| Progressive percentage of total fluid produced in the oil industry is formation brine. Ever, increasing water cut will reduce oil recovery, diminish wells' productivity and increase cost of eventual artificial lift and produced water handling. This study investigates the problem of early water development in layered and heterogeneous reservoirs and determines the effect of various reservoir parameters on the development of water front movement in the presence of thief zones. A water injection in a line-staggered pattern was simulated to analyze these effects on the water breakthrough time, WBTT, and the evolution of water front in the thief zone. To achieve this, sensitivity analysis is conducted to investigate and determine the effect of some reservoir parameters that would explain the experience of having unpredicted advancement of injected water resulting in early water breakthrough and high water cut wells. These reservoir parameters included layers' horizontal permeability, Kv/Kh ratio, thickness of high permeability layers, water gravity effect, oil API gravity effect, and injection/production rate ratio (or IPR). Threshold of these parameters beyond which its effect would be constant is also determined to help operators to better estimate the water breakthrough time and hence better decision making process in waterflooding projects.;Moreover, water shut-off, WSO, treatments are simulated to determine their effectiveness in delaying the water breakthrough time, and reducing water cut percentages for maximum possible time, under different thief zones' conditions. Extreme thief zone cases are selected from the first part of the study for this purpose. Also, the optimum WSO treatment thickness is identified at which a maximum delay in time is achieved which yields to best treatment practices in the fields. Finally, this study summarizes the applicability of these WSO treatment methods, and it identifies the level of effectiveness based on specific given petrophysical data that are presented as dimensionless variables to be able to generalize the predictions. |
