Research On Elevator Brake Fault Diagnosis Method Based On Vibration Feature Extraction

Elevator brake is an important part of elevator traction system.The normal operation of brake is related to the safety of passengers’ life and property.The elevator brake is not only responsible for the stable stop of the elevator during the normal operation of the elevator and preventing the occurrence of accidents such as car sliding on the designated floor,but also one of the important safety devices in case of elevator failure.The normal operation of elevator brake is very important to the life and property safety of elevator passengers.Therefore,it is more important to carry out the research on the condition monitoring of elevator brake in fault diagnosis.Aiming at the problem of fault diagnosis of elevator drum brake,this paper carries out the analysis of working mechanism and kinematics mechanism of drum brake,the feature extraction of brake vibration signal by using vibration signal analysis method,and the pattern recognition and fault diagnosis of signal under fault condition by using support vector machine.This paper deeply analyzes the mechanism of holding brake and releasing brake of elevator drum brake,analyzes the defects of single degree of freedom kinematic model and two degree of freedom model of elevator drum brake,establishes six degree of freedom kinematic model according to the mechanism of elevator drum brake,and deduces the kinematic equation.According to the kinematics equation,the two different fault mechanisms of belt brake operation and insufficient braking force are analyzed,which lays a foundation for the fault diagnosis of elevator brake.Aiming at the selection of the number of modes K and the penalty coefficient alpha in the variational modal decomposition method,a hybrid adaptive function composed of energy index and correlation coefficient is proposed to select the two parameters.The experimental verification of the constructed mixed signal with noise shows that this method can adaptively select the parameters and has a good decomposition effect on the mixed signal with noise.The elevator brake fault test platform is built.By artificially changing the compression amount of the compression spring,two different faults with brake operation and insufficient braking force are created,and their vibration signals are collected.The vibration signals under different working conditions are analyzed in time domain and frequency domain,and the vibration signals with different braking forces are solved by sample entropy.By calculating the decomposed sample entropy of the modal function based on parameter optimization VMD in the selected frequency domain,the fault eigenvalue of elevator brake is successfully extracted.A fault diagnosis model based on sample entropy parameter optimization VMD and squirrel search optimization algorithm support vector machine is proposed,which is verified by the measured elevator brake vibration data.The vibration data under different working conditions are divided into four categories,and the data set is constructed.Finally,the comparative analysis verifies the effectiveness of VMD based on sample entropy and support vector machine model based on squirrel search optimization algorithm.