Wei Zhao

Date of Award

Spring 2003

Document Type


Degree Name

Doctor of Philosophy (PhD)


Materials and Infrastructure Systems

First Advisor

David Hall


Thin-walled plastic liners are routinely used to rehabilitate structurally sound host pipes that have lost their hydraulic integrity. The liner/host pipe structure is treated as a ring encased in a rigid wall. Significant research work dealing with liner stability has been completed in North America and Europe. These research works include laboratory experiments, numerical analysis, and statistical modeling. The objective of this dissertation is to achieve a better understanding of liner buckling phenomenon using the finite element method and statistical analysis.

Short-term liner buckling models have been applied by several researchers to study the influence of liner geometric imperfections on the buckling pressure. Two-dimensional, plane-strain models were predominantly used in these studies. In this research, a three-dimensional liner buckling model was constructed using ABAQUS. This model was applied in studying the influence of liner thickness variations and local circular defects on the buckling pressure.

At first, the thickness variation of the liner wall was modeled in the longitudinal and circumferential directions separately. Next, liners with thickness variations in both directions were simulated. Two factors, the frequency and the magnitude of thickness variation, were studied quantitatively. A comparison between the buckling pattern and thickness variation pattern was conducted to investigate the relationship between the buckling and thickness.

Local circular defects of reduced thickness and reduced flexural modulus were also modeled using the three-dimensional model. Three aspects, the frequency, the size, and the magnitude of defects were studied with respect to their influence on buckling pressure.

Long-term buckling experiments of liners have shown a large amount of scatter when the buckling time is plotted against the applied pressure. The presence of this scatter and other uncertainties are typically accounted for in liner design by applying a factor of safety. Moreover, an additional “factor of safety” is sometimes implied since the design equation in ASTM F1216 is generally considered to be conservative. The scatter observed in recent long-term, liner buckling experiments conducted at the Trenchless Technology Center was studied to associate reliability factors with selected confidence levels. This work resulted in a set of reliability factors that can be directly applied to the ASTM design equation for the partially deteriorated case. The reliability factors allow a designer to quantitatively estimate the influence of observed scatter on liner design and provide the designer with confidence that their design does not fall within the region of scatter observed in liner buckling experiments. Two separate statistical methods were applied to same set of experimental data, and the results were compared.