Development of procedure for structural safety assessment of energy saving device subjected to nonlinear hydrodynamic load. (1st April 2016)

• Record Type: Journal Article
• Authors: Lee, DongBeom; Jang, Beom-Seon; Kim, HyunJin
• Is Part Of: Ocean engineering. Volume 116(2016)
• Page Start: 165
• Page End: 183
• Publication Date: 2016-04-01
• Subjects: ESD (Energy saving device) -- CFD (Computational fluid dynamics) -- Neural network -- Long-term analysis -- Gumbel distribution; Ocean engineering -- Periodicals; Ocean engineering; Periodicals; 620.4162
• DOI: 10.1016/j.oceaneng.2016.02.038
• Languages: English
• ISSNs: 0029-8018

Abstract

Abstract: There have been many attempts worldwide to reduce ships׳ fuel consumption. With fuel-efficiency improvement as their designed purpose, a variety of energy saving devices (ESD) have been developed and deployed. Much research with respect to ESD performance already has been carried out; however, the issue of ESD structural safety has received relatively little attention. According to the current approach to ESD structural safety assessment, the Morison equation, which is calculated for a certain probability-level velocity, is applied, or alternatively, a spectral method based on the assumption of a linear system between wave and response is utilized. Unfortunately, this methodology does not take into account the nonlinearity of hydrodynamic loads. Therefore, a new ESD structural safety assessment procedure that utilizes the neural network and time-domain simulation using the Gumbel fitting method is herein proposed. The procedure entails four main steps: sea-keeping analysis, hydrodynamic load analysis and neural network, ultimate strength analysis, and fatigue strength analysis. The important features of the proposed procedure are, in order, as follows. First, to consider the nonlinearity of hydrodynamic force acting on the ESD, computational fluid dynamics (CFD) analysis is carried out on samples consisting of various wave heights and periods. The neural network is then trained based on the CFD analysis results for the prediction of hydrodynamic loads. Second, to take into account the randomness of the peak hydrodynamic force, a three-hour time-domain simulation is repeated 20 times for each sea state of a wave-scatter diagram, and Gumbel parameters are calculated for long-term analysis. Third, approximate long-term analyses using a contribution coefficient and short-term analysis are performed for an efficient long-term analysis. Highlights: No proper procedure for an evaluation of structural safety of an energy saving device. A new procedure of strength assessment of an energy saving device subject to nonlinear hydrodynamic load. Time domain simulation is simplified using neural network and CFD simulation. Long-term extreme of hydrodynamic load is estimated by Gumbel fitting method. Application of approximate long-term analysis using contribution coefficient and short-term analysis is examined.