Predicting production performance of CBM reservoirs

Prediction of gas production from the coalbed methane (CBM) reservoirs is challenging due to the complex interaction of storage and transport mechanisms. The vast majority of the gas in CBM reservoirs is stored by adsorption in the coal matrix which practically has no permeability. The flow to production wells however takes place through the cleats or the natural fracture system which store relatively small amounts of gas. These unique coal characteristics have resulted in classification of CBM as an unconventional gas resource. Gas production from CBM reservoirs is governed by gas diffusion through coal matrix followed by gas desorption into the cleat system through which the gas flows to the wellbore generally under two-phase conditions. As a result, the production profile of the CBM reservoirs greatly differs from conventional gas reservoirs. This precludes the use of common techniques such as decline curves to forecast the recovery, future revenues, and well performance. Numerical reservoir models (simulators) that incorporate the unique flow and storage characteristics of CBM reservoirs are by far the best tools for predicting the gas production from the CBM reservoirs. It is however cumbersome, time consuming, and expensive to use a complex reservoir simulator for evaluating CBM prospects when the required reservoir parameters are not available. Therefore, there is a need for a quick yet reliable tool for predicting production performance of CBM reservoirs. This paper presents a set of production type curves that can be used for predicting gas and water the production from CBM prospects. The type curves are particularly useful for parametric studies when the key characteristics are not well established. A numerical reservoir model that incorporated the unique flow and storage characteristics of CBM reservoirs was employed to develop the type curves. The impact of various reservoir parameters on the type curves was investigated to confirm the uniqueness of the type curves. The application and limitation of the type curves have been also discussed. (C) 2009 Elsevier B.V. All rights reserved.