A Practical Workflow for Characterizing Stress-Dependent Behaviour of Coal from Changes in Well Productivity

Abstract

Abstract Stress (or pressure) dependence of coal permeability is a commonly observed and generally accepted dynamic behaviour that is often ignored from production performance forecasting. Reasons for this omission typically include (a) the difficulties in reliably characterizing stress dependent effects from a limited number of pressure buildup tests, and (b) large uncertainties in our understanding of both the porosity and compressibility of coals. This paper demonstrates a new analytical workflow, and proposes a new set of equations that overcomes some of those limitations through acknowledging and accounting for stress dependent permeability (SDP) behaviour with radius - during testing and production - by translating expected permeability changes with pressure, to productivity changes with pressure gradient. This paper utilizes a slightly modified form of Palmer-Mansoori (P&M) model in a workflow that includesEstimation of coal cleat volume compressibility using permeability-depth trendsCharacterization of mechanical skins from interpreted apparent skinsCalculation of "stress dependent pseudo pressure" (SDPP) – converting permeability changes with pressure to productivity changes with pressure gradient – enabling the use of well productivities over time as part of a SDP characterization processThe matching through regression of relative changes in well productivity indices in groups of wells - utilizing the SDPP approach – with a single stress dependent controlling parameter, in a way that is suited to extrapolating away from well control. The paper provides the theoretical support for this approach via derivations from published models, and then outlines the methodology – sharing the results of a field example – demonstrating the relative ease with which the analytical process can be applied. Further, it highlights pitfalls of using well productivities as a direct proxy for permeability changes, or even utilizing coarse grid numerical simulation in the matching process. Finally, it also discusses further applications and limitations of its application. This paper addresses existing knowledge gaps in the CSG industry by providing a simple, yet efficient, workflow for characterizing and incorporating stress-dependence of permeability in CSG Reservoir Engineering. This is achieved through the application of new analytical equations to characterize stress dependent pseudo pressure, enabling the direct use of well productivity changes, and can be used standalone, or as a means to accelerate a numerical history matching workflow.