Worlds growing energy demand has pushed oil companies to explore and produce hydrocarbons in complex and technologically challenging deepwater environments. These difficult and complex operations involve the risk of major accidents as well, demonstrated by disasters such as the explosion and fire on the UK production platform Piper Alpha and capsizing of the Deepwater Horizon rig in the Gulf of Mexico (GoM). Accidents cause death, suffering, pollution of the environment, disruption of business and bad reputation to oil industry. A quantitative risk analysis technique has been used in this study to identify and categorize risk associated with different life phases of a deepwater well. Volume of oil released to the environment is used as a risk indicator. Five oil spill scenarios related to drilling and production life phases of a deepwater well are modeled. Risks associated with drilling an exploratory well in the deepwaters of GoM are analyzed in Scenario-1. A representative well location and corresponding reservoir properties were used to estimate the worst case discharge rates (WCD). Fault tree analysis (FTA) was performed to identify and categorize different hazards. Unexpected pore pressure and delayed response to an emergency situation were identified as two most important parameters contributing to overall risk of the system. In Scenario-2 an underground blowout was modeled by using representative geological settings from Popeye-Genesis field. A shallower low pressure zone is exposed to a deeper high pressure zone during drilling. The time to recharge the shallower zone to its fracture pressure is estimated. The shallower zone will transmit hydrocarbons to sea floor once its fracture pressure is reached. Risks associated with production life phase of a deepwater well are modeled in scenario-3. A representative well location and corresponding reservoir properties were used to estimate the WCD. FTA showed that sand screen and subsea tree control failures were main elements contributing to risk. In scenario-4 risk associated with floating production and offloading (FPSO) system for GoM are quantitatively and qualitatively presented. Scenario-5 deals with oil spill risk associated with severe weather conditions. An example mudslide calculation for SP-70 block of GoM is presented.

Modeling subsurface particle transport and retention is important for many processes, including sand production, fines migration, and nanoparticle injection. In this study, a pore-scale particle plugging simulator is concurrently coupled with a streamline reservoir simulator to predict the behavior of particles in the subsurface. The coupled simulators march forward in time together. The automated communication between the two models enables the prediction of spatially and time dependent parameters that control the particle transport process. At each time step, the reservoir simulator provides the inlet velocity and particle concentration of the fluid suspension to the pore-scale model which outputs the permeability, porosity, and retention coefficient. This permits the reservoir simulator to include pore-scale physics at selected locations to determine the number of particles retained and the formation damage. The pore-scale simulator tracks the path of individual particles as they are simultaneously injected into the sample and produces an effluent particle concentration curve that is fit with a continuum-scale advection-dispersion model. The advection-dispersion model is matched to the pore-scale data by adjusting two parameters: the dispersion and retention coefficient. The retention coefficient dictates the number of particles retained across a grid block in the reservoir simulator. Incorporating fundamental pore-scale physics into the streamline reservoir simulator improves its predictive ability by updating the particle retention and formation damage of a grid block at each time step.

A numerous laboratory and field tests revealed that foam can effectively control gas mobility, improve sweep efficiency, and increase oil production, if correctly designed. It is believed that there is a significant gap between small laboratory-scale experiments and large field-scale tests because of two main reasons: (i) typical laboratory flow tests are conducted in linear systems, while field-scale foam EOR processes are performed in radial (or spherical partly) systems in general; and (ii) through the complicated in-situ lamella creation and coalescence mechanisms and non-Newtonian behavior, foam rheology is thought to depend on geometry and dimensionality and, as a result, it is often not clear how to translate laboratory-measured data to field-scale applications. Therefore, this study for the first time investigates how foam rheological properties change in different dimensions and geometries and how such dimensionality-dependent properties are affected by different foam flowing conditions by using mechanistic foam fractional flow analysis. Complex foam characteristics such as three foam states (weak-foam, strong-foam, and intermediate state; sometimes referred to as foam catastrophe theory) and two steady-state strong-foam regimes (high-quality regime and low-quality regime) lie in the heart of this analysis. The calculation results from a small radial or spherical system showed that (i) for strong foams in the low-quality regime injected, foam mobility decreased (or mobility reduction factor increased) significantly with distance which improved sweep efficiency; (ii) for strong foams in the high-quality regime, the situation became more complicated near the well foam mobility decreased, but away from the well foam mobility increased with distance, which eventually gave lower sweep efficiency; and (iii) for weak foams injected, foam mobility increased with distance which lowered sweep efficiency. The results also implied that the use of fixed value of mobility reduction factor, which is common practice in reservoir simulations, might lead to a significant error, especially for strong foams in the low-quality regime. When the method was applied to the large field-scale applications, it was first shown why strong foams would eventually turn into weak foams. Then additional results showed that strong foams could propagate deeper into the reservoir at higher injection rate, higher injection pressure, and at lower injection foam quality. Foam propagation distance was very sensitive to these injection conditions for foams in the high-quality regime, but much less sensitive for foams in the low-quality regime. This study uses a mechanistic foam model similar to Afsharpoor et al. (2010) which is an updated version of Kam and Rossen (2003), Kam et al. (2007), and Kam (2008). In all calculations, gas and liquid phases are assumed to be incompressible and the presence of oil is not considered at this stage.

Reservoir characterization is one of the most important tasks that determines the recovery plan for a specific reservoir. This process incorporates a significant amount of data acquisition and processing to finally develop an acceptable model that matches the production history and can forecast the future production behavior. The model also should be able to adapt to changes along the way: adding or removing producers or injectors, changing the injection pattern, recompletions and converting wells are all examples of possible changes that are common in the oil and gas industry. Usually these changes are modeled by running field-scale simulations and providing the model with the daily data from the field to keep the model up to date and to reduce prediction errors.

Among numerous foam applications in a wide range of disciplines, foam flow in porous media has been spotlighted for improved/enhanced oil recovery processes in petroleum-bearing geological formations and shallow subsurface in-situ NAPL (non-aqueous phase liquid) environmental remediation in contaminated soils and aquifers. In those applications, foams are known to reduce the mobility of gas phase by increasing effective gas viscosity and improve sweep efficiency by mitigating subsurface heterogeneity. This study investigates how surfactant/foam process works fundamentally for environmental remediation purpose by using MoC (Method of Characteristics) based foam modeling and simulation techniques. It consists of two main parts: Part 1, developing foam model using three-phase fractional flow theory accounting for foam flow rheology such as foam strength and stability at different phase saturations; and Part 2, extending the model to investigate the mechanisms of surfactant/foam displacement in multi-layer systems. Part 1 investigates six scenarios such as different levels of foam strength (i.e., gas mobility reduction factors), different initial conditions (i.e., initially oil/water or oil/water/gas present), foam stability affected by water saturation (Sw), oil saturation (So), and both together, and uniform vs. non-uniform initial saturations. The process is analyzed by using ternary diagrams, fractional flow curves, effluent histories, saturation profiles, time-distance diagrams, and pressure and recovery histories. The results show that the three-phase fractional flow analysis presented in this study is robust enough to analyze foam-oil displacements in various conditions, as validated by an in-house numerical simulator built in this study. The use of numerical simulation seems crucial when the foam modeling becomes complicated and faces multiple possible solutions. Part 2 first shows how to interpret theoretically the injection of surfactant preflush and following foams into a single-layer system at pre-specified rock and fluid properties, and then extends the knowledge gained into multi-layer systems where the properties vary in different layers. The results in general show that the mechanisms of foam displacement strongly depend on foam properties such as gas-phase mobility reduction factors (MRF), limiting water saturation (Sw*), critical oil saturation (So*), and so on as well as petrophysical properties of individual layers such as porosity (φ), permeability (k), relative permeability and so on. The overall sweep efficiency in a multi-layer system is very difficult to predict because of the complexity, but the mathematical framework presented in this study is shown to be still reliable. The in-house foam simulator is also extended to compare with modeling results.

The continuous researvoir model updating is widely used to calibrate reservoir simulation models to production data, but many challenges remain. First, few real field data are available to test the new history matching method, and most of the data sets are synthetic cases. Second, computational cost may be high when using non-Gaussian priors or nonlinear models. Third, with large complex models, the simulation runs and history matching method require huge memory allocations. This dissertation achieves a continuous reservoir model updating workflow with a meter-scale , two-phase flow experiment. Both production and seismic data are collected in the experiment. Because the data are high-frequency sequential data with noise, the EnKF method is used to efficiently integrate them. To better understand the problem, scaling analysis is done on the capillary transition zone. Two new dimensionless numbers are introduced-capillary time and capillary length. We found that for different models, if their capillary time and gravity number are equal, the capillary length would be the same. The scaling analysis results help us find a proper flow rate for the sand tank experiment. Two experiments are conducted to test the workflow and the EnKF method. In the first one, both the production and seismic data are collected and analyzed. The production data have large errors in the flow rate and they are integrated to improve reservoir models using EnKF method. The history matching results are in an acceptable range which demonstrate that even if the observation data has large error, the EnKF method still works. In the second experiment, the errors of flow rate are reduced by measuring manually with a graduated cylinder. Because the data quality are much better in the second experiment, the observations can be matched easily.

Sustained Casing Pressure (SCP) is the undesirable casing head pressure of a well annulus that rebuilds when bled-down. As the conventional methods for SCP removal using rigs are expensive, there is a need for improvement. Annular intervention for replacing the fluid above the leaking cement with a heavier fluid to stop gas migration is a solution for SCP removal; however, previous attempts failed due to miscibility of injected fluids. Using hydrophobic heavy fluids for the purpose is a newly proposed technique to the technology. Potential of theoretically selected and produced immiscible heavy fluids are investigated in characterized annular fluids. A transparent laboratory scaled-down hydraulic analog of wells annulus provided visual evidence for displacement geometry and did the first stage testing of heavy fluid injection into clear synthetic-clay muds. A 20-foot physical model then tested the performance of the displacement process. Settling of various heavy fluids with densities from 11 to 23 ppg in drilling fluids with densities from 9 to 13 ppg provided quantitative bottom pressure data. Finally, a full-size test in 2750-foot well examined the viability of the technology. Visualization experiments proved that the counter-current flow in annulus leads to up-lifting of heavy fluid droplets and must be minimized for a desirable displacement process. Selection of injection geometry and rate are also essential to maintain a controlled transport of heavy fluid downwards. Pilot experiments developed mathematical correlations relating the process performance to fluid properties and rate. Full-size test shows that hydrophobic heavy fluids are able to slip in long columns; however, bridge-over of buoyant settling may occur due to high injection rates and/or flotation effect of migrating gas that was entrapped in annular fluid. The findings in this research present solid support to the viability of immiscible gravity displacement of annular fluid for remediating a well annulus affected with SCP. For given fluid properties and in confined annular space, injection rate is the key to a successful displacement. Finally, the research proved that the duration of a complete displacement process and required heavy fluid volume are inversely correlated. For any operation design; time and killing material restrictions must be considered.

Dispersion, deposition, and suspension of particulate materials in the carrier fluid play a significant role in the oil industry. Increasing the cuttings transport performance in deviated wells is difficult due to the rolling/sliding transport, and cuttings settling on the low side of the annulus. Insufficient cuttings transport may lead to some crucial problems such as pipe sticking, increasing in torque and drag, material damage and bed cementing quality. Increasing flow rates and improving mud properties may not be applicable for a proper hole cleaning because of the hydraulic and mechanical limitations. In such cases, additional pressure may be generated, and this causes formation fractures and drilling fluid losses. Under these circumstances, the other major contribution to cuttings transport is provided by drill-pipe rotation. In this study, the effect of drill-pipe rotation on cuttings transport behavior is investigated for eccentric horizontal wells. Whirling motion of drill-pipe is also analyzed. During drilling, drill-pipe is subjected to axial, lateral and torsional loads due to the dynamic vibrations. These loads cause that drill-pipe to lose its stability and generate snaking and/or whirling type of motion. Dynamic behavior of drill-pipe plays a significant role on cuttings transport and stationary bed removal. Turbulence modeling becomes very complicated when cuttings transport includes deposition and sliding effects. Advanced turbulence models are required to get accurate flow predictions while optimizing computational resources requirements. Unsteady SST k-ω turbulence model is applied due to its practicability and reliability in predicting cuttings transport behavior. Discrete phase is modeled with discrete element method (DEM) by including particle-particle and particle-fluid interactions with a commercial ANSYS FLUENTTM 15.0 CFD package using LSU high performance computing (HPC) resources. It is concluded that cuttings concentration significantly decreases with increasing flow rate. Drill-pipe rotation around its own axis causes cuttings swaying and distribute asymmetrically along the circumferential direction. Orbital motion of the drill-pipe contributes more to cuttings transport performance. Low whirling rotary leads to increase in annular pressure losses in low flow rates. In the turbulent flow regime, however, annular pressure losses increase with increasing whirling speed.

The cement sheath failures and nearby wellbore failures may lead to upward flow of drilling fluid or formation fluid, which may have significantly adverse consequences like loss of reserve and environmental hazards. In order to maintain wellbore integrity in the long term, it is expedient to examine the causes of failures around the wellbore and propose suitable numerical models to predict annulus cracks around the casing. The complex failure behavior of cement/rock interfaces observed in the laboratory experiments does not look like the behavior of linear or simple nonlinear mechanical interfaces. Cohesive zone method (CZM) with BK-form bilinear traction separation law can be a good candidate to reproduce the complicated failure behavior around the casing. Then fracture critical energy, cohesive strength, and the deformability can be derived for cohesive zone constitutive equations by reproducing the loading-displacement curves from laboratory and inverse analyses. In this work, the comprehensive analysis for microannulus formation is presented by utilizing axisymmetric or three-dimensional poroelastic finite element models with CZM. This dissertation investigated these aspects: 1) Two and three-dimensional analysis of cement sheath integrity around wellbores due to presence of a leakage point; 2) Stimulation multi-zone fracturing and its cement sheath integrity during hydraulic fracturing. In this research, the physical mechanism of the loss of wellbore integrity is explained by the combined effects of fluid pressure, tensile and shear stresses, as well as failures. The excessive fluid pressure induced by leakage or hydraulic fracturing fluid acts as the drive for failures. The intensified tensile stress and shear stress occur at the crack tip initiate failures if they satisfy with the failure initiation criterion. Moreover, Lab and field scaled sensitivity analysis extract the influential parameters involved in failure development. Furthermore, the matching between failure patterns from numerical analysis and real field measurements using radioactive tracer-logs provides a comparison basis for model accuracy. Additionally, micro-annulus cemented systems are further analyzed by considering interface strength heterogeneity, anisotropic in situ stresses, wellbore inclination and eccentricity. The proposed approach provides a tool for more accurate predictions of cement integrity in the subsurface conditions to quantify the risk of wellbore integrity issues.

<p> Polymer flooding is one of the most common chemical EOR techniques used in EOR projects worldwide. However, field applications of polymer flooding are usually limited by the cost of polymer and its considerable loss during the injection. The injectivity and efficiency of polymer flooding are significantly affected by polymer degradation, polymer retention, and high velocity near the wellbore. A function of nanoparticles&mdash;encapsulation&mdash;can be used to reduce these adverse impacts. Encapsulating polymer in the nanocapsules can isolate polymer from the environment and from impact near the wellbore for a designated releasing time, thereby reducing the impact near the wellbore. To explore the effect of nanocapsules on polymer injection, numerical models for polymer flooding were applied to test the decrease of injection pressure between encapsulated polymer injection and pure polymer injection, under different values of simulation parameters. The viscosity model was then integrated into an in-house grid-based simulator to simulate the transport of capsules and released polymers in the formation. The result indicates that for an engineered releasing time, encapsulating polymer in the nanocapsules can prevent polymer from contacting the environment near the wellbore, thus reducing the injectivity loss occurring near the wellbore and transporting polymers to further areas in the reservoirs.</p><p>