Publications

Safety in the oil and gas industry is a critical concern, particularly in confined spaces where hazardous gases such as methane (CH₄) and hydrogen sulfide (H₂S) pose significant threats. Past incidents involving CH₄ explosions and H₂S toxicity highlight the necessity for real-time gas monitoring and predictive modeling. Conventional gas detection systems, which rely on fixed or portable sensors and manual inspections, often result in delayed responses, increasing safety risks in dynamic industrial environments.

This study introduces a novel approach: an ESP32-based Wireless Sensor Network (WSN) integrated with gas sensors and an AI-driven Extreme Gradient Boosting (XGBoost) model to enable continuous, real-time monitoring of CH₄ and H₂S concentrations. The ESP32, characterized by its low power consumption and built-in Wi-Fi/Bluetooth capabilities, serves as the central processing unit, aggregating and wirelessly transmitting data from multiple distributed sensors. These sensors, precisely calibrated for specific gas concentrations, and facilitate real-time data acquisition essential for predictive analytics.

The XGBoost model improves predictive accuracy by utilizing sensor readings from locations proximal to the gas source to estimate concentrations at a distant sensor before a critical increase occurs. Leveraging gradient-boosted decision trees, the model effectively captures complex correlations in gas dispersion patterns, providing accurate short-term forecasts. This early warning mechanism enables proactive mitigation strategies, reducing the risk of exposure before hazardous thresholds are reached.

Designed for cost-effectiveness, scalability, and ease of deployment, the proposed system demonstrates high applicability in high-risk industrial environments. This paper provides a comprehensive analysis of the technical implementation of the ESP32-based WSN with XGBoost, including sensor calibration, and safety implications. By enabling continuous monitoring and early gas detection, this technology contributes to enhanced worker safety and improved risk management in the oil and gas industry.