Fuel consumption of ship has been of great interest due to the reinforced environmental regulations. Estimation of fuel consumption requires calculating the added resistance in actual irregular waves. However, most of relevant researches remain in regular wave since nonlinear interaction is hard to be included in numerical computation. In this paper, the added resistance in irregular head sea has been investigated using Reynolds-averaged Navier-Stokes based Computational Fluid Dynamics. The results were compared to those of the model tests conducted in Samsung Ship Model Basin. In the simulation, the irregular waves were generated by the linear superposition of a number of incoming wave components. Since the computation times are highly increased when a large number of waves are used, total time window was divided into a number of partitions, and irregular waves were continuously generated by overlapping the neighboring windows. Three degrees of freedom were considered for ship's motion: heave, pitch and surge. Motion responses from the computation show fairly good agreement to those from the model test. In addition, the simulation predicts the added resistance in the similar level of accuracy to the experiment. The stepwise analysis is made and key findings are discussed.


Recently, due to regulations of Energy Efficiency Design Index (EEDI), energy conservation and reduction of CO2 emissions have emerged as a major interest for ship design. In general, a ship is designed for specific speed and draft based on her resistance and propulsion in calm sea. However, when a ship advances in a seaway, ship experiences added resistance defined as an increased resistance due to environmental influence relative to that in calm sea. Since the added resistance result in a speed loss to the sailing ship, the precise prediction of that is required to achieve the economical sailing of the vessel.

Traditionally, added resistance in real sea has been treated by the spectral analysis employing the transfer function and sea spectrum, for example in the sea trial analysis (ISO, 2015). In this sense, when the transfer function of added resistance is estimated in regular waves, total amount in real sea is estimated by the moment of response spectra. Most of added resistance research has been focused on calculating transfer function accurately at regular waves. Theoretical and numerical method has been developed as like panel method (Jonquez et al., 2009; Kim and Kim, 2011; Söding et al., 2014; Pan et al., 2016), Reynolds- averaged Navier-Stokes (RANS) based Computational Fluid Dynamics (CFD) (Orihara and Miyata, 2003; Guo et al, 2012; Sadat-Hosseini et al., 2013; Hu et al, 2014; Yang and Kim, 2016) and empirical formula (Fujii and Takahashi, 1975; Faltinsen et al., 1980; Tsujimoto et al., 2008; ISO, 2015). Such methodology has shown their validity especially in moderate sea condition. However, their approach is inherently not complete since the concept of transfer function cannot reflect the wave-body interaction dependent on sea environment. For example, it has been reported from numerical and experimental studies that even the transfer function is dependent on the wave steepness of regular wave (Tsujimoto et al., 2008; Kashiwagi, 2013; Yasukawa et al., 2016; Kim et al., 2017; Lee et al., 2017). In addition, nonlinear interaction in irregular sea is hard to be defined by the linear superposition of responses at each regular wave. Although necessity of direct estimation is obvious, a few studies have been performed due to the limitation of basin facility and computation capacity. Kobayashi (2007) and Kuroda et al. (2016) tried to construct the time histories of added resistance in irregular sea by the approximation using response in regular wave. Despite of the partial satisfactory in mild sea, discrepancy is inevitable in harsh waves. Yasukawa et al. (2016) conducted model test in regular waves at two different wave heights as well as in irregular wave. Their results clearly demonstrate that spectral summation is not complete to estimate the added resistance in actual irregular sea appropriately as well as the transfer function is dependent on wave height. Dalzell (1974), Hirayama and Wang (1993) also showed that transfer function obtained in regular waves does not coincide with that indirectly estimated from irregular and transient waves. Consequently, added resistance needs to be verified in irregular wave directly and thus active study is consistently requested.

This content is only available via PDF.
You can access this article if you purchase or spend a download.