Numerical Modeling of Radial Fracturing of Cement Sheath Caused by Pressure Tests

To achieve an acceptable level of zonal isolation, well integrity should be guaranteed in hydrocarbon production and geological CO2 sequestration. Well pressure test can cause different types of failures in the well system leading to leakages through these failures. Laboratory evidences have revealed that occurrence of radial tensile fractures is likely during pressure tests. In this paper, we use a numerical code call MDEM which was formulated based on discrete element method. The code can model discontinuum feature of fractures. A model of a lab-sized pressure test was built and compared to an experiment previously published. The model was tested under different confinement levels and effect of the tensile strength of rock on the radial fracture was investigated at the same lab-scale. Fracture opening profiles are also presented showing the leakage potential of these fractures under different pressure level.

Additives to Enhance Cement Sheath Durability

The primary goal of the oil and gas well cementing is zonal isolation. During the production life of a well, the cement experiences various severe conditions affecting its permeability. These conditions include cracking, debonding, and shear failure which can be worsened by pressure fluctuations during hydraulic fracturing operations. Any of these conditions by forming micro-cracks within the cement or micro-annuli at the casing/cement or cement/rock interfaces create cement permeabilities far beyond the intrinsic permeability of the intact cement sheath. Recently, some studies have been devoted to improving the overall mechanical behavior of the cement by adding carbon nanotubes and carbon nano-fibers. Although these nano-additives offer considerably high strength and modulus, the high costs of these materials persuade us to find alternatives at relatively low costs, such as, graphite nanoplatelets (GNPs). Our preliminary laboratory studies show the effectiveness of GNPs in the enhancement of durability characteristics of the prepared nanocomposite cement paste by improving its compressive strength, ductility and toughness resistance. Considering the importance of dispersion of nanoadditives within the cementitious matrix, we physically or chemically manipulate the surface properties of GNPs to prevent the agglomeration of nanoparticles.

Increased Recovery Using Autonomous Inflow Management

Inflow control devices (ICD) were first introduced 26 years ago on the Troll field. The main purpose was to reduce water coning to delay water production. This technology is commonly used in long horizontal wells. An ICD is a passive orifice. More recently several types of active devices have been developed. The choking effect here depends on viscosity, fluid density or pressure contrasts. They are called autonomous devices as they react on changes inside the reservoir, without signal from surface. The main objective is to maximize oil recovery, before water production is so large that the wells are abandoned. A master thesis study conducted at the University of Stavanger together with Neptune Energy has investigated the applications of passive and autonomous inflow devices, to see which tool actually provides the highest oil recovery. The analysis was based on existing products and tools under development. Areas where a specific tool works most optimally were identified. Wells from a producing field were used as candidates for the analysis. A considerable portion of the work was to build a realistic reservoir simulator from production data. This paper will present the work and discuss the results of the study.

The Calving Events of Petermann Glacier From 2008 to 2012: Ice Island Drift Characteristics, Assessment of Fracture Events, and Geographical Data Analysis

The eastern Canadian Arctic is an ice-prone environment that is a vital part of Canadian Arctic shipping lanes. A better understanding of the ice environment and ice characteristics in this region is essential for supporting safe and economical marine activities. This study presents a first analysis of the drift of ice islands that originated from the Petermann Glacier calving events in northwest Greenland between 2008 and 2012. These massive calving events generated numerous smaller ice islands and icebergs through subsequent deterioration and break-up events. Surviving ice features drifted further southward into the Baffin Bay and reached as far as offshore Newfoundland (∼47 °N) for the case of the 2010 calving event. The drift characteristics of Petermann ice islands are evaluated through the analysis of the recently developed Canadian Ice Island Drift, Deterioration and Detection (CI2D3) database. The average drift distance, speed, and directions of the ice islands that resulted from the 2008, 2010, and 2012 calving events were estimated using successive observations of the monitored ice islands in the CI2D3 database. This study also includes an assessment of fracture events, including the total number of ice island break-up events following each massive calving event and the average number of daughter ice islands resulting from each break-up event. A geographical analysis of the data was also performed to present the location of the fracture events, as well as the time series of latitude change of Petermann ice islands from their origin (northwest Greenland ice tongues) to where until they became too small (< 0.25 km2) to be delineated in the CI2D3 database. This information is of particular interest to marine activities in the eastern Canadian Arctic, and oil and gas operations offshore Newfoundland and Labrador.

Understanding the Phenomenon of Dissolved Gas Migration of Gas in Riser During Drilling Operations

The prevention and control of gas kicks is a major concern in the petroleum industry during deepwater drilling operations. The problem is further aggravated when dealing with synthetic and oil-based muds (SOBM and OBM) that can dissolve a gas influx entering the wellbore. Due to the solubility of formation gases in drilling fluids, the gas cut mud resulting from gas absorption has a density lower than that of overlaying unsaturated drilling fluid. Lab scale experimental tests were conducted in order to understand whether buoyancy-induced convection and diffusion attribute to mass transfer of a dissolved influx. Experiments were performed on a low-pressure mass transfer apparatus using carbon-dioxide (CO2) and mineral oil to study the extent of mass transfer due to buoyancy induced convective flow and diffusion. Measurements were made on the axial distance travelled by the dissolved carbon dioxide and gas concentration over the length of a pipe by measuring the mass of gas accumulated in different test sections of the experimental apparatus. This arises due to a concentration gradient developed when contaminated fluid comes in contact with a fresh column of drilling fluid. Experimentally obtained measurements made on the mass transfer coefficient are used to tune simulations carried out using a computational fluid dynamic (CFD) software — ANSYS Fluent. This enables us to replicate field scenarios to study the extent of well control issues that could arise when a gas influx enters the wellbore, even when circulation has ceased. Results obtained here can be used as a base case to understand a similar phenomenon occurring when extended to other fluid systems such as that of a natural gas influx in synthetic oil-based drilling fluids.

An Experimental Investigation of Filtercake Reinforced Wellbore Strengthening and Fracture Sealing

Drilling-induced lost circulation should be managed before and during fracture initiation rather than after they propagate to form large fractures and losses become uncontrollable. Recent studies indicated the potentially critical role of filtercake in strengthening the wellbore through formation of a pressure-isolating barrier, as well as plugging microfractures during fracture initiation. In this study, an experimental investigation was conducted to understand the role played by filtercake in the presence of lost circulation materials (LCMs). A modified permeability plugging apparatus (PPA) with slotted discs was used to simulate whole mud loss through fractures of known width behind filtercake. Cumulative fluid loss upon achieving a complete seal and the maximum sealing pressure were measured to evaluate the combined effects of filtercake and LCMs in preventing and reducing fluid losses. The effects of some filtercake properties (along with LCM type, concentration and particle size distribution) on filtercake rupture and fracture sealing were investigated. The results indicate that filtercake can accelerate fracture sealing and reduce total mud loss. Efficiently depositing filtercake while drilling can reduce the concentration of LCM that is required to plug and isolate incipient fractures.

Numerical Simulations of Riser Gas Behavior in Non-Aqueous Muds Using a Modified Drift Flux Model

Real-time prediction of riser gas behavior is of great importance in well control. Single bubble models have, thus far, been used to describe gas-in-riser events and define riser equilibrium. These models have however not considered the transient nature of desorption of gas influx from non-aqueous fluids (NAFs) during migration or circulation in a riser. This paper uses a modified drift-flux model (DFM) to more properly describe gas-in-riser events by incorporating time-dependent mass transfer processes in NAFs. In this paper, we modified the DFM to account for the gas-liquid mass transfer due to the time dependent desorption of the gas phase. The advection upstream splitting model (AUSMV) hybrid scheme was used to solve the model. The time dependent mass transfer is calculated using a kinetic model developed based on recent experimental data. The capability of this model to improve riser gas management is demonstrated using a case study and the simulations are compared to when mass transfer between gas influx and NAF is not considered. Results also show that the severity of unloading and depth of the riser equilibrium can be underestimated if a time dependent desorption is not considered. The concept of riser equilibrium has been, thus far, developed without due consideration of mass transport of gas phase in the mud. This paper factors in the time-dependent desorption of the gas phase in the mud for a more realistic prediction of riser gas unloading events.

Feature Selection for Kick Detection With Machine Learning Using Laboratory Data

Conventional kick detection methods mainly include monitoring pit gains, surface flow data (flow in and flow out), surface and down-hole pressure variations, and outputs from physics-based models. Kick detection times depend on a driller’s individual ability to interpret these drilling measurements, symptoms and model predictions. Furthermore, testing a novel data-driven solution in a full-scale operation may induce non-productive time, safety risks and crew fatigue adding to false alarms that inevitably occur during testing. Therefore, the development of better, faster and less human intervention-dependent kick detection on a laboratory scale system is a valuable step before full-scale testing. We have generated a dataset containing seven typical drilling measurements and a sequence of gas kicks from experiments conducted in the laboratory scale. First, we employ data analysis tools following data pre-processing steps, data scaling, outlier detection, and natural feature selection. Next, we consider additional “engineered features” and apply different feature combinations to logistic regression with an ensemble method (boosting) for developing kick detection algorithms. In our data analysis, ‘Delta flow’ (difference between flow in and flow out of the well) and ‘Rate of change of delta flow’ designed features, combined with logistic regression and boosting, give promising results in detecting kicks. Finally, we propose an intelligent algorithm and alarm architecture for a complete kick alarm system, which draws from both data analysis and machine learning models developed in this work.

Rheology of Brine-Based Fuzzy-Ball Drilling Fluids in Deepwater Drilling

In deepwater drilling, the rheology of traditional drilling fluid is uncontrollable since the fluid usually mixes with brine and encounters low temperature. A solution may be to use the newly designed brine-based fuzzy-ball drilling fluids (BFDFs) since these have a well-adapted rheology under high salinity and low temperature condition. This has the potential to make drilling safer and more efficient. In this experiment, the rheological properties of BFDFs under test conditions were characterized with a rheometer by varying salinity (2 to 20 mass%) and temperature (4 to 80 °C). The rheological parameters considered are apparent viscosity (AV), plastic viscosity (PV), yield point (YP), and θ6 reading. To characterize the magnitudes of changes of the rheological parameters and their low temperature dependence, their ratios at 4 and 25 °C, and 4 and 80 °C were calculated. The results showed that the apparent viscosity (AV), the plastic viscosity (PV), the yield point (YP), and θ6 reading of BFDFs increased slightly with the decrease of salinity and temperature. The ratios of rheological parameters at 4 and 25 °C were close to unity, while the ratios at 4 and 80 °C were about two. The flow behavior of BFDFs under high salinity and low temperature condition was stable. Therefore, brine could be used as the base fluid for BFDFs. Theoretically, the flow behavior of BFDFs under low temperature condition seems to follow the Herschel-Bulkley model. Practically, the tests indicated that the BFDFs possess a strong tolerance to sandstone cuttings and Cabentonite, an excellent inhibitive property to shaly cuttings, weak corrosive characteristics against N80 casing steel, excellent lubricity properties, and remarkable biodegradability. In summary, the empirical results showed that the newly designed fuzzy-ball working fluid can use brine instead of fresh water as based fluid and maintain remarkable properties under high salinity and low temperature condition. Properties of BFDFs could basically satisfy the requirement of deepwater drilling work.

Optimizing Well Locations in Green Fields Using Fast Marching Method: Optimize Well Locations for Millions of Cells Using Hundreds of Scenarios and Realizations With High Accuracy in Seconds

To perform an optimization study for a green field (newly discovered field), one must collect the information from different parts of the field and integrate these data as accurately as possible in order to construct the reservoir image. Once the image, or alternate images, are constructed, reservoir simulation allows prediction of dynamic performance of the reservoir. As field development progresses, more information becomes available, enabling us to continually update and, if needed, correct the reservoir description. The simulator can then be used to perform a variety of exercises or scenarios, with the goal of optimizing field development and operation strategies. We are often confronted with important questions related to the most efficient well spacing and location, the optimum number of wells needed, the size of the production facility needed, the optimum production strategies, the location of the external boundaries, the intrinsic reservoir properties, the predominant recovery mechanism, the best time and location to employ infill drilling and the best time and type of the improved recovery technique we should implement. These are some of the critical questions we may need to answer. A reservoir simulation study is the only practical means by which we can design and run tests to address these questions in sufficient detail. From this perspective, reservoir simulation is a powerful screening tool. The magnitude, time and complexity of a reservoir simulation problem depends in part on the available computational environment. For instance, simple material balance calculations are now routinely performed on desktop personal computers, while running a field-scale three-dimensional simulator may call for the use of a supercomputer and may take many days to finish. We must also take into account the storage requirements and limitations, CPU time demand and the general architecture of the machine. The problem arises when there is a large amount of data available with a study objective that requires running several scenarios incorporating millions of grid cells. This will limit the applicability of reservoir simulation as it will be computationally very inefficient. For example, determining the optimum well locations in a field that will result in the most efficient production rate scenario requires a large number of simulation runs which can make it very inefficient. This is because one will have to consider multiple well scenarios in multiple realizations. The main purpose of this paper is to use a novel methodology known as the Fast Marching Method (FMM) to find the optimum well locations in a green oil field that will result in the most efficient production rate scenario. The concept of radius of investigation is fundamental to well test analysis. The current well test analysis relies on analytical solutions based on homogeneous or layered reservoirs. The FMM will enable us to calculate the radius of investigation or pressure front as a function of time without running any simulation and with a high degree of accuracy. The calculations can be done in a matter of seconds for multi-millions of cells.