In-Situ Stresses and Geomechanical Properties Characterization Using Robust Optimization Approach, Tested in a Tight Gas field, Sultanate of Oman

Abstract

Knowledge of in-situ stresses and geomechanical properties is important for wellbore stability and hydraulic fracture optimization applications. Both mechanical rock properties (e.g., Young's modulus and Poisson's ratio) and the stresses represent the initial step in constructing a geomechanical model that will eventually require static calibration from the lab or field tests. Nonetheless, a wellbore deformation-based inverse analysis solution has become an alternative method that characterizes in-situ stress in particular. In this paper, a genetic algorithm and probabilistic analysis methods are proposed and integrated into a well-drilled known analytical method to characterize both stresses and geomechanical properties. Systematic steps have been applied to this analysis. First, borehole geometry (i.e., multi-arm caliper), mud weight, and vertical pressure (from the density log) are well-defined inputs for deformation-stress relationships. Unknown parameters have also been determined and include horizontal stress, Poisson's ratio, and Young's modulus. Subsequently, the minimum and maximum expected values for each unknown parameter have been defined. Thousands of combinations have been created by the analytical equation (fitness function). In addition, the semi-genetic algorithm concept was used as an optimization method to find the best solution from a wide range of inputs for a given fitness function. The first hundred strongest fitness combinations were then chosen for the next level, which had a noticeably higher frequency number using the statistical analysis technique. The approach was checked with a real field example, the results indicated the measured values of geomechanical properties, and horizontal stress were reasonably consistent with the actual field data and previous studies in the field. In particular, the proposed approach allows for a realistic estimate of the most difficult stress (i.e., Max horizontal stress), which was ~45 % higher than minimum horizontal stress. The proposed technique was developed to reduce in-situ pressure uncertainties and geomechanical properties for the studied area. Results from this paper presented a simple and practical alternative method for the determination of geomechanical parameters using a simple logging tool (e.g., a caliper) that theoretically provides a robustness guide for wellbore stability and hydraulic fracture models for tight gas fields.