Well Location Selection from a Static Model and Multiple Realisations of a Geomodel Using Productivity-Potential Map Technique

Abstract

Abstract One of major field development decision is choosing the number of wells required to efficiently drain an oil or gas reservoir. It is an interactive process in which various development scenarios are chosen and their performance analysed. Formation geology and zone connectivity have a major impact on the choice of well location since they determine well productivity. The industry's current well placement selection process is time consuming and costly. It requires analysing numerous development options by performing a large number of flow simulations. This paper describes a technique to partially automate this well placement process. It has been found that a new map which ranks the reservoir zones based on their productivity potential can speed-up, and hence reduce the cost, of this decision making process. This map, termed the Productivity Potential Map (PPM), is based on fundamental petroleum engineering principles. It is generated from the numerical reservoir models developed from standard data measured during the exploration and appraisal process. It incorporates both static and dynamic properties (e.g. porosity and saturation respectively). A static and stochastic numerical reservoir model are coupled with the PPM and well locations that maximise production are identified. The technique was tested using flow simulation models of two UK reservoirs by generating PPM and identifying the drilling targets that could deliver the maximum, sustained production potential. The first example uses a static reservoir simulation model of a field that had been production history matched for 18 years. Compared to the development plan implemented many years ago, the PPM map reduced the number of development wells by 15% while increasing the cumulative oil production by 2.6% at 2.5 years since production started. The second example employs multiple realisations developed from exploration well data using PETREL™. Well locations were chosen from a PPM map derived from these multiple realisations. The chosen well locations clearly matched the reservoir geology. Well locations were also chosen from a STOIIP based map. The performance of the STOIIP and PPM based field development were compared - the PPM based well placement consistently performed best. PPM thus reduced the flow simulation effort required, improved the flow forecast and reduced the uncertainty in flow performance. This paper will show that the use of a PPM is a quick and cost-effective technique for analysis of the reservoir production performance and the generation of drilling targets. The technique may also aid reservoir management decisions e.g. water flood front location and selecting the preferred water flood front direction.