Comparison of Artificial Neural Networks and Genetic Algorithms for Predicting Liquid Sloshing Parameters

• Published in: Journal of marine science and application Vol. 23; no. 2; pp. 292 - 301
• Main Authors: Saghi, Hassan, Nezhad, Mohammad Reza Sarani, Saghi, Reza, Sahneh, Sepehr Partovi
• Format: Journal Article
• Language: English
• Published: Berlin/Heidelberg Springer Berlin Heidelberg 01.06.2024; Springer Nature B.V; Department of Civil Engineering,Hakim Sabzevari University,Sabzevar,Iran%Department of Electrical and Computer Engineering,University of Birjand,Birjand,Iran%State Key Laboratory of Coastal and Offshore Engineering,Dalian University of Technology,Dalian,116024,China%Department of Marine Engineering,Amirkabir University of Technology,Tehran,Iran
• Subjects: Algorithms; Artificial neural networks; Boundary conditions; Electrical Machines and Networks; Engineering; Finite element method; Free surfaces; Genetic algorithms; Geotechnical Engineering & Applied Earth Sciences; Laplace equation; Liquid sloshing; Machinery and Machine Elements; Mathematical analysis; Offshore Engineering; Power Electronics; Research Article; Sloshing loads; Artificial neural network; Fluid structure interactions; Genetic algorithm; Potential flow analysis
• ISSN: 1671-9433; 1993-5048
• DOI: 10.1007/s11804-024-00413-6

This paper develops a numerical code for modelling liquid sloshing. The coupled boundary element-finite element method was used to solve the Laplace equation for inviscid fluid and nonlinear free surface boundary conditions. Using Nakayama and Washizu’s results, the code performance was validated. Using the developed numerical mode, we proposed artificial neural network (ANN) and genetic algorithm (GA) methods for evaluating sloshing loads and comparing them. To compare the efficiency of the suggested methods, the maximum free surface displacement and the maximum horizontal force exerted on a rectangular tank’s perimeter are examined. It can be seen from the results that both ANNs and GAs can accurately predict η max and F max .