Theoretical and numerical models for the influence of debonding on the interaction between hydraulic fracture and natural fracture

Abstract

Purpose

The interaction between hydraulic fracture (HF) and natural fracture (NF) in naturally fractured rocks is critical for hydraulic fracturing. This paper aims to focus on investigating the development of tensile and shear debonding zone on the NF caused by the stresses produced by HF, and the influence of NF’s debonding behavior on the interaction between HF and NF.

Design/methodology/approach

Theoretically, tensile and shear debonding modes of NF are considered, two dimensionless parameters are proposed to characterize the difficulty of tensile and shear failure of NF, respectively. Numerically, a finite element model combining the extended finite element method and cohesive zone method (CZM) is proposed to study NF’s debonding behavior and its influence on the interaction between HF and NF.

Findings

Both theoretical analysis and numerical simulation show the existence of two debonding modes. The numerical results also show that the HF can cross, offset or propagate along the NFs depending on the parameters’ value, resulting in different fracture network and stimulated reservoir volume. When they are large, the NF’s debonding area is small, HF tends to cross the NF and the fracture network is simple; when they are small, the NF’s debonding area is large, HF will propagate along the NF. In addition, HF is easier to propagate along with NF under tensile debonding mode while it is easier to pass through NF under shear debonding mode.

Originality/value

The theoretical and numerical considerations are taken into account in the influence of the debonding of NFs on the interaction between HFs and NFs and the influence on the formation of the fracture network.

Citation

Wang, T., Liu, Z., Gao, Y., Ye, X. and Zhuang, Z. (2019), "Theoretical and numerical models for the influence of debonding on the interaction between hydraulic fracture and natural fracture", Engineering Computations, Vol. 36 No. 8, pp. 2673-2693. https://doi.org/10.1108/EC-07-2018-0290