Research On Condition Monitoring And Fault Diagnosis System Of Scraper Conveyor

With the application of automatic control technology in mine,the production speed of mine has been improved obviously.The smooth and fast transportation of coal directly affects the production efficiency of fully mechanized mining face.Scraper conveyor,shearer and hydraulic support constitute three kinds of fully mechanized mining machinery,which can undertake materials and transportation workers.Due to the poor working condition of the production face,large load and many shocks in the transportation process,it is easy to fail in the operation process,which seriously reflects the production efficiency and output of the mine.Therefore,it is necessary to carry out real-time remote monitoring of the parts of the scraper conveyor that are prone to failure.This paper studies the structure and working principle of scraper conveyor,combined with the failure statistics and failure causes of scraper conveyor are analyzed,the fault parts of scraper conveyor are easy to appear,the design scheme of scraper conveying and condition monitoring system is establelished by monitoring the running state of motor,reducer and scraper chain.The parameter values to be monitored are collected through Hall sensor,temperature sensor,current sensor and Siemens s7-200 smart.The remote monitoring interface of the monitoring system is establelished by using force control configureuration software,motor reducer temperature and current status interface and alarm recording interface.The monitored parameter values are displayed in real time and stored and exported locally by renting con Figureuration software.Based on the condition monitoring of the scraper conveyor,using a Gaussian probability density function to detect outliers in the data combined with the requirements of fault diagnosis,the neural network is used to diagnose the status parameters of the scraper conveyor.The BP neural network optimized by genetic algorithm is selected for fault diagnosis,the GA-BP model is establelished,and the fault diagnosis process and training model are compiled by Matlab.By comparing the results before and after optimization,it is concluded that the accuracy of fault diagnosis by the optimized neural network is higher,based on the existing model,the data in different scenarios can be fault diagnosed by transfer learning.It provides a theoretical basis for the application of fault diagnosis of scraper conveyor.Test and debug the performance of scraper conveyor monitoring system in laboratory,and the performance of sensor,controller,monitoring software and other system components is tested.The test results show that the monitoring system has good performance and can transmit and display data accurately and quickly in real time,and lays a foundation for realizing unmanned underground.