This paper presents the results of a simulation study to evaluate the effects of gas-water fracture relative permeabilities on post-fracture performance for a well stimulated with a water-frac. More specifically, we investigate the effects of the fracture relative permeability curve shape on both short-term and long-term post-fracture production performance. We utilize a reservoir flow simulator coupled with a geomechanical model which allows us to account for fracture growth during the stimulation treatment and to model realistic fluid distributions in both the reservoir and fracture. We also incorporate new fracture relative (gas-water) permeability data measured in the laboratory for a range of proppant types. Unlike many previous studies that simply assume a linear shape, the new measured fracture relative permeability data are quite non-linear.
We compare short-term fracture cleanup and long-term water-frac production performance using both the measured non-linear as well as hypothetical linear gas-water fracture relative permeability curves. Generally, the non-linear relative permeability data---combined with low initial absolute fracture conductivities---create very low effective fracture conductivities to gas and cause ineffective fracture cleanup. Although fractures with high absolute conductivities clean up more effectively, we still observe significant residual water saturations---even after several hundred days of production. We also observe differences in the longer-term production performance caused by residual fracture water saturations that are much higher than originally thought. Finally, we assess the effects of relative permeability curve shape on post-fracture diagnostics using pressure transient testing. Evaluation of simulated pressure buildup tests suggests the computed fracture half-lengths are essentially equal to the model inputs, but the computed effective fracture conductivities are much lower.
Wells producing from tight gas sands require hydraulic fracturing to achieve economic rates and to maximize ultimate recoveries. Depending on the type and size of the treatment, hydraulic fracturing may be expensive---often representing a significant percentage of the total completion costs. Since the economic viability of wells completed in tight gas sands depends on minimizing costs, then it is essential that we optimize fracture treatments---i.e., find the proper balance between stimulation costs and well productivity. A key component in achieving the optimum stimulation treatment is the creation of high effective fracture conductivity to gas.
Several previous studies have demonstrated the importance of high effective fracture conductivity on well performance. Montgomery, et al. and Sherman, et al. suggest low effective fracture conductivities will result in low recoveries of fracturing fluids and an associated increase in fracturing fluid damage[3,4] to the reservoir, especially in tight gas sands with low or immobile connate water saturations. Furthermore, Soliman and Hunt showed the fracture conductivity required to clean up a fracture effectively is often much higher than is necessary to recover the gas. Therefore, we conclude that optimum fracture treatments must focus on creating sufficient effective fracture conductivity during both short-term cleanup and long-term production. [6–8]
Early hydraulic fracture treatments utilized polymer gel fluids with large proppant concentrations in an attempt to create long, highly conductive fractures. Although excellent for transporting proppant, the gels often damaged both the formation and fracture, thereby creating low effective fracture conductivities. Moreover, conventional gel treatments are very expensive. Under these conditions, minimal effective stimulation may be achieved---sometimes resulting in sub-economic or even uneconomic wells.