Steam-Sensitive Flow-Control Device Implemented in Surmont SAGD Project

_ This article, written by JPT Technology Editor Chris Carpenter, contains highlights of paper SPE 230259, “Field Implementation of a New Inflow-Control-Device Design in Surmont I SAGD,” by Matthew French, SPE, and Alex Colleaux, SPE, ConocoPhillips, and Marco Melo Llanos, SPE, Baker Hughes, et al. The paper has not been peer-reviewed. _ This paper introduces a steam-sensitive flow-control device designed to restrict the production of steam and low-subcool liquids while allowing higher mobility of oil-phase fluids. By preferentially limiting flow from zones with high steam saturation or low subcool, the device helps retain steam energy within the reservoir, directs heat to colder regions, minimizes sandface erosion, and supports the growth of a more-uniform steam chamber. This paper presents the first field implementation of the featured inflow control device (ICD) in the Athabasca McMurray oil-sands reservoir, specifically within the Surmont I steam-assisted gravity drainage (SAGD) operation. Challenges of ICD Use in SAGD Operations At first adoption by SAGD operators, available ICDs were not designed specifically for such applications and, in some cases, were poorly suited for the operating conditions. Moreover, the modeling methods currently in use have significant limitations that hinder accurate ICD design and understanding. While these models can often be history-matched, they typically fail to provide meaningful insights into the actual pressure drops encountered or to support evaluation of the benefits of different ICD types. The two main limitations are, first, grid-related constraints that prevent accurate modeling of fine-scale gridding, and, second, difficulty in capturing steam breakthrough because of local sandface-subcooling concerns. The use of ICDs in the Surmont SAGD project began with trials in 2008, gaining attention after a 2012 report showed a 33% increase in oil production and a 20% improvement in the steam-to-oil ratio (SORI) (Fig. 1). Although ICDs have shown potential for improving recovery in SAGD production wells, field results have been inconsistent. Vendors often design ICDs using varying approaches that do not always account for the behavior of the liquid pool, leading to cases wherein no performance improvement could be justified. These shortcomings were largely the result of limited understanding of the specific design conditions or the absence of tools tailored for SAGD applications. In previous studies, the effectiveness of steam-trap or subcool control was assessed based on production rates without accounting for actual steam flow. Consequently, earlier simulation models—where the well grid is connected directly to the reservoir grid—assumed that steam could only be produced once the producer was fully exposed to the steam chamber, predicting zero steam production before full contact. However, in reality, steam can be produced even without full exposure because of flashing, which leads to localized steam or vapor breakthrough. In this study, the flashing mechanism is extended to ICDs, enabling its integration into reservoir and wellbore simulations. The complete paper presents a framework to evaluate existing ICDs and to guide the design and optimization of future ICDs for SAGD applications. Analysis of results using diagnostic charts and quantitative parameters specifically developed for SAGD and other thermal recovery processes is essential. Much of the complete paper is devoted to a discussion of a diagnostic dedicated to the variation of pressure-drop ratio vs. production rate for ICDs.