Doctor of Philosophy (PhD)


Petroleum Engineering

Document Type



Although water coning is well understood, it is difficult to control in field operations resulting in low recovery and large volumes of waste produced water. A solution - proposed here - is a multi-functional well with the in-situ bottom water drainage and injection installations - Downhole Water Loop (DWL). Theoretically, DWL greatly improves well performance (for example, a two-fold increase of DWL well’s water drainage rate would increase the critical (water-free) oil rate by 80%). However, DWL has practical limitations that must be quantified for actual well design. The objective of this work is to: (1) find maximum water drainage rate to ensure separation of a small amount of under-drained oil from the drainage water; (2) learn how the small oil contamination would impact water injection and how to set criteria for oily water disposal to the bottom aquifer; and, (3) develop a method for assessing feasibility of DWL for oil reservoirs with bottom-water coning problem. Counter-current oil water separation experiments have been to simulate the flow of oil droplets in the downhole water looping section of DWL wells. From the results, an analytical model calculates the maximum water drainage rate that prevents carry-over of oil by the injection water. Aquifer injectivity decline is described by a mathematical model based on mass balance of oil phase in the injected water by considering the effects of oil droplets capture due combined effect of advection, dispersion and adsorption (ADA model) coupled with the two-phase relative permeability relationship. For comparison, a two-phase flow model based on the Buckley-Leverett theory describes aquifer permeability decline during oily water injection process. The two models are in a good agreement for linear flow and excellent agreement for radial flow. Consequently, the aquifer permeability damage is converted to time-dependent skin factor and injection pressure. A comparison of the injection and fracturing pressure gives an estimate of the well stimulation cycle and a criterion for screening reservoir-aquifer candidate for DWL. In order to assess DWL feasibility, a dimensionless model of movable oil recovery vs. seven scaling groups has been built using the inspectional analysis method and multivariable regression technique. The model is used as a final step in the five-step procedure for finding good reservoir candidates for DWL application. Six real reservoirs were used to demonstrate the procedure with three reservoirs becoming good candidates for DWL technology.



Document Availability at the Time of Submission

Release the entire work immediately for access worldwide.

Committee Chair

Wojtanowicz, Andrew