Doctor of Philosophy (PhD)


The Craft and Hawkins Department of Petroleum Engineering

Document Type



Recent developments in downhole temperature measurements open new alternatives contributing to reservoir characterization. In this dissertation, novel forward and inverse models to analyze production- and injection-induced temperature signals are developed for conventional and unconventional reservoir applications. Important limitations of the proposed models are addressed by accounting for fluid property variations and complex production strategies.

Forward modeling approaches involve making relevant assumptions that allow rigorous analytical solutions to be constructed using Laplace transform, Method of Characteristics, and control volume analysis. Our results of the analytical models are benchmarked with those from commercial numerical simulation software. Multiple possible scenarios of conventional reservoirs are addressed including single-layer reservoir, multi-layer reservoir, near-wellbore damaged zone, and non-Darcy flow effect. To treat temperature signals associated with complex production history, we introduce methods with underlying theories of superposition principle and production rate normalization borrowed from pressure transient analysis while developing a new analytical approach when these theories are not applicable. Besides the transient flow period, boundary dominated flow is incorporated to extend the application of the proposed temperature transient analysis. We further extend the temperature transient analysis to fracture diagnostics during production and flow-back periods for unconventional reservoirs and CO2 leakage detection and characterization from storage zones.

From the analysis results, we identify major mechanisms for thermal signals associated with production/injection of fluids from/into the subsurface: Joule-Thomson (JT) effect, adiabatic expansion/compression, heat conduction, and advection. We determine the significance of these mechanisms depending on the application of interest and the dominating flow regime (transient versus boundary dominated). For conventional reservoir production cases with high drawdown and strong temperature signals, the developed fluid property correction method improves the accuracy of the forward models. The interpretation and inversion processes are mainly conducted on semi-log plots with temporal temperature signals. For conventional reservoirs, the inverse modeling estimates permeability, porosity, damaged zone permeability and radius, non-Darcy flow coefficient, drainage area, and reservoir shape. Other outputs from the inversion procedures include leakage rate and transmissibility for CO2 leakage, and inflow fluid temperature, surrounding temperature field, and after-flow velocity of each fracture during unconventional reservoir production and flow-back.



Committee Chair

Zeidouni, Mehdi