A Thorough investigation of cleanup efficiency of hydraulic fractured wells using response surface method

Poor cleanup efficiency of injected fracture fluid (FF) has been considered as one of the main factors contributing to the poor performance of many hydraulically fractured wells (HFWs). Limited parametric studies evaluating the efficiency of FF cleanup have not embarked on a much needed extensive investigation of variation of all pertinent parameters.

In this work we present the results of over 130000 simulations of the process, for a HFW model that was constructed using a reservoir simulator. A computer code has been developed, which automatically, read input data, link the injection and production periods and create the output data. The impact of 16 parameters describing the gas and FF effective permeability of matrix and fracture, pressure drawdown, capillary pressure, and porosity have been studied for two injected FF volume values. Different statistical experimental design methods have been used to sample a reasonably wide range of variation of pertinent parameters. Linear and quadratic response surface models have been used to map the gas production loss (GPL), compared to 100% cleanup case.

The results indicate that GPL is mainly controlled by parameters related to FF cleanup inside the fracture particularly fracture permeability. In some cases increased back flow of FF from matrix into fracture increases GPL. As production continues, the impact of matrix permeability and gas exponent of Corey type relative permeability curve in the matrix become more pronounced. The fracture residual gas saturation and matrix gas end points have negligible effect. The relative importance of pertinent parameters is less for lower FF injection volume and especially at higher production periods. These practical findings can be used to make a better decision on the performance of such costly operations, suggested methods for improving the cleanup efficiency of FF and the optimum fracture design practices.