Using a Metamodel-Based Approach for Optimization of Stimulated Rock Volume Geometry, Hydrocarbon Production, and Related Field Development Costs

Pourpak, Hamid (Total SA) | Will, Johannes (Dynardo) | Eckardt, S. (Dynardo) | Mottet, Nicolas (Total SA)



The success of an unconventional hydrocarbon development depends on effective hydraulic fracturing, which highly depends on reservoir properties and the stimulation procedure. In the beginning of shale development, the industry practice was to conduct a large number of field trials, which was a very expensive and time consuming practice. Advanced integrated studies are being performed today by industry on rock and SRV (Stimulated Rock Volume) characterization, however the topic remains still vastly challenging, due to the complex nature of fracturing in shales as well as because of the complexity of multiple physics and the number of operational parameters involved in shale development from formation characterization to SRV creation and production.

This project used a numerical simulation approach, based on truly 3D reservoir modelling of fracture network generation and stimulation, to optimize hydrocarbon production through the investigation of a large number of virtual well stimulations. Starting with a calibration workflow, taking into account potential reservoir and geomechanical uncertainties, a calibrated reservoir model was built. The goal of this approach is to find the optimal stimulation parameters much faster with much less investments compared to the industry standard of simply undertaking a trial and error well drilling and completion process. The reservoir model calibration, the multi-realization runs, together with the metamodel analyses have been performed using a workflow and a range of advanced software tools developed since 2010 (Bai et al, 2011, Gao et al, 2011, Yeh et al, 2018).

In this study, the proposed workflow was applied to a major Unconventional Oil and Gas field. A multi-stage hydraulic fracturing operation has been modelled and calibrated based on the data from a real hydraulic fracture shale gas operation. Geomechanical and geological uncertainties have been taken into account in the calibration process of the reservoir model. Furthermore, microseismic monitoring results and fracture treatment pressure data have been used to calibrate important parameters in the fracturing modelling process. Utilization of multi-realization runs while scanning parameters uncertainties, enabled to rank the parameters influencing the stimulated rock creation process.

In a second step, a sensitivity study has been performed within a predefined window of variation of operational parameters. From this sensitivity study, important operational parameters influencing fracture network geometry and related hydrocarbon production have been identified. Based on the sensitivity study, meta models were then generated which represent the influence of the variation of operational parameters on fracture network geometry and hydrocarbon production. The meta models have been combined with costs to optimize operational parameter taking into account the conflicting nature of EUR, NPV, VIR. The results of this meta model-based optimization may help improve the decision-making process of hydraulic fracturing operations and shale play development, including unit development costs and unit profitability.