Experimental and numerical investigations of axial pullout behavior of buried ductile iron pipe

Abstract

Buried pipelines often cross active landslide areas, which are subjected to additional loads due to ground movements. The effect of ground loads on the performance of buried pipelines is an important consideration for pipeline integrity assessment. Though experimental and analytical studies were conducted to understand the maximum loads caused by the ground movement on the pipelines, design practices for the assessment of pipes subjected to ground movements are not yet well developed. This thesis presents the design of a new laboratory test facility developed for pullout testing of buried pipelines that investigate the behavior of pipelines subjected to axial ground movements. The test facility is first assessed using finite-element modelling to identify the effects of the size of the facility and the rigidity of the boundary walls on the pullout study of pipelines. The results show that a test cell having the dimensions of 2 m (width) x 1.5m (height) x 4 m (length) is adequate for the current purpose of tests if the wall stiffness is adequately designed. The facility is then used to conduct a pullout test of a 178-mm diameter ductile iron pipe. During the test, horizontal and vertical earth pressure in the soil is measured using Tekscan pressure sensors. Pipe deformation and wall strains are measured using strain gauges. A finite element modelling technique is developed to simulate the test conditions for investigating the soil-pipe interaction during the axial pullout test. The finite element model is then employed to study the pipe-soil interaction mechanism. The study reveals that arching effect and dilation of sand in the pipe–soil interface can affect the mobilized soil load on the ductile iron pipe. The unit interface shear resistance is found to be 16.0H to 17.0H in dense sand, 13.0H in medium dense sand and 5.0H in loose sand for the pipes tested.