Uncertainty Assessment of Production Performance for Shale Gas Reservoirs

Accurate assessment of uncertainty and optimization of production performance in shale-gas reservoir are critical for successful planning and development of shale-gas assets. Compared to the conventional assets, shale-gas reservoirs display significant and different challenges for flow simulation, particularly in modeling of multi-stage hydraulic fractures and transport of gas from micro or nano pores to the fracture network. Stimulated fracture half-length, spacing, conductivity (initial and also during later times, i.e., during production), diffusivity, as well as adsorption parameters are highly uncertain in practice which have a huge impact on recoveries. Thus, specific methods or treatments are needed for efficient uncertainty quantification and optimization of production for shalegas reservoirs, such as the handling of key controlling parameters of fracture geometry, diffusion, and adsorption/desorption. This paper presents an integrated workflow for uncertainty assessment for well production and field development based on a newly developed approach for modeling and simulation of shale gas production in multi-staged hydraulic-fractured formations. In this approach, fracture system is modeled using three different fracture groups: the primary fractures with known geometry, the secondary fractures created by hydraulic fracturing process, and the tertiary small fractures that contribute to the enhancement of diffusion rate. The transport mechanism of gas from micro or nano pores to fracture network is also explicitly modeled through molecular diffusion and convection. Experimental design (ED) and probabilistic collocation method (PCM) are used to systematically analyze the impacts of different uncertainty parameters on gas production. Key uncertainty parameters (heavy hitters) are identified, which can be used as guidance for the field data collection process in order to reduce key uncertainties. The technologies and workflow developed in this paper are shown to be able to improve the efficiency & accuracy in uncertainty assessment, as well as to optimize field development.