On Line Detection And Fault Diagnosis Of Ring Of Ring Spinning Machine

Ring spinning machine is one of the best and the largest used spinning equipment in China.Wire ring as one of the most important part in it,which is easy to be imbalance,rubbing and cracks happening during the operation in a state of high-speed friction circumference without stopping.This state is easy to be overheated and excessive worn.The breakdown will not only affect the quality of spinning products,but also cause a great waste of productive resources,hinder the continuous operation of spinning machine,block the high efficiency of production,and even affect the life of spinning equipment,when the wire ring failed.Different kind of spinning machines match different types of wire wheels in practice,which also means different operating rules.It is very difficult in the data acquisition and fault diagnosis,while taking the high speed of wire travel,vast data volume,the complexity of information and other factors into account.Therefore,a reliable,efficient,comprehensive,universal ring spinning on-line data detection and fault diagnosis system is in urgent needed in the spinning industry,which also could provide scientific production guidance for the production of textile enterprises.In view of the above production requirements,the main contents of this paper are as follows:1)With the restriction of the wide range of wire ring,complicated movement process,fast running speed and huge quantity,it is difficult to obtain the real-time data of all the wire rings in a round,accurate and rapid way.In this paper,a three-tier distributed data acquisition and control system based on CAN(Controller Area Network)bus,which has reliability,timeliness,sensitivity and versatility,is designed based on field research and theory research.After the actual enterprise application shows that the system has a great enterprise value.2)The backward and too simple wire coil fault diagnosis can't guarantee the accuracy and timeliness.In order to solve this problem,the fault diagnosis algorithm is deeply researched based on the characteristics of the wire ring.An intelligent fault diagnosis algorithm based on SVM(Support Vector Machine)is introduced,and a joint time-domain frequency-domain feature extraction method is proposed for feature selection,grid search technique and cross-validation are adopted for parameter selection.Through the construction,analysis and testing of the SVM fault diagnosis model,the result shows that the diagnosis time of our proposed method is very short and the correct rate arrives 88.0%,which can basically solve the problem of wire ring fault diagnosis in small and medium-sized textile enterprises.3)According to the production requirements of textile enterprises,a set of fully functional and friendly software control system is developed based on the data acquisition and fault diagnosismodel subsystem.From the software perspective,by combining with C/S and B/S technology models,the integration and docking of the acquisition and fault diagnosis subsystems are implemented.The preliminary debugging results of spinning enterprise in Shaoxing show that the system basically implement on-line detection and fault diagnosis processing of wire ring in ring spinning machine.