Research On Intelligent Fault Diagnosis System For Elevator Control Cabinet

With the development of the construction industry and the continuous improvement of people’s living standards,the demand for elevators is also growing.As an important part of the elevator,the elevator control cabinet will have certain faults during the use,so it is necessary to carry out regular inspection and maintenance of the elevator control cabinet.The development of modern power electronics technology and computer technology has made the composition and structure of electrical equipment more and more complicated.Once equipment problems occur,ordinary maintenance personnel cannot.repair and solve them in time,which poses a huge threat to people’s lives.With the continuous improvement of industrial automation level,people have higher and higher requirements for the intelligentization of electrical equipment.Therefore,it has important practical significance for the intelligent research of elevator control cabinet fault diagnosis.This paper is based on understanding the structure of the elevator control cabinet and its various faults and failure mechanisms,aiming at the problems of relay malfunction caused by high harmonics and the difficulty of distinguishing transformer temperature rise from electrical component fault itself,mapping phenomenon of high-order harmonics based on transformer vibration signal,a joint fault diagnosis system based on Lab VIEW framework and MATLAB algorithm is designed.The system takes the vibration,temperature,current and voltage signals of elevator control cabinet as the research object.Firstly,the time-frequency analysis of vibration signals collected by vibration sensors on transformers is carried out by empirical mode wavelet packet transform method.Then the energy entropy is reconstructed and used as input vector to train the BP neural network model based on thought evolutionary algorithm in MATLAB.Finally,the final fault diagnosis system of elevator control cabinet is constructed by cooperating with temperature,current and voltage threshold model.This method solves the problem that traditional wavelet transform lacks self-adaptability,improves the phenomena of endpoint effect and mode aliasing in empirical mode decomposition,more accurately reflects the characteristics of vibration signals,and combines the strong non-linear fitting ability of neural network with the advantages of local convergence accuracy and speed of thought evolutionary algorithm,accurate classification results are obtained.Finally,the system is verified by experiments,and the fault diagnosis model of elevator control cabinet is built through data samples.Then,the model is applied to the diagnosis system.Experiments show that this method has higher accuracy and faster speed than traditional fault diagnosis methods.This method ensures the self-diagnosis and fault early warning of the elevator control cabinet in use.Once the hidden trouble or equpment failure is found,the elevator control cabinet automatically detects the cause of the fault,reduces the time of equipment fault detection,locates the fault quickly and accurately for the elevator control cabinet,improves the efficiency of fault treatment and avoids the occurrence of major hidden trouble.