Numerical simulation of pipe-soil interaction during pulling back phase in horizontal directional drilling installations. (June 2018)

• Record Type: Journal Article
• Authors: Xu, Guangli; Cai, Liangxue; Ji, Rui; Wang, Zhongliang
• Is Part Of: Tunnelling and underground space technology. Volume 76(2018)
• Page Start: 194
• Page End: 201
• Publication Date: 2018-06
• Subjects: Horizontal directional drilling -- Pipe -- Soil -- Interaction -- Numerical simulation; Tunneling -- Periodicals; Underground construction -- Periodicals; Tunnels -- Periodicals; Underground areas -- Periodicals; 624.193
• DOI: 10.1016/j.tust.2018.03.022
• Languages: English
• ISSNs: 0886-7798

Abstract

Highlights: The wedging effect in HDD installations can amplify resistances significantly. The effect of several factors on the wedging coefficient is studied. Prediction accuracy of pulling forces is improved by considering wedging effect. Abstract: The displacement of pipe-soil contact points in the radial direction and the wedging effect of surrounding soil on the pipe perimeter during pipe pulling-back may have a substantial effect on predicting pulling forces for horizontal directional drilling installations. However, the influence of pipe-soil interaction on pulling force predictions is not considered in existing methods to predict the required pulling force. This paper provides a model representing the interaction between pipe and borehole. The mechanical characteristics of the pipe-soil interaction were numerically investigated with program ANSYS. Two parameters, the wedging coefficient and pipe displacement, were defined and associated parametric studies were performed to analyze the influences of relative factors based on a Yangtze River crossing project. The wedging coefficient is shown to increase with external load and ovality of the borehole, and decrease with increasing over-cut ratio, Poisson's ratio and elastic modulus of soil. It can improve the prediction accuracy to consider the wedging effect since the difference among predicted results caused by wedging effect can differ up to 7.7% in example analysis.