Characterization of Multi-Scale Discrete Fracture Network Systems in Unconventional Reservoirs Using Dynamic Data

近年来，由于世界范围内的大量资源估计，非常规致密或页岩油藏的开发受到了极大的关注。钻水平井进行多级水力压裂，以增加油层接触面积，达到经济产量的目的。如果由于现有裂缝的存在，水力裂缝可以连接到甜蜜点，这种增产技术就特别有效。致密/页岩储集层可以概念化为具有两个不同子系统的双重介质：基质和裂缝。由于渗透率的高对比度，两个系统中流动的时间尺度有很大的不同。基质的原地渗透率很低，而裂缝是渗透率很高的局部不连续面。裂缝可以是位于井筒周围的大裂缝(如水力裂缝)，也可以是远离刺激区的小裂缝(如微裂缝)。需要注意的是，这项建议侧重于原地开发页岩/致密油/气藏，而不考虑油页岩的地面开采。 总体目标是开发一套适合于(1)使用基于有限元的技术模拟离散裂缝中的多相流；(2)捕捉不同尺度上发生的裂缝的影响；以及(3)整合动态数据(例如，流动和压力测量)以在稳健的历史匹配工作流程中表征裂缝网络的一套数值模型。 这项研究提出了一种新的分析低渗透多孔介质中多相渗流和流体分布的综合方法，其中裂缝网络，即水力裂缝和微裂缝，在多个尺度上同时存在。这项研究将彻底改变目前主要侧重于分析模型和有限体积/有限差分模拟的做法，这些模型通常忽略多相效应，而有限体积/有限差分模拟不处理与致密/页岩储层相关的裂缝配置的复杂性。这种模型质量的损失使得数值模拟中对裂缝系统的描述不完整，从动态数据推断裂缝特征具有挑战性。许多工业从业者熟悉非常规油藏开发的最新研究，他们表示需要更复杂的方法来分析和模拟这些油藏。随着计算技术的快速进步，预计有利于当前实践的计算负载正在消失，更多的计算密集型技术，如本提案中开发的技术，可能在不久的将来在商业代码中实现。这一结果将影响我们优化压裂设计的能力，这也是地热能源开发中的一个重要考虑因素。

Development of unconventional tight or shale reservoirs has received much attention in recent years because of the large resource estimates worldwide. Horizontal wells with multi-stage hydraulic fracturing are drilled to enhance reservoir contact area and to achieve economic production rates. This stimulation technique is particularly efficient if hydraulic fractures can connect to sweet spots due to the presence of existing fractures. A tight/shale reservoir can be conceptualized as a dual-medium with two distinct sub-systems: matrix and fracture. The time scales for flow in the two systems differ significantly due to the high contrast in their permeability. The matrix has a very low in-situ permeability, while the fracture is a local discontinuity with a very high permeability. Fractures can be large and localized around the wellbore (e.g., hydraulic fractures) or small and located away from the stimulated zone (e.g., micro fractures). It should be noted that this proposal focuses on in-situ development of shale/tight oil/gas reservoirs, while surface mining of oil shale is not considered. The overall objective is to develop a suite of numerical models suitable for (1) simulating multi-phase flow in discrete fractures using finite-element based techniques; (2) capturing the impacts of fractures occurring over different scales; and (3) integrating dynamic data (e.g., flow and pressure measurements) for characterizing fracture networks in a robust history-matching workflow. This research proposes a novel integrated approach for analyzing multi-phase flow and fluid distribution in low-permeability porous media, where fracture networks, i.e., hydraulic fractures and micro fractures, are present over multiple scales simultaneously. This research would revolutionize the current practice that focuses predominantly on analytical models, which typically ignore multi-phase effects, and finite-volume/finite-difference simulations, which do not handle the complexities in fracture configuration pertinent to tight/shale reservoirs. This loss of model quality renders the description of fracture systems in numerical simulation incomplete and the inference of fracture characteristics from dynamic data challenging. Many industrial practitioners, who are familiar with the state-of-the-art research in unconventional reservoir development, have expressed the need for more sophisticated approaches to analyze and model these reservoirs. It is expected that computational loads that favor current practice are disappearing with the rapid advances in computing technology, and more computationally-intensive techniques, such as those developed in this proposal, could be implemented in commercial codes in the near future. The outcomes would impact our capability to optimize fracturing design, which is also an important consideration in geothermal energy development.