Optimization of Fault Diagnosis of Electrical Motors Using Adaptive Control Based on IOT Monitoring System

Document Type : Original Article

Authors

Electrical Engineering Department, Faculty of Engineering, Benha University

Abstract

Induction motors are popular in industry due to their robustness, reliability, and low maintenance. Like all machines, they can fail and cause downtime, production losses, and safety hazards. Early detection and diagnosis of motor faults prevents catastrophic failures, reduces maintenance costs, and improves efficiency. This paper presents the feasibility and effectiveness of using vibra-tion, temperature, and current (VTC) measurements to obtain a comprehensive picture of the motor's condition and predict faults early. Internet of Things (IoT) sensors and adaptive control supervision protect induction motors by detecting and classifying faults in real-time based on experimental data obtained in the lab. This IoT system monitors and diagnoses electrical motor conditions by measuring VTC to predict functional abnormalities. Sensors are connected to a universal, low-cost microcontroller to obtain the required results. Data is stored on a cloud platform and accessed via a web dashboard and a smartphone application. An efficient adaptive control technique using Artificial Neural Network (ANN) learning identifies fault types even in uncertain diagnosis situ-ations. Simulation results demonstrate its effectiveness in diagnosing the target fault type among the three types. Overall, the paper's results prove that the pro-posed method improves the reliability and efficiency of motor systems by providing accurate fault diagnosis. This can result in significant economic and environmental benefits by reducing maintenance costs and preventing cata-strophic failures.

Keywords

Main Subjects

Fayoum University Journal of Engineering
Volume 7, Issue 2
Special Issue, Selected papers from the Third International Conference on Advanced Engineering Technologies for Sustainable Development ICAETSD, held on 21-22 November 2023
2024
Pages 106-119

Files

Share

How to cite

Statistics