井漏
井筒
钻探
欠平衡钻井
石油工程
钻井液
地质学
孔隙水压力
断裂(地质)
压力(语言学)
岩土工程
有限元法
圆筒应力
压力梯度
工程类
结构工程
机械工程
语言学
哲学
海洋学
作者
Yongcun Feng,Cagdas Arlanoglu,Evgeny Podnos,E. B. Becker,K. E. Gray
摘要
Summary In the drilling of deep horizons, the mud-weight window between pore-pressure gradient and fracture gradient often narrows because of rock properties and underground-stress state. Nonproductive-time events such as lost circulation, wellbore instability, kicks, underground crossflow, and pipe sticking are more likely. Such problems greatly increase drilling costs. Bridging pre-existing natural fractures or drilling-induced fractures with lost-circulation materials (LCM) is often performed to increase fracture gradient and widen the mud-weight window. This technique, referred to as wellbore strengthening, includes the stress cage, fracture-closure stress, and fracture-propagation-resistance methods. Although these methods are often used, several aspects of these approaches are still not thoroughly understood. To reduce the risk of lost circulation while drilling in formations with narrow mud-weight windows, such as pressure-depleted reservoirs and deepwater formations, a good understanding of the mechanisms of wellbore strengthening in different downhole scenarios helps engineers to optimize the design of drilling fluids and operational procedures. This paper discusses the mechanism of wellbore strengthening, with a focus on hoop stresses at and near the wellbore, in elastic and poroelastic models, by use of the finite-element method to evaluate wellbore and near-wellbore stresses during fracture creation and propagation, and after plugging fractures with LCM. Factors affecting fracture behavior, such as horizontal-stress anisotropy, LCM-bridging location, initial pore pressure, and fluid leakoff, are investigated. A better understanding of the several interacting events local to the wellbore and near-wellbore regions can result in improved operational practices related to lost-circulation prevention and mitigation.
科研通智能强力驱动
Strongly Powered by AbleSci AI