Integration of microseismic data with geomechanical model

Hydraulic fracturing is a technology that is applied to increase production from unconventional resources such as shale gas and tight oil. Many hydraulic fracturing models have been developed in recent years that can be used to predict the performance of a hydraulic fracturing treatment. Inverse modeling involves conditioning simulation models to field data. The solution to the inverse problem provides estimates for formation properties and other model input parameters. The tuned model can then be used for predictive simulations. In this research, we developed a framework that can solve inverse problems using a Gibbs sampler and a radial basis function proxy model. In multidimensional inverse problem, the solution is often not unique, which means that there are multiple combinations of input parameters that can provide matches to the data. The goal of this research is to find a practical algorithm that can find all possible parameter combinations that match the available data, within bounds specified by the user. This is better than obtaining a single solution that matches the data, because it allows us to better account for uncertainty. For finding precise matches to data, our algorithm would be outperformed by other methods. However, our algorithm is very good at efficiently mapping out the global solution space and solving highly nonunique problems. The framework that we propose also allows us to visualize the relationship between input variables in order to better understand the underlying physical processes. In this report, two example cases are presented and discussed. The example cases demonstrate the efficacy of our approach.