Settlement-based framework for long-term serviceability assessment of immersed tunnels

• Published in: Reliability engineering & system safety Vol. 228; p. 108801
• Main Authors: Tang, Cong, He, Shu-Yu, Zhou, Wan-Huan
• Format: Journal Article
• Language: English
• Published: Barking Elsevier Ltd 01.12.2022; Elsevier BV
• Subjects: Algorithms; Bayesian analysis; Bayesian approach; Calibration; Computer applications; Elastic foundations; Evolutionary algorithms; Evolutionary computation; Hong Kong-Zhuhai-Macao tunnel; Immersed tunnel; Long-term serviceability; Markov chains; Mathematical models; Reliability; Reliability analysis; Reliability engineering; Robustness (mathematics); Safety; Settlement; Settlement analysis; Structural damage; Tunnel construction; Tunnels; Settlement; Bayesian approach; Long-term serviceability; Immersed tunnel; Reliability; Hong Kong-Zhuhai-Macao tunnel
• ISSN: 0951-8320; 1879-0836
• DOI: 10.1016/j.ress.2022.108801

•A framework for long-term serviceability assessment of immersed tunnel is proposed.•Two serviceability failure modes are considered.•A novel beam on elastic foundation model is developed for settlement estimation.•The Bayesian method with the DE-TMCMC algorithm is used to qualify uncertainties.•The proposed framework is verified by field data from the HZM tunnel. In immersed tunnels, the considerable settlement that can develop during their long-term service period may induce structural damage that affects normal operations (i.e., serviceability performance). The long-term serviceability assessment from the perspective of settlement is thusly crucial to ensure the safety of immersed tunnels, whereas a practical assessment framework is lacking to date. This study presents a feasible reliability analysis framework for the long-term serviceability assessment of immersed tunnels for the first time, in which a novel beam on elastic foundation model (BEFM) is developed as a mathematical model for tunnel settlement analysis. The BEFM is easy to be integrated into the probabilistic framework due to its simple modeling procedure and high computational efficiency. The Bayesian approach with an efficient and robust algorithm, differential evolution transitional Markov chain Monte Carlo (DE-TMCMC) algorithm is adopted to guarantee effective model calibration and uncertainty quantification. The application potential of the proposed framework is demonstrated by a case study of the Hong Kong-Zhuhai-Macao (HZM) immersed tunnel. Results show that the developed method can provide a feasible and effective assessment of the long-term serviceability of immersed tunnels, offering a practical tool to enhance the reliability and safety of immersed tunnels during their operation periods.