Research On Fault Diagnosis Of Mine Asynchronous Motor

With the implementation of the Made in China 2025 and the advancement of the construction of smart mines in China,the electromechanical equipments under the mine are getting closer and closer,and they are developing in the direction of automation and intelligence,which puts forward higher requirements for the operation of the mining system.As the power core of the entire coal mine production,whether the asynchronous motors can work reliably in harsh environments will directly affect the stability of the mining system.Carrying out fault diagnosis research on asynchronous motors under the mine is of great significance for discovering early faults of motors,preventing the expansion of faults,and ensuring the safe operation of mining systems.In this paper,firstly,the fault mechanism of the broken rotor bar fault,static air gap eccentric fault,and dynamic air gap eccentric fault is analyzed from the perspective of electromagnetics and derives the fault characteristic frequency of each fault state reflected in the stator current.The ANSYS software is used to model the faults of the explosion-proof motors commonly used in underground mines,and the simulated stator current data was obtained.Through the analysis of the simulated stator current spectrum,the correctness of the fault mechanism analysis is verified,and the data support is provided for the follow-up algorithm and simulation.Next,due to Fourier analysis has certain limitations in extracting local features of non-stationary signals,so the MEEMD method is used to extract the fault characteristics.the MEEMD algorithm is used to decompose the motor stator current signal into a series of IMFs,then IMF components with the most abundant information are selected by cross-correlation criterion and their energy entropy is calculated to construct Fault feature vectors.Then input the feature vectors of the four states including normal motors into the BPNN model for training and fault type recognition.The simulation results show that the MEEMD-BPNN algorithm proposed in this paper is a feasible method for fault diagnosis of mine motors,which realize the accurate recognition of asynchronous motor in normal state,broken rotor bar,air gap eccentricity,with a comprehensive recognition rate of 99%.Compared with the EEMD-BPNN method,MEEMD-BPNN reduces the false components and training iterations,and improves the calculation accuracy by 4.25%.Compared with the MEEMD-SVM,MEEMD-BPNN improves the accuracy by 3.75%Finally,with the STM32F429 chip as the core,the overall design plan of the fault diagnosis system for the mine inverter motor is completed,including the signal acquisition and conditioning circuit,analog-to-digital conversion circuit,communication circuit and other peripheral hardware circuits.The design ideas of software are given for each hardware module.Using the code optimization scheme based on STM32CubeMX software,the embedded porting of BPNN is realized through the AI toolkit,which significantly improve the development efficiency.An experimental platform for a faulty motor is designed and built,and the experimental data is analyzed from multiple perspectives to further verify the fault mechanism and the feasibility of the MEEMD-BPNN algorithm...