Vehicle Engine Health Monitoring Using Deep Learning

This project presents the design and implementation of an intelligent vehicle engine health monitoring system leveraging deep learning techniques. The system continuously collects multi-modal sensor data — including vibration, temperature, oil pressure, RPM, coolant levels, and exhaust emissions — from an on-board diagnostic (OBD-II) interface and a network of IoT sensors mounted on the engine. The acquired data is pre-processed and fed into a Convolutional Neural Network (CNN) and Long Short-Term Memory (LSTM) hybrid deep learning model, which performs real-time anomaly detection, fault classification, and remaining useful life (RUL) estimation of critical engine components. The system is capable of identifying a wide range of engine faults including misfires, overheating, oil degradation, turbocharger failures, and valve timing irregularities with an accuracy exceeding 96%. Results are transmitted via Wi-Fi to a cloud dashboard and a mobile application, enabling remote monitoring by vehicle owners and fleet managers. The proposed system significantly reduces unplanned downtime, enhances vehicle safety, and lowers maintenance costs by enabling condition-based, predictive maintenance — a major advancement over traditional scheduled maintenance approaches.