Real-Time Self-Adaptive & Autonomous Calibration of CO₂ Leak Detection Sensors in the Presence of Impurities Using Wavelet Transform and ML Algorithms

Protecting Qatar’s air and environmental quality is crucial for sustainability, especially given its reliance on oil and gas. This sector significantly contributes to greenhouse gas emissions, particularly CO₂ from the LNG value chain. Achieving net-zero carbon emissions requires innovative, sustainable technologies, making carbon capture and storage (CCS) essential. Captured CO₂ is transported and injected into subsurface reservoirs, often using existing pipelines and wellbores for cost efficiency. However, impurities in CO₂, such as N₂, O₂, H₂, hydrocarbons, CO, SO₂, H₂S, and particulates, can alter its phase envelope and thermodynamics, affecting sensor accuracy in leak detection. AI-based self-calibration offers real-time sensor adjustments using machine learning, multi-sensor fusion, and wavelet transform, enhancing accuracy and reducing maintenance. This research focuses on self-adaptive calibration for CO₂ leak detection sensors in pipe and porous media flows. Experiments will assess CO₂ leakage under varying conditions, supporting Qatar’s CCS efforts in Ras Laffan and Dukhan reservoirs. The project aims to improve CO₂ monitoring, ensuring safer and more efficient Carbon Capture, Utilization, and Storage (CCUS).

Hamad Bin Khalifa University (HBKU)