Abstract:
Accurate estimation of in situ stresses of a subsurface formation is important to get a basic knowledge of formation structure and position of anomalies, groundwater flows, performing fracturing operations, drilling operations, oil and/or gas production stimulation, wellbore stability analysis, and coupled geomechanics-reservoir simulation in petroleum engineering. In this paper, at first a new method for estimation of minimum and maximum horizontal stresses in tectonically active area based on the modification of linear poroelastic model and minifrac test results is presented. The rock mechanical properties used in poroelastic model are determinded using the artificial neural networks. Then, this method is applied to field data in order to verify the applicability of the modified linear poroelastic model. The results indicated that the agreement between the results of minifrac test and modified linear poroelastic model is satisfactory. Furthermore, application of artificial neural networks in this methodology increases the accuracy of linear poroelastic model for estimation of horizontal stresses.
