This paper presents a smart gas leakage detection system that integrates Internet of Things (IoT) hardware with machine learning to improve detection accuracy and reduce false alarms. The proposed system uses an MQ-series gas sensor, Arduino Uno microcontroller, GSM module for SMS alerts, and a servo motor for automatic gas valve shutoff. A Random Forest classifier is trained on labeled gas sensor time-series data to distinguish between safe and leakage conditions while mitigating transient noise and environmental interference. Experimental validation demonstrates high detection accuracy (>95%), rapid response time (<2 seconds), and a significantly reduced false alarm rate compared to threshold-based detection approaches.
Vasudevan et al. (Tue,) studied this question.