Motor Fault Diagnosis System Based on Single Chip Microcomputer and Artificial Intelligence

An artificial intelligence motor fault diagnosis system based on STM32 single-chip microcomputer and deep learning is designed. After algorithm optimization and a small number of sample learning, the real-time recognition rate of fault diagnosis can reach 96.07%. First, the hardware circuit is designed, the ADXL335 acceleration sensor is used to collect the motor running vibration signal, and the Kalman filter is used to improve the sampling accuracy. Then the Kalman filtered signal is output to the host computer through the serial port, and then converted into the deep learning model for training and recognition using FFT. Finally, experiments are carried out on a self-made motor fault diagnosis simulation experiment platform, and the results show that the system has a better recognition effect.