Abstract
As the QCLNG project has moved from the production ramp up phase to a plateau maintenance phase, well spacing optimisation has been recognised as key concept select decision to improve capital efficiency and reducing UTC for future development wells. This paper focuses on the workflow adopted by the project team to define the optimum well spacing for greenfield areas. The future development wells are geographically widespread and heterogeneous in character. As a result, different subsurface settings, risks and uncertainties are also observed across the scope which can potentially impact the well spacing decision.
The workflow integrates a reservoir property clustering analysis, sector model simulation and Decision Framework Model. The reservoir clustering analysis seeks to partition similar parts of the reservoir using five key subsurface parameters using the K-means clustering method. This enables a manageable number of representative sector models to be built for the well spacing study, using average properties from the defined reservoir clusters. The sector models are further sense-checked and benchmarked against empirical data in similar producing areas. The benefit of using a sector model in this workflow rather than the full field model to run all simulations, is an ability to test significantly more subsurface realisations and well spacing options, as well as addressing the key risks in each of the different catchments. The full field model requires significant simulation run time which makes it challenging to evaluate a wide enough solution space that will address the relevant subsurface risks and uncertainty at play.
The Decision Framework model is used to rapidly screen and compare the different well spacing options based on the defined value drivers. A ranking methodology is then adopted to identify the optimal well spacing option for each reservoir cluster. The workflow is concluded by running a limited number of recommended well spacing options for each catchment through the full field simulation model to test the impact on the mid-term production and understand the full project value trade off. The overall outcome of this workflow is that a wider range of subsurface realisations can be created and efficiently screened in simulation against key value drivers, thus allowing faster and more robust well spacing decisions.
