Leakage in bolted pipe flange connections, particularly those involving materials with dissimilar stiffness like high-density polyethylene (HDPE) and steel, is a persistent challenge in infrastructure systems. This study investigates a computational framework for evaluating and improving the leak-tightness of HDPE flange assemblies by employing the tetra-parametric assembly method (TAM), which iteratively calculates non-uniform bolt load distribution. The approach accounts for key time- and temperature-dependent phenomena, including viscoelasticity and stress relaxation in HDPE, using a calibrated non-linear three-network (TN) constitutive model. Finite element analysis (FEA) is used to simulate the long-term mechanical response of HDPE connections under various isothermal conditions (23 ◦ C, 40 ◦ C, 60 ◦ C, and 80 ◦ C), revealing increased susceptibility to leakage at elevated temperatures due to loss of bolt preload. To reflect real-world operating conditions, an annual temperature profile representative of an above-ground piping system was applied to the model, enabling prediction of leak-tightness over a full year of service. The results highlight the importance of accounting for thermal and mechanical degradation over time and inform several bolt re-torquing strategies to mitigate leakage risks. The proposed framework offers practical guidelines for optimizing HDPE flange connection performance and provides engineers with a robust tool for improving the reliability of thermoplastic piping systems.
Ezzeldin et al. (2026) studied this question.