Expanding the Use of Nuclear Magnetic Resonance (NMR) and Machine Learning for Reservoir Characterization of an Offshore Gas Field – Rock Typing and Capillary Pressure Profiling

One of the key inputs to reservoir characterization work is the understanding of reservoir rock quality and its capillary pressure (Pc) profile(s) to define fluid saturation characteristics and its distribution across the reservoir(s) of interest. This is commonly performed by utilizing core measurement results such as porosity and permeability profiles as well as its corresponding Pc vs water saturation (Sw) profiles. When core data are not available, alternative solution is required to fill the gap in the input data required for reservoir modelling. In this study, we have integrated Logging While Drilling (LWD) Nuclear Magnetic Resonance (NMR) data, the Hydraulic Flow Unit (HFU) methodology and Machine Learning (ML) tool and processes to generate and populate such rock properties on multiple wells for field wide application and integrated reservoir modelling of an offshore gas field in East Malaysia.