AI-IoT Enabled Methane Emission Prediction and Carbon Footprint Reduction in Underground Coal Mines: A Case Study

Unofficial AI-generated summary based on the public title and abstract. Not an official translation.

🔬研究者:Demonstrates state-of-the-art AI methods (LSTM, RL) for methane prediction in mining, with high accuracy and real-world deployment potential.

🏢実務担当者:Mining companies can adopt AI-IoT systems to reduce methane emissions, lower energy costs, and improve safety, contributing to ESG reporting and carbon credit generation.

🏛政策担当者:Highlights the role of AI in achieving national climate commitments (India's Net-Zero 2070) and suggests regulatory support for smart mining technologies.

Methane emissions from underground coal mining represent one of the most significant contributors to greenhouse gas (GHG) emissions within the mining sector. As a greenhouse gas, methane possesses a global warming potential approximately 28 times greater than carbon dioxide over a 100-year period, making its monitoring and mitigation essential for achieving climate goals. India, the world's second-largest coal producer, continues to rely heavily on coal for energy security. Coal India Limited (CIL), responsible for more than 80% of India's coal production, faces increasing pressure to reduce fugitive methane emissions while maintaining operational safety and productivity. Recent sustainability disclosures indicate that Scope-1 emissions remain a major component of CIL's carbon footprint, with underground gassy mines contributing substantially through methane leakage. This study proposes an Artificial Intelligence–Internet of Things (AI-IoT) integrated framework for real-time methane monitoring, prediction, and carbon footprint reduction in underground coal mines. The framework combines wireless methane sensing networks, edge computing, cloud analytics, and advanced machine learning algorithms including Long Short-Term Memory (LSTM), Random Forest (RF), and Reinforcement Learning (RL). A case study involving representative gassy mines of Coal India Limited, including Jharia and Raniganj coalfields, demonstrates the practical applicability of the proposed approach. Simulation and field-based analyses indicate that the LSTM model achieved prediction accuracy exceeding 93%, outperforming conventional statistical models. Implementation of AI-enabled ventilation optimization reduced methane-related emissions by approximately 35%, while energy consumption associated with mine ventilation systems decreased by nearly 27%. Additionally, safety performance improved through predictive hazard identification and proactive decision-making. The results highlight the transformative potential of AI-IoT technologies in enabling sustainable mining operations, supporting India's Net-Zero 2070 commitment, enhancing worker safety, and creating pathways for carbon credit generation within the coal mining sector. Keywords: Methane Emissions, Underground Coal Mining, Artificial Intelligence, Internet of Things, LSTM, Carbon Footprint Reduction, Smart Mining, Sustainable Mining, Predictive Analytics

• openalex https://doi.org/10.55041/ijcope.v2i7.044first seen 2026-07-07 04:48:06