Neural network for the estimation of the inelastic buckling pressure of loosely fitted liners used for rigid pipe rehabilitation

Khaled M. El-Sawy, Abdel Latif Elshafei

Research output: Contribution to journalArticlepeer-review

4 Citations (Scopus)

Abstract

One of the structural design aspects of most of the liners is to check their stability under external uniform pressures. This requires the definition of the critical pressure at which the liner destabilizes. A neural network based on the results of a previous parametric study using the Finite Element Method (FEM) is developed. The neural network provides an estimate for the critical pressure of an elasto-plastic loosely fitted liner. The inputs for the network are the liner's thickness-to-radius, gap-to-radius, and the equivalent yield stress-to-Young's modulus ratios. The network results are checked against the FE results and compared to Jacobsen solution. The results of the neural network show excellent agreement with the FE results and Jacobsen solution for thick liners. This network provides a new tool that can be used in the structural design of loosely fitted liners.

Original languageEnglish
Pages (from-to)785-800
Number of pages16
JournalThin-Walled Structures
Volume41
Issue number8
DOIs
Publication statusPublished - Aug 2003

Keywords

  • Inelastic
  • Loose liner
  • Neural network
  • Rehabilitation
  • Rigid pipes

ASJC Scopus subject areas

  • Civil and Structural Engineering
  • Building and Construction
  • Mechanical Engineering

Fingerprint

Dive into the research topics of 'Neural network for the estimation of the inelastic buckling pressure of loosely fitted liners used for rigid pipe rehabilitation'. Together they form a unique fingerprint.

Cite this