Optimizing well-stimulation treatment size using mixed integer linear programming

Abstract

This thesis presents a model for optimum sizing of well stimulation treatments. The optimization scheme is based on a combination of a well stimulation type selection and treatment size, therefore determining the optimum type of stimulation with an optimum treatment size for a given vertical well located in the center of a square drainage area. Although it considers a well producing under pseudosteady-state or stabilized flow, the optimum stimulation treatment will result in the best option even during transient flow. The model evaluates three alternatives: (1) no stimulation, (2) matrix acidizing, and (3) hydraulic fracturing. It was implemented into a Mixed Integer Linear Programming (MILP) problem and is solved through the well-known MILP algorithm, Branch-and-Bound (BB). Our approach integrates the relationship between well performance, reservoir, and proppant characteristics, considering the effect of treatment size both on well performance and treatment cost through the unifying concept of proppant number (N[] ). The model was programmed and evaluated in Mosel language. A review of field data indicates that the results of our optimization scheme are realistic. The results suggest that this approach can be a useful tool for practicing "a-priori" optimization for a stimulation treatment sizing. An appropriate application of this optimization scheme might represent potential improvements of at least 10-20% in terms of profit.