Advancing Reservoir Simulation Capabilities for Tight Gas Reservoirs

Abstract

Abstract It is critical to identify optimum well spacing to develop cost-effective full field development plans for tight gas reservoirs. Accurate prediction of well performance is a major challenge that arises during the development of such reservoirs. Understanding well performance is needed for both well design and depletion planning. Almost all wells in tight gas fields are hydraulically fractured. We have investigated methods of simulating hydraulic fractures using coarse grids since explicitly gridding these fractures can easily lead to an impractical number of grid blocks for large, full field models with many and complex wells. We have developed practical solutions for accurately representing both flowing and shut-in performance of hydraulically fractured vertical wells and multiply fractured horizontal wells in full-field models. We have validated these approximations in trials with actual field examples. We found it necessary to enable wellbore cross-flow in the simulator for coarse grid field-scale models to more accurately simulate observed shut-in pressures. We also found differences in pressure distribution results of finite difference simulation between models using a pressure formulation vs. those using a pseudo-pressure formulation when simulating tight gas reservoirs under certain conditions. In summary, the existing commercial reservoir simulation tools can be adapted to model challenging tight gas depletion scenarios and optimize well spacing. However, care must be taken to ensure that the models are properly constructed and results frequently checked for consistency. Introduction Tight Gas Simulation Needs As the industry looks to increase investment in lower quality tight gas rocks, integrated reservoir modeling studies become increasingly important to understand and improve reservoir management of tight gas fields for optimum depletion planning (see, for example, Iwere et al. 2006). For reservoir management purposes, an accurate simulation of the performance (history and forward prediction) of hydraulically fractured wells in such tight sands poses several challenges which are not encountered in routine simulation applications. One of the important challenges is appropriate representation of the hydraulic fractures. In this paper we describe a practical method of simulating hydraulic fractures that requires fewer gridblocks than methods that explicitly model the fracture. There are numerous reasons to attempt to simulate performance of tight gas reservoirs. Two of the most common areTo compare the performance of various completion optionsTo determine/evaluate optimum infill well locations with corresponding incremental reserves and economic benefits The first of these can often be done with single well models in which the number of gridblocks used to simulate each hydraulic fracture can be quite large (and a single horizontal well with a large number of hydraulic fractures will require even bigger number of grids) but the second may require a large number of wells with a correspondingly large number of fractures thus results in an impractical number of gridblocks.