Research Of Fault Diagnosis Of Automatic Car Washing Machine Based On Data Driven

With the increase of domestic car ownership year by year,the market demand for automatic unmanned car washing machine is growing.Although the automatic car wash machine brings us convenience,it also faces great challenges.Automatic car washing machine works for long periods of time in humid and stained environment,and equipment failure occurs from time to time.With large quantities of equipment put into use,due to unmanned and large laying locations across domains,faulty equipment cannot be diagnosed and repaired in time,which seriously affects user experience and operating income.The traditional fault diagnosis method,which relies on electrical engineers to troubleshoot faults one by one,has been unable to meet the demand of high-efficiency and high-accuracy fault diagnosis for large quantities and distributed equipment.In this thesis,the fault characteristic information of equipment operation data is deeply excavated,the fault diagnosis algorithm based on two-dimensional singular value decomposition(2DSVD)and k-nearest neighbor(KNN)is proposed,and the algorithm is applied to practice,and the fault diagnosis system of automatic car wash machine based on data-driven is designed and implemented.The main research contents of this paper are as follows:(1)Fault analysis and sample set design of automatic car washing machine.On the basis of introducing the composition and working principle of automatic car wash equipment,this paper in-depth analysis the fault characteristics of sensor,motor and user misoperation,summarizes the relationship between fault type and partial fault signs,four typical fault types,including the broken signal line of the encoder,the spotty lens of the reflection sensor,the broken connector of the motor and the misoperation of parking,are determined as the fault types studied in this paper.Three types of equipment operation data,including sensor signal value,controller output state value and process attitude information,which are closely related to the occurrence of faults,are selected to form the fault diagnosis sample set of automatic car washing machine.Finally,the sample acquisition method is introduced.(2)Establish and analyze the fault diagnosis model of automatic car washing machine based on 2DSVD+KNN.Aiming at the problem of insufficient feature extraction from fault diagnosis samples,a fault diagnosis algorithm based on 2DSVD feature extraction and KNN classification is proposed.In order to verify the advantages of this algorithm in automatic car washing machine fault diagnosis,three fault diagnosis models based on KNN,SVD+KNN and 2DSVD+KNN has been built respectively,and comparative experimental analysis was carried out.The experimental results show that the accuracy of fault diagnosis based on 2DSVD+KNN model has obvious advantages.(3)The design and implementation of automatic car washing machine fault diagnosis system which is based on data driven.Aiming at the shortcoming of traditional stand-alone equipment,such as low fault diagnosis efficiency,small amount of historical data,which are unable to meet the needs of large quantities of automatic management,The 2DSVD+KNN fault diagnosis model is integrated to design the fault diagnosis system of automatic car wash machine based on data driven.The system has the functions of remote real-time monitoring,equipment history status query and cloud fault diagnosis.At the same time,the cloud can share a large number of equipment sample data,which improves the fault diagnosis performance of the overall system,and finally realizes the research goal of efficient fault diagnosis of data-driven automatic car wash machine.