Uncertainty in Geomechanical Modeling
Abstract The active application of geomechanical modeling in the oil and gas industry began in the early 2000s. Geomechanics is used in all life cycles of the field-from exploration drilling to field development models, construction of production wells and monitoring of the development process. The success of geomechanical modeling directly depends on the quantity and quality of input information. In contrast to the geological and hydrodynamic models in geomechanics, there is still no common approach and algorithm for the quantitative determination of the error of the model. This paper presents an algorithm for determining the 1D error of a geomechanical model, taking into account the measurement errors of the devices and the correlation dependencies used. This is the first step towards the formation of a unified approach to the quantitative determination of the error of 1D and 3D geomechanical models and the construction of uncertainty corridors. In this paper, we propose a mechanism for estimating the error of correlation dependencies, which includes not only the error of the measuring equipment, but also the resulting convergence of the dependence and the rate of growth or decline of this convergence. The confidence parameter is used to estimate the contribution of a particular dimension to the calculations. The sensitivity of the resulting error to the quality of the source information is analyzed. In the application of this technique, it is possible at the initial stage to make a conclusion about the quality of the simulation results and to take measures to improve the reliability of the calculations.