Research On On-line Monitoring And Fault Diagnosis System Of Motorizedspindlebasedon Web

With the rapid development of modern industry,mechanical equipment is large,complicated and automated.The degree of automation of modern equipment is increasingly improving.While saving manpower and cost at the same time,the modern equipment is also facing the situation of lack of sufficient site supervision,but the equipment structure is gradually becoming larger and more complicated equipment is also facing the situation of lack of sufficient site supervision,,industrial production in the occurrence of mechanical failure is inevitable,not found in time or processing may cause serious consequences,and it is of great significance to explore online monitoring and re-mote fault diagnosis at the technical level.Mechanical equipment malfunction can't first finish the functions,and hides a huge risk,.Equipment failures can not be handled in time,which could cause fire,explosion and other damaging great production safety accidents,which will bring people's life and property safety hidden trouble,so industrial safety requirements effective supervisions in today's large machinery,artificial field regulation actual operating cost is relatively high.The tasks of fault diagnosis are fault detection,fault type judgment and location processing.The development of detection instruments,such as sensors,provides good off-site monitoring means for fault detection,and at the same time,provides strong data support for the acquisition and storage of monitoring data for fault diagnosis.Nc machine is widely used in industrial production and relatively typical mechanical equipment,under the background of the Internet of things,the realization of the industrial Internet is the keynote.In view of the motorized spindle,one of the common components of Nc machine,as an object,this paper carried out the study of the fault diagnosis system based on web,has realized the purpose of the data sharing,the mechanical equipment distributed monitoring and centralized processing.This paper first analyzes the characteristic of motorized spindle,selects characteristic signals that can reflect the characteristics of the component fault signal,uses collection and monitoring equipment such as the transformer for original signal acquisition and processing,through the network transmission protocol,the collected signal and the signal corresponding to the number belongs to mechanical unity is transmitted to the database storage and centralized management,in order to optimize the speed of data processing,this paper designs the real-time data table and historical data table,and a series of table to store the monitoring and database tuning design processing,in addition,the stator current signal is restored by the amplitude recovery method.Carry out the operation of removing the fundamental frequency,and effectively establish the characteristic data table of the fault.Then,combing with artificial intelligence algorithm,BP neural network is used as the model to build the fault diagnosis system in this paper.PSO particle swarm optimization algorithm is used to optimize BP's slow iteration and easy to fall into the minimum when building the system model.Matlab simulation is used to verify itseffectiveness.Finally,the background program based on Java language,and the front-end program based on HTML,and it realizes the real-time online detection,fault diagnosis and derivative system auxiliary functions under the browser-end B/S architecture with the help of web technologies such as echart,easyui,ajax,interceptor,timer and cross-application call matlab.