Diagnostic de Défauts des Machines par Analyse Vibratoire

Abstract

This work presents an intelligent vibration monitoring system for rotating machinery fault diagnosis. A prototype based on two GY-LSM6DS3 sensors and a Raspberry Pi 4 was developed to acquire vibration signals under normal, imbalance, and misalignment conditions. A database of 180 signals was created and analyzed in both time and frequency domains. KNN, SVM, and DNN models were used for fault classification. A web application named VibraTech Monitor was also developed for real-time monitoring and diagnosis. The results confirm the effectiveness of the proposed system for predictive maintenance.