Please use this identifier to cite or link to this item:
|Title:||Data driven modeling and simulation for TBM reliability analysis in tunnels||Authors:||Yeong, She Chan||Keywords:||Engineering::Civil engineering||Issue Date:||2021||Publisher:||Nanyang Technological University||Source:||Yeong, S. C. (2021). Data driven modeling and simulation for TBM reliability analysis in tunnels. Final Year Project (FYP), Nanyang Technological University, Singapore. https://hdl.handle.net/10356/150453||Project:||CT-03||Abstract:||Tunnel-induced damage has become one of the most critical problems in the world, due to the rapid growth of urban metro systems. Tunnel boring machines (TBMs) have become increasingly common in recent years. To perform single objective optimization and multi-objective optimization for minimizing tunnel-induced damages under uncertainty, a hybrid approach combining fuzzy cognitive map (FCM) with real-coded genetic algorithm, and FCM with non-dominant sorting genetic algorithm-II (NSGA-II) is proposed. Using RCGA on a data-driven modelling approach, FCMs are learned from historical datasets. There are 16 identified concept nodes and 2 key objectives, building tilt rate and accumulative settlement, are used for optimization and to find the optimal solutions among them. To assess the applicability and efficacy of the suggested strategy, in this study, TBM parameters are extensively explored. Results shows that (1) grouting volume have the highest influence on both building tilt rate and accumulative settlement. (2) A combination of concept nodes will give a better optimization result. (3) Multi-objective optimization with a 13.01% improvement is better overall to achieve the optimal solutions than single objective optimization with 6.66% improvement. The established method is effective that it is capable of not only predicting the severity of tunnel-induced losses, but also allows stakeholders to minimize them when multiple objectives are optimized at the same time.||URI:||https://hdl.handle.net/10356/150453||Fulltext Permission:||restricted||Fulltext Availability:||With Fulltext|
|Appears in Collections:||CEE Student Reports (FYP/IA/PA/PI)|
Updated on Jan 20, 2022
Items in DR-NTU are protected by copyright, with all rights reserved, unless otherwise indicated.