A Neural-Genetic Technique for Coastal Engineering: Determining Wave-induced Seabed Liquefaction Depth

Abstract

In the past decade, computational intelligence (CI) techniques have been widely adopted in various fields such as business, science and engineering, as well as information technology. Specifically, hybrid techniques using artificial neural networks (ANNs) and genetic algorithms (GAs) are becoming an important alternative for solving problems in the field of engineering in comparison to traditional solutions, which ordinarily use complicated mathematical theories. The wave-induced seabed liquefaction problem is one of the most critical issues for analysing and designing marine structures such as caissons, oil platforms and harbours. In the past, various investigations into wave-induced seabed liquefaction have been carried out including numerical models, analytical solutions and some laboratory experiments. However, most previous numerical studies are based on solving complicated partial differential equations. In this study, the proposed neural-genetic model is applied to wave-induced liquefaction, which provides a better prediction of liquefaction potential. The neural-genetic simulation results illustrate the applicability of the hybrid technique for the accurate prediction of wave-induced liquefaction depth, which can also provide coastal engineers with alternative tools to analyse the stability of marine sediments.