Fault Analysis And Diagnosis Of Scraper Conveyor Based On Fuzzy Clustering

With the development of science and technology,the automation of the coal working condition is improved.The scraper conveyor trends to the direction of automation and intelligence.The main function of the scraper conveyor is transshipment and delivery,moreover,with the crusher,shearer and other equipment forms a series system.But the scraper failures frequently,which makes the entire coal mining system often to get into trouble,leads to the huge economic losses,brings security risks to the staff at the same time.In order to make the scraper safe and reliable operation,Fault diagnosis of equipment is researched with concentrating more energy and financial resources.In this thesis,the SGZ1250/2565 coal mine scraper machine is studied,and the significance of fault diagnosis is analyzed.On the basis of studying the current research status of domestic and international,the future development direction and the common methods of fault diagnosis,a fault diagnosis method is put forward which is based on fuzzy clustering analysis.Firstly,the dependence of the overall structure of the scraper conveyor and the fault mechanism of the transmission system is researched based on the mechanism.And the causes of the fault and the result of the fault are explained.Then,the concept of fuzzy clustering algorithm is introduced.The fuzzy C-means clustering is applied in the fault diagnosis of scraper conveyor.The result shows that the algorithm has good classification ability and fast convergence rate in the case that the sample data is less,the difference among samples is large or there exists no interleaved data.When the amount of sample data is large,overlapping with each other,FCM algorithm error rate is relatively higher,and the classification results are not good enough.Finally,aiming at the problems of the FCM algorithm,a genetic simulated annealing algorithm with optimal classification property is proposed and applied to the fault diagnosis of scraper.It is proved that the optimized fuzzy clustering algorithm can solve the clustering problem quickly and effectively.