Modeling and Prediction of Borehole Collapse Pressure during Underbalanced Drilling in Shale
MetadataVis full innførsel
Drilling with a bottomhole pressure less than the formation pore pressure (underbalanced) will usually increase the risk of borehole collapse due to yielding or failure of the rock adjacent to the borehole. Understanding borehole collapse mechanisms during underbalanced drilling (UBD) in shale is becoming increasingly important for the petroleum industry, especially due to the implementation of the UBD technique in operational practices. The UBD technique has become a bit of an art in the modern oil industry, often considered to avoid or mitigate formation damage, reduce the risk of lost circulation, enhance recovery and increase rate of penetration (ROP). So far, the potential of UBD has yet to be fully realized by the industry, particularly because borehole stability has not been well addressed. In this study, we will therefore focus on mechanical borehole stability and improving the prediction of the borehole collapse pressure while drilling in shale. From an operational point of view, rotating BOP, drill string vibration, downhole restrictions and the selection of an accurate borehole stability model are important concerns when trying to increase drilling efficiency and safety in UBD wells. Of these four major issues related to UBD, only the one related to wellbore stability were considered in this PhD project. The other three issues are complex and represent individual research topics. Other issues, such as the inclusion of a borehole stress model, a geo-mechanical model, in-situ stresses, true pore pressure in shale, thermal or cooling effects on rock stiffness and strength, dilatancy, lamination, anisotropy, heterogeneity and shale permeability, are important to include in conjunction with borehole stability models. To prevent wellbore instability problems (i.e., borehole collapse, sloughing shale, stuck pipes, hole cleaning or fracturing formation), accurate predictions of stresses, pore pressure, rock stiffness and strength, and deformations around the wellbore are essential. Recent research shows that several mechanisms contribute to shale instability including shearing, swelling, pressure diffusion, heterogeneity, anisotropy, capillary effects, osmosis and physico-chemical alteration. During UBD, it is believed and assumed that there is no chemical effect and the rock is inert. Thus, there is no sloughing due to chemical instability, only mechanical failure through shear or radial tensile failure. We show in this study that several factors need to be considered to properly understand borehole stability issues in UBD wells. Physical models can address the UBD hypothesis and help understand the borehole stress model. Rock strength, mud weight and pore pressure are the most important parameters used for both the UBD and borehole stability hypotheses. Existing geomechanical models have been implemented to verify the behavior of shale during underbalanced drilling. However, when applied to the problem areas, several weaknesses in these models become apparent. On the basis of these models, an assumption of shale permeability was made to derive further estimates for shale consolidation and its effect on time-delayed wellbore stability. The specific behavior of shale was further investigated to define material characterizations that affect yielding and dilation. These studies were supported by testing of samples that verify previous results and allowed us to develop a new procedure for shale testing. A combined experimental/numerical investigation was performed to determine how the directional properties of shale impact borehole stability in weak shale formations. In all, three triaxial undrained tests for each of four different sample-inclinations (0, 45, 60 and 900) relative to the bedding plane, as well as two drained triaxial tests and Brazilian tests, were performed to obtain an accurate description of the anisotropic properties of the shale. Thick-walled hollow cylinder tests at underbalanced conditions were carried out on samples drilled parallel and perpendicular to the bedding to calibrate the material model on a developed virtual borehole. Laboratory observations of material failure were compared with the numerical outcomes. An improved understanding of the behavior of shale during UBD will improve stability analysis models and facilitate more rational predictions of borehole collapse risk. Several material constitutive models have been considered for rock failure studies, including the Mohr-Coulomb, the Mogi-Coulomb, the Mohr-Coulomb elasto-plastic and the Modified Cam-Clay models. The effects of material plasticity and mud cooling on mud weight were quantified. Shale specimens were investigated experimentally for the development and organization of consistent input parameters intended for model calibration and the improvement of accuracy. This research also developed model-fitting parameters for the Mogi-Coulomb model and reorganized the Mohr-Coulomb closed-formed analytical solution to predict borehole collapse pressure. Additionally, the timedelayed borehole failure, when subjected to transient pore pressure effects, was discussed. This study provides in-depth information and techniques for a developed mud design tool for UBD wells in shale. This was achieved through the development of various concepts on UBD and borehole instability, the implementation of linear and non-linear material models to predict collapse pressure and the evaluation of mechanical properties of the rock, such as the stiffness and strength and parameters of failure. A combination of theoretical, analytical, numerical and experimental work was performed to achieve the overall goal of this project. The project workflow is presented briefly in an introduction of the findings and conclusions, followed by several conference proceedings and submitted journal articles. Numerical studies were carried out in MATLAB and ABAQUS. An inhouse mechanical analysis calculator for rock was developed and used during the simulations of analytical solutions of poroelastic model. The main contribution of this PhD project was to improve the confidence and reduce confusion when choosing an appropriate material model to predict borehole failure, with the ultimate goal of minimizing wellbore stability problems experienced by drilling communities. Extensive experimental data related to the characterization of shale anisotropy and heterogeneity is provided and will be useful to other researchers working with shale.