On the Fracture Height Migration Under Unified Fracture Design Optimization

Authors:

Abstract
We have published already a three-layer p-3D fracture design optimization using the Unified Fracture Design (UFD) approach; fracture height, resulting from fracture geometry optimization and therefore net pressure, was matched against the value resulting from fracture height migration at the same pressure. For a multi-layered formation a more robust and more general approach is suggested in this work. First a range of admissible reference treating pressures is prepared. For each pressure level we determine the up-down tip locations and hence we obtain the fracture height map. Thus, there is a relationship between induced height and net pressure and this relationship can be calculated from Linear Elastic Fracture mechanics before the actual design optimization is carried out. According to the applicable fracture propagation model, for a given fracture half length, there exists a width and henceforth a net pressure. The ancillary height is a geometric byproduct. Matching the net pressure and height with the LEFM independently determined pair of height and net pressure results in a solution that satisfies both. The dimensionless productivity index, JD, is then calculated. For the fracture half lengths, ranging from a minimum (corresponding to aspect ratio one with respect to the perforated thickness) to a laterally fully penetrating fracture, we repeat the above procedure to calculate JD and so forth, until a maximum JD is accomplished. Introduction Fracture geometry optimization is a key to advance hydraulic fracture design. Traditional means to predict fracture geometry include: Two-dimensional (2D) models, Perkins, Kern and Nordgren (1961), (the "PKN?? model); Khristianovitch and Zheltov, Geertsma and De-Klerk (1955), (the "KGD?? model) and the Radial or "Penny-shape?? Model. Three-dimensional (3D) and pseudo-3 dimensional (p-3D) models, such as the "Cell?? approach, (Cleary 1994); overall fracture geometry parameterization and meshed full 3D models (Johnson et al., 1993) are also available. While a 2D model focuses on the fracture length and width by assuming a value for fracture height, the 3D and p-3D models attempt to predict height along with length and width (Gidley et al., 1979). For design purposes where the expected fracture geometry must result in the optimized fracture performance, Pitakbunkate (2011) used the Unified Fracture Design approach and extended it to a p-3D model in a three-layer lithology reservoir, predicting fracture height, half-length and width at the end of the fracture job. The present work has primarily focused on incorporating a multi-layered, p-3D, hydraulic fracture propagation model into the systematic design of a fracturing treatment for both oil and gas reservoirs. There are a number of considerations in optimizing fracture dimensions to maximize the production. The reservoir deliverability, well producing systems, fracture mechanics, fracturing fluid characteristics, proppant transport mechanism and operational constraints should be considered and integrated to achieve the physical optimization of the system. Incorporating economics in the design would maximize the benefits of a stimulation treatment. One drawback in the work of Pitakbunkate et al. 2011 by using a three-layer reservoir lithology is that some constraints are needed to ensure a solution, especially if the fracture height appears to "run away?? in certain conditions.

This publication has 18 references indexed in Scilit: