Optimum Fluid and Proppant Selection for Hydraulic Fracturing in Shale Gas Reservoirs: a Parametric Study Based on Fracturing-to-Production Simulations

摘要

Production from shale gas reservoirs depends greatly on the efficiency of hydraulic fracturing treatments. The cumulated experience in the industry has led to several best practices in treatment design which have improved productivity in these reservoirs. However, further advances in treatment design require a deeper understanding of the complex physics involved in both hydraulic fracturing and production, such as stress shadow, proppant placement and interaction with natural fractures. This paper investigates the non-linear physics involved in the production of shale gas reservoirs by improving the understanding of the complex relation between gas production, the reservoir properties, and several treatment design parameters, with a focus on proppant and fluid selection. A fracturing-to-production simulation workflow integrating the Unconventional Fracture Model, with the Unconventional Production Model is presented. This workflow has shown qualitative consistency with real production data. In this paper we applied the workflow on a realistic reservoir with characteristics from the Marcellus play, and then studied the relation between production and treatment design parameters such as proppant size, proppant concentration, the treatment volume of the treatment, fracturing fluid viscosity, pumping rate and proppant injection sequence. Since this paper focuses on fluid and proppant selection, our methodology was to run 28 simulations to cover the 2D parametric space of proppant size and fluid viscosity for every parameter. More than four hundred simulations were run in this parametric study and the results provide guidelines for optimized treatment design. The behaviors observed confirm several best practices in treatment design for shale. For example, combination of different sizes of proppant optimizes production by maximizing initial production and slowing down production decline. Simulations also confirm the best practice of injecting the smallest proppant first. Another key finding is that the optimum fluid viscosity increases with treatment volume, and decreases when pumping rate increases.