A depleted gas field in the Sarawak basin, offshore Malaysia, was selected as a candidate geological site for CO2 storage. The selection and design for the CO2 injection well locations are an important decision making in the business planning, which involved a complicated risk assessment system covering subsurface and surface integration. The objective in this paper is to apply the probabilistic risk assessment method in the optimization of CO2 injection well locations for the CO2 storage development plan.

Risk comes from uncertainty, so the workflow with probabilistic risk assessment (PRA) methodology includes: 1) the risk identification and sensitivity analysis: the definition of uncertainty elements; 2) the risk analysis: the uncertainty ranges and distributions; and 3) Risk evaluation: including the individual element and composite risk evaluation. Similar to risk matrix method, two parameters used: 1) the likelihood (Pi) of occurrence of each consequence, expressed by the percentile of a threshold; and 2) the severity magnitude (Ii) of the possible adverse consequences, expressed by the coefficient of variation. The total risk is the sum of the products of Pi and Ii, displayed on the risk map.

The workflow and methodology were applied in a depleted gas reservoir for CO2 storage. The main elements identified for uncertainty analysis include: 1) structural model including the top of reservoir depth and internal reservoir zonation; 2) reservoir parameters based on porosity and permeability; 3) fluid contacts. Combing all possible cases for each element, a number of scenarios of reservoir models were constructed, which are the foundation of risk evaluation.

Risk score, expressed by probability, was calculated by all elements, generating a composite risk probability map, where there are different risk levels at different locations. Combining with engineering constraints, the CO2 injection well locations were selected by optimization, avoiding the area with high risk.

You can access this article if you purchase or spend a download.