A hybrid surrogate-assisted integrated optimization of horizontal well spacing and hydraulic fracture stage placement in naturally fractured shale gas reservoir

Abstract

Horizontal well drilling and hydraulic fracturing are the most frequently adopted technologies for the commercial development of shale reservoirs. However, expensive production strategies are still implemented which results in high risk in investment. Integrated optimization of horizontal well spacing and hydraulic fracture stage placement helps in striking a balance between gas production and economic benefits. Previous studies relied heavily on numerical simulation models which are computationally expensive. In this study, a novel hybrid surrogate-assisted shale gas horizontal well spacing and hydraulic fracture stage placement multi-objective optimization method based on transfer stacking (SATS-WSF) is proposed to lessen the computational burden of the numerical simulation model run. In the SATS-WSF method, three widely used machine learning models, namely Gaussian process regression (GPR), radial basis function network (RBFN), and support vector regression (SVR) were applied to approximate the numerical simulation model as the source tasks. Then, a hybrid surrogate model transferring the three source tasks to the computationally expensive numerical simulation model was adopted to guide the optimization process of shale gas horizontal well spacing and hydraulic fracture stage placement. In addition, two sampling infill strategies called promising and uncertain were used to accelerate the searching process and to improve the quality of the final optimal solutions. Furthermore, two cases with different wells and fracture types based on the Barnett shale gas reservoir properties were employed to verify the effectiveness and efficiency of the SATS-WSF method. This method provides an intelligent approach for efficient decision-making of shale gas well space and fracture scheme.