Internet of Things-Based Induction Motor Diagnosis Using Convolutional Neural Network

Abstract

We experimented analyzing motor vibration with aid of Raspberry Pi when, at that time, the engine vibration was abnormal. The Pi signal is transmitted to a relay by the motor supply disconnection. The control unit, nevertheless, monitors and sends the data to the storage system in good form with proper temperature. A FO-PID controller is utilized to analyze the effects of IM due to harmonic current, vibration, and noise. The induction motor’s response to harmonic and current fluctuations is stabilized by a FO-PID controller. The findings can be displayed on the mobile. The tests were carried out in a static state of vibration condition, and fast Fourier transformation is used to analyze the measured vibration data signals. The results of this model were based on the convolutional neural network (CNN), which considerably monitors early diagnostics of the vibration. With a maximum delay of around 1 s, the controller can forward cloud vibration data. Using the CNN model train to analyze the performance of the classification accuracy, the stored data are collected. This article offers a novel way of building tools for measuring vibration in real time based on the schematic architecture provided by the Python mode.