Laboratory Studies on the Stability of Vertical and Deviated Boreholes

Abstract

*SPE Member Abstract Polyaxial or true triaxial tests have been performed on Polyaxial or true triaxial tests have been performed on weak and strong sandstones in order to investigate the stability, of model boreholes. Both vertical and deviated boreholes in anisotropic stress systems have been studied. The results indicate the decrease in stability of the deviated boreholes with increasing inclination. Boreholes oriented parallel to the maximum horizontal stress direction also produces less stable boreholes compared to those drilled parallel to the minimum horizontal stress. Comparisons of the laboratory test results with simple predictions using linear elasticity, and a Mohr Coulomb failure criterion, show that the boreholes in the strong sandstones are approximately 4 times stronger than predicted; while the weak sandstones are approximately 8 times stronger than predicted. predicted. 1.0 Introduction Increasing use of highly deviated, extended reach and horizontal drilling, has focussed attention on many aspects of drilling, not least on the ability of highly deviated holes to remain open. Associated with this issue, is the potential for hole cleaning problems or the drill packing off, both of which are aggravated by excessive spalling of the wellbore wall. The challenge of minimising such problems, has resulted in the need for more analytical tools, with which to plan the drilling of the boreholes, and assess their feasibility. The analytical and numerical prediction of borehole stability predominantly considers the stresses develsped in a plane perpendicular to the borehole axis, and the comparison of these stresses with the strength of the surrounding material. The comparison of such analyses with laboratory tests, shows an under estimation of the strength of the boreholes by factors of two to four times for competent rocks (Guenot and Santarelli. These conservative predictions are generally attributed to the use of linear elasticity in the analyses and models. Analytical solutions and numerical models incorporating pressure dependent elastic moduli and elasto-plaslicity show less conservatism associated with their predictions of mud weights required to ensure wellbore predictions of mud weights required to ensure wellbore stability. The validation, and the assessment of the accuracy of numerical models and analytical solutions, is of prime importance if these techniques are to be used as predictive tools. Validation of the predictions of these tools predictive tools. Validation of the predictions of these tools would most satisfactorily be performed by comparison with the behaviour and response of drilling a monitored borehole. However, such field validations would not only be expensive, but possibly inconclusive due to difficulties in accurately determining the three in-situ stresses and the in-situ strength and behaviour of the rock, Fleming et al. These latter parameters are of fundamental importance in the input to current analytical and numerical methods. These limitations restrict our present day validation of predictive tools to comparisons present day validation of predictive tools to comparisons with laboratory test results. P. 19