of equipment,its internal structure and working environment become more and more complex,and the requirements for equipment condition monitoring become more strict.This situation presents a new challenge to the fault analysis and early warning of industrial equipment.To meet these challenges,enterprises must anticipate device status changes in advance so that maintenance personnel have enough time to maintain or replace components before a fault occurs.Based on the above understanding,this thesis takes the production equipment in the discrete manufacturing workshop as the research object,and establishes the equipment twinning model based on the digital twinning technology.Based on the establishment of equipment twin model,the equipment fault early warning model is created,and the equipment fault early warning method is studied.The early warning model is combined with the twin model to judge the real-time state of equipment more intuitively.On this basis,this thesis designs and develops an intelligent early warning system for equipment based on digital twin,which realizes the monitoring and management of equipment in discrete manufacturing workshop.The main research contents of this thesis are as follows:(1)Research on real-time device state simulation technology based on digital twinning.Firstly,to solve the problem of higher requirements for data collection,the architecture,workflow of data acquisition system based on MQTT(Message Queuing Telemetry Transport)protocol and data transmission method based on Redis are designed.In order to solve the problem of outliers in the historical data of equipment,an abnormal data processing method based on One Class SVM algorithm was proposed.On the basis of data acquisition,the modeling process of virtual equipment is designed according to the components of workshop equipment.Based on Unity 3D and the actual running environment of the device,the construction process of the device virtual scene was designed,and the control script of the driving model was established.Based on the above research,the real-time running state of the device is displayed in the virtual space,which provides an intuitive and realistic observation window for the state monitoring of the device.(2)Research on equipment fault warning method in equipment intelligent warning system.Firstly,the main equipment fault warning methods are summarized.After comprehensive consideration,the data-driven fault warning method is selected,and the AO-SOM neural network is proposed.An improved Aquila Optimizer(AO)algorithm was used to optimize the neurons of SOM(Self-Organizing Map)neural network,and an early warning model based on adaptive threshold was constructed.(3)The device intelligent warning system is developed.According to the actual demand,the design principle of the system is constructed,and the overall structure and specific objectives of the device intelligent warning system are designed.The main functions of the system are designed according to the specific objectives of the system,and the prototype system of the intelligent early warning system of equipment is developed by the previous research.The main functions of the system are shown.