Predictive modelling of naturally fractured reservoirs using geomechanics and flow simulation

Abstract

Abstract To optimise recovery in naturally fractured reservoirs, the field-scale distribution of fracture properties must be understood and quantified. We present a semi-deterministic method to systematically predict the spatial distribution of natural fractures and their effect on flow simulations. This approach enables the calculation of field-scale fracture models. These are calibrated by geological, well test and field production data to constrain the distributions of fractures within the inter-well space. First, we calculate the stress distribution at the time of fracturing using the present-day structural reservoir geometry. This calculation is based on geomechanical models of rock deformation such as elastic faulting. Second, the calculated stress field is used to govern the simulated growth of fracture networks. Finally, the fractures are upscaled dynamically by simulating flow through the discrete fracture network per grid block, enabling field-scale multi-phase reservoir simulation. Uncertainties associated with these predictions are considerably reduced by constraining and validating the models with seismic, borehole, well test and production data. This approach is able to predict physically and geologically realistic fracture networks. Its successful application to outcrops and reservoirs demonstrates there is a high degree of predictability in the properties of natural fracture networks. Several examples show the success of the method in singleand multi-phase fields. In cases of limited data - where stochastic models typically fail - this method remains robust. Introduction Natural fracture systems can have a dramatic impact on reservoir performance - they may act as highly permeable flow conduits or act as baffles and seals. The complexity of a fracture network typically leads to an extremely heterogeneous and anisotropic permeability distribution within the reservoir. Successful management of these reservoirs is impossible without substantial knowledge of the natural tensile and shear fracture systems. It is essential to know their spatial distribution and hydraulic properties on an inter-well scale to properly simulate the field-wide recovery processes. This paper presents a new method for predicting natural fracture distributions and their effect on reservoir simulations (Figure 1). The first step uses geomechanical models of rock deformation to calculate the field-scale distribution of stress responsible for fracturing from the observed structural geometry of the field. Fracture network geometries are then obtained by simulating the initiation, growth, and termination of fractures within the calculated stress field. These predicted network geometries are partially constrained and validated by core, borehole image, mud loss, and outcrop data. Thereafter, multi-phase, well-scale or field-scale flow simulations of the fracture model are validated and calibrated against well test and production data. Close integration of fracture prediction and flow simulation enables significant reductions in uncertainty by using all the available static and flow data to constrain a single model. In this way, for instance, standard ambiguities in borehole fracture data due to sampling bias can be overcome by the use of well inflow data. Moreover, as the fracture model is field-scale, the greater the number of wells available the smaller the uncertainty in fracture prediction becomes across the whole field and not just around the wells. Such reduction in uncertainty allows improved field development through:better assessment of the recovery mechanism,more reliable production forecasts,well placement for optimal drainage,minimal water-cut, andrecognition of drilling hazards associated with fractures. The following sections describe how structural geometry is used to predict stress, how stress is used to predict fractures, and how fractures are used to predict flow. We finish by presenting two applications of this method to producing carbonate reservoirs. 2. Fracture prediction It is neither possible nor generally necessary to accurately predict individual fractures within a reservoir. Rather, we restrict our attention to predicting just the properties of those tensile and shear fracture networks that are hydraulically conductive. We calculate the stress field responsible for reservoir fracturing using geomechanics. Brittle fractures form where this stress field exceeds the local material strength as characterised by the brittle failure envelope for both tensile and shear fractures.