This repository houses the code and documentation for a comprehensive IoT-based predictive maintenance system designed for machinery health monitoring and fault diagnostics. The project focuses on developing a low-cost, low-power consumption, and small-size vibration monitoring system using Micro-Electromechanical Systems (MEMS). The system's performance is evaluated through conventional rolling-element bearing fault diagnostics, with a subsequent goal of implementing a supervised deep learning model using Convolutional Neural Network (CNN) for fault classification.
Key Objectives:
- Design a MEMS-based vibration monitoring system for rotating machinery.
- Evaluate system performance through rolling-element bearing fault diagnostics.
- Develop a supervised deep learning model (CNN) for accurate fault classification.
Hardware:
- ADXL1002 MEMS Accelerometer
- ADC 4 Click (AD7175-8 Development Board)
- STM32 Nucleo-L432KC Microcontroller
Contributions:
- MEMS-based hardware design.
- Vibration signal processing algorithms.
- Deep learning model for fault classification.
- Calibration and signal conditioning techniques.
Note: This repository is a work in progress, and contributions or feedback are welcome.