Research On Key Technology Of Intellignet Elevator Cloud Data Collection And Diagnosis System

With the increasing shortage of urban land area and the increasing number of high-rise buildings,the number of elevators in China continues to grow,how to effectively monitor,manage and maintain such a huge number of elevators has become an urgent problem to be solved.With the development of the Internet of Things and artificial intelligence,transmitting elevator operation data to the cloud to realize remote real-time monitoring of elevator operation status and fault diagnosis based on artificial intelligence mining can effectively improve the safety and reliability of elevators and enhance elevator management and maintenance efficiency,which has become the development trend of elevators in the future.Based on this,with the support of the Industrial Transformation and Upgrading Project of Jiangsu Province Department of Industry and Information Technology and the Key Industrial Technology Innovation Project of Suzhou City,this paper conducts research on the key technologies of intelligent elevator cloud data collection and fault diagnosis system.First,according to the functional requirements of the system,considering the underlying sensor type and elevator peripheral interface,an ARM-based intelligent elevator cloud data collection device is designed to realize the functions of elevator sensor signal acquisition,data network transmission and elevator operation control.Secondly,the traction machine vibration signal de-noising algorithm is studied to provide an accurate data source for elevator traction machine fault diagnosis.The complex operating electromagnetic environment of elevator makes the collected vibration signal superimposed with so many disturbing noises,for this problem,an improved signal de-noising method based on wavelet transform and EMD(empirical mode decomposition)is proposed in this paper.The original signal is decomposed into a series of IMF(intrinsic mode function)components by EMD,and then the high frequency IMF components containing more noise are processed by wavelet transform,and finally the signal is reconstructed to achieve the de-noising of vibration signal.The experimental tests show that the improved de-noising algorithm based on wavelet transform and EMD improves the de-noising performance of traditional wavelet de-noising and EMD filtering algorithms.Finally,based on the elevator operation data collected and transmitted to the cloud,the study realizes real-time monitoring of elevator operation and fault diagnosis.Firstly,the analog quantities such as elevator operation voltage,current and temperature and switching quantities such as limit signal and overspeed signal collected to the cloud are analyzed to realize real-time monitoring of the elevator core components such as elevator inverter,safety circuit,controller and door machine.Meanwhile,by extracting the time-domain characteristics of vibration signals,the traction machine fault diagnosis algorithm based on convolutional neural network is adopted to effectively improve the safety and reliability of elevator.The intelligent elevator cloud data collection and diagnosis system developed in this dissertation has been deployed in Wujiang District,Suzhou,and passed the inspection and testing of the National Center for Quality Supervision and Inspection of Software Products,and the system has achieved the intended design goal and significantly improved the intelligent level of elevator supervision.