TY - JOUR
T1 - Multi-objective optimumdesign of double-layer perforated-wall breakwaters
T2 - Application of NSGA-II and bargaining models
AU - Nikoo, Mohammad Reza
AU - Varjavand, Iman
AU - Kerachian, Reza
AU - Pirooz, Moharram D.
AU - Karimi, Akbar
PY - 2014/8
Y1 - 2014/8
N2 - Double-layered perforatedwall (DLPW) breakwaters have mostly been studied by limited experimental methods due to complexities in their mathematical modeling. In this paper, a methodology based on data-driven simulation modeling, multi-objective optimization and game theory is proposed for optimum design of DLPW breakwaters. This methodology converts data to model and model to optimum decision in terms of design variables. Experimental results are converted to data-driven simulation models based onANFIS (adaptive neuro-fuzzy inference system), which can model the relationship between wave parameters, DLPW design variables and performance criteria considering inherent uncertainties in the experiment and results. The ANFIS-based simulation models are trained and validated based on experimental results of a physical DLPW breakwater. The simulation models are included in a powerful multi-objective optimization algorithm, Non-dominated sorting genetic algorithm-II (NSGA-II), to determine the trade-off between wave transmission and reflection as two important criteria in efficiency assessment of the DLPW breakwaters. Finally, two efficient multi-person decision-making models, namely unanimity fallback bargaining (UFB) and condorcet social choice method, are utilized to find the best agreed-upon design point on the trade-off curve from wave transmission and reflection point of view. The results indicate that the proposed methodology can be effectively used for determining the optimal characteristics of DLPW breakwaters.
AB - Double-layered perforatedwall (DLPW) breakwaters have mostly been studied by limited experimental methods due to complexities in their mathematical modeling. In this paper, a methodology based on data-driven simulation modeling, multi-objective optimization and game theory is proposed for optimum design of DLPW breakwaters. This methodology converts data to model and model to optimum decision in terms of design variables. Experimental results are converted to data-driven simulation models based onANFIS (adaptive neuro-fuzzy inference system), which can model the relationship between wave parameters, DLPW design variables and performance criteria considering inherent uncertainties in the experiment and results. The ANFIS-based simulation models are trained and validated based on experimental results of a physical DLPW breakwater. The simulation models are included in a powerful multi-objective optimization algorithm, Non-dominated sorting genetic algorithm-II (NSGA-II), to determine the trade-off between wave transmission and reflection as two important criteria in efficiency assessment of the DLPW breakwaters. Finally, two efficient multi-person decision-making models, namely unanimity fallback bargaining (UFB) and condorcet social choice method, are utilized to find the best agreed-upon design point on the trade-off curve from wave transmission and reflection point of view. The results indicate that the proposed methodology can be effectively used for determining the optimal characteristics of DLPW breakwaters.
KW - Double-layer perforated-wall breakwater
KW - Fallback bargaining
KW - Multi-objective optimization
KW - NSGA-II
KW - Social choice method
UR - http://www.scopus.com/inward/record.url?scp=84898998696&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84898998696&partnerID=8YFLogxK
U2 - 10.1016/j.apor.2013.12.001
DO - 10.1016/j.apor.2013.12.001
M3 - Article
AN - SCOPUS:84898998696
SN - 0141-1187
VL - 47
SP - 47
EP - 52
JO - Applied Ocean Research
JF - Applied Ocean Research
ER -