IoT-Based Signal Enhancement and Compression Method for Efficient Motor Bearing Fault Diagnosis

Abstract

Continuous condition monitoring and fault diagnosis of motor bearings are vital to guarantee motor safety operation and reduce breakdown losses. With numerous Internet of things (IoT) sensors being installed on motors for condition monitoring, data transmission and storage problems have become new challenges. This study designed a signal enhancement and compression (SEC) method and implemented on an IoT platform for motor bearing fault diagnosis. First, vibration signal is acquired from an accelerometer installed on the motor. The bearing signal is demodulated using an online demodulation algorithm. Second, an envelope signal is downsampled and enhanced using a stochastic resonance-based nonlinear filter. The enhanced signal is compressed using an Opus encoder and transmitted to a receiver. Lastly, the received signal is decompressed using the Opus decoder, and the bearing fault type can be recognized. The effectiveness and efficiency of the proposed SEC method are verified on an IoT platform compared with a conventional method. The proposed method improves 3.83 dB of the average signal-to-noise ratio (SNR), and reduces 94.7% of the total time and 94.6% of the dissipative power. The advantages of the proposed SEC method include high output SNR, low power consumption, and compatibility with edge computing. These advantages show potential applications in remote motor fault diagnosis using battery power supply.