Enhancing PEHD pipes reliability prediction: Integrating ANN and FEM for tensile strength analysis

Srii Ihssan,Nagoor Basha Shaik,Naoual Belouaggadia,Mustapha Jammoukh,Alanssari Nasserddine
DOI: https://doi.org/10.1016/j.apsadv.2024.100630
2024-08-10
Applied Surface Science Advances
Abstract:In the pipe industry, pressure pipes have long made use of High-Density Polyethylene (HDPE), which is used extensively. Currently, HDPE pipes are installed in higher numbers in comparison with other plastic pipes. The purpose of this study is to evaluate and compare the predictive capabilities of two methods, including the finite element method (FEM) and artificial neural network (ANN) techniques, for predicting the tensile strength of HDPE pipes used in water distribution systems. Attempts have been made to improve prediction models to better predict the mechanical behavior of these pipes by improving our understanding of the structure and surface characteristics as well as the interactions between the interface and the operating environment. The results show that experimental trial results are in perfect agreement with machine learning techniques. The findings of this study highlight the benefits of using ANN to predict the behavior of HDPE pipes, which may have significant ramifications for the plastics and water distribution industries.
English Else
What problem does this paper attempt to address?