College of Petroleum Engineering, Yangtze University, Wuhan, China; Department of Civil Engineering, Kumasi Polytechnic, P.O. Box 854, Kumasi, Ghana
Owusu, P.A., College of Petroleum Engineering, Yangtze University, Wuhan, China, Department of Civil Engineering, Kumasi Polytechnic, P.O. Box 854, Kumasi, Ghana; De Hua, L., College of Petroleum Engineering, Yangtze University, Wuhan, China; Nagre, R.D., College of Petroleum Engineering, Yangtze University, Wuhan, China, Department of Civil Engineering, Kumasi Polytechnic, P.O. Box 854, Kumasi, Ghana
A model which approaches the recovery performance of waterflooding in communicating stratified reservoirs with log-normal permeability distribution is presented. The reservoir permeability is characterized by the Dykstra-Parsons variation coefficient (V) and standard deviation of the distribution (σ_k). The hydrocarbon pore volume is also characterized by normal distribution. The model allows for effect of variation of hydrocarbon pore volume. Comparison of the results obtained is made with the Dykstra-Parsons model. The performance of the reservoir is expressed in term of vertical sweep as a function of the producing water-oil ratio. Expressions for pseudo-relative permeability function and fractional flow curves are derived. The recovery performance is found to be dependent on mobility ratio and permeability variation. For a given water-oil ratio the vertical coverage increases as mobility ratio increases but decreases as permeability variation increases. Oil recovery therefore is enhanced for system with mobility ratio less than unity and retards for system with above unit mobility ratio. Below unit mobility ratio, Dykstra-Parson model gives reduced vertical coverage but higher at above unit mobility ratio compared to the model presented. The variation of porosity is found to increase oil recovery over the case of constant porosity and it is dependent on the rise of permeability variation and mobility ratio. Truncating the permeability significantly increases the vertical coverage and telling effect on the behavior on the performance prediction. © 2015 CAFET-INNOVA Technical Society All rights reserved.