Article ID Journal Published Year Pages File Type
787838 International Journal of Pressure Vessels and Piping 2012 9 Pages PDF
Abstract

In the present work an analytical method is developed to study the mechanical response of thick composite pipes under internal pressure. The present method considers that the surface tension must satisfy, on the interface between two adjacent layers, all the boundary conditions imposed on the radial and hoop stresses, which are computed using a specific model for thick multi-layered composite cylinders suggested by Tsai. This model uses the plane strain states of cylindrical tubes under internal pressure containing a number of cylindrical sub-layers, each of which is cylindrically orthotropic. Such analytical method is considered as simple, fast and precise compared to other analytical and numerical methods. Thereafter, the effects of random design variables on the pipe behavior are quantified by probabilistic analysis. Using these stress components for each ply, the Monte Carlo method is used to predict the distribution function of the mechanical response, in addition to the failure probability. Five random design variables are considered, namely, the composite elastic constants, the ply angles, the inner and outer diameters, thickness and the mechanical loading. The improvement of multi-layered filament-wound composite pipes design based on the probabilistic analysis is also discussed. According to the obtained results, we note that the thickness of the pipe and the internal pressure variation are the principal factors influencing the scatter of stress distribution.

► Effect of the pipe thickness on the hoop stress variations is evaluated. ► Mechanical response of thick composite pipes under internal pressure is studied. ► Effect of random design variables on the pipe is quantified by probabilistic analysis. ► Monte Carlo method is used to predict the failure probability distribution.

Related Topics
Physical Sciences and Engineering Engineering Mechanical Engineering
Authors
, , ,