Research Article  Open Access
Gang Xu, TianRui Mei, MingLiang Hu, Zhen Chen, JunMing Hu, "Hydrodynamic Analysis of Ship with Well Deck in the Linear Numerical Wave Tank", Mathematical Problems in Engineering, vol. 2020, Article ID 3804748, 15 pages, 2020. https://doi.org/10.1155/2020/3804748
Hydrodynamic Analysis of Ship with Well Deck in the Linear Numerical Wave Tank
Abstract
In recent years, the development and construction of islands and reefs has been proposed by the government and commercial company. However, as a large cargo carrier cannot reach islands and reefs if the harbor is not available, such type of carrier which has well deck is designed to meet the requirements of delivering people and equipment. It is a possible way to connect the island and supply cargo ships. This paper firstly summarizes the domestic and foreign research progress of hydrodynamic analysis of ships with well deck. Then, based on the CFD (Computational Fluid Dynamics) tools, we set up a linear numerical wave tank and study the hydrodynamic performance of original WigleyIII ship and modified WigleyIII ship with well deck. The hydrodynamic effect of the floating body in the well deck has been investigated and discussed.
1. Introduction
Recently, with the development of computer technology, various numerical modeling tools are becoming possibility. Compared with the physical model test, the study period of the numerical method is short, the cost is small, and the resource occupancy is low. Moreover, the traditional design mode has been changed by the numerical method and its process is becoming simple.
Longuethiggins and Cokelet [1] used the Euler–Lagrange method to simulate plunging breaker based on the idea of direct solution of boundary integral. Baker et al. [2] added the artificial damping viscous term to the boundary of a specific free surface in the same way as the sponge absorbed wave energy and presented the concept of damped wave elimination region in the twodimensional boundary element mode for the first time. Vinje and Brevig [3] proposed the modal decomposition method, which makes the numerical wave tank be capable of dealing with hydrodynamic calculation and motion simulation of floating bodies. Based on completely nonlinear free surface boundary, Jin and Zou [4] established a new vertical twodimensional wave equation by taking wave surface rise and free surface velocity potential as the variables to be determined. Yan et al. bib5 used the twopoint method to separate the wave morphology of each component wave after the submerged body and established a twodimensional fully nonlinear numerical model to simulate the wave effect on a submerged horizontal cylinder based on the indomain wavemaking technology. Fang et al. [6] used the mixed finite difference and finite volume numerical schemes to solve the twodimensional fully nonlinear distributed water wave equation and established the fully nonlinear Boussinesq wave propagation numerical model based on the MUSTA scheme. Zhang et al. [7] established a twodimensional viscous numerical water trough based on the CIP method, which can be used to simulate linear, weakly nonlinear, and strongly nonlinear waves. Li and Ning [8] simulated and studied the deformation of long wave train propagation and the energy exchange between waves of various orders.
Bai and Taylor [9–11] analyzed the phenomenon of fully nonlinear regular wave and focusing wave, respectively, on vertical cylindrical diffraction and the effect of completely nonlinear wave on fixed and floating structures. Li et al. [12] used highorder spectral methods to simulate threedimensional fifthorder deepwater Stokes waves to verify accuracy and convergence. Chen et al. [13], based on the CIP method for the gasliquid twophase flow numerical model, carried out numerical simulation of the propagation and deformation of solitary waves on coastal slopes. Wu et al. [14] studied the effects of wavelength, water level, effective wave height, and friction resistance on the propagation deformation of nearshore waves.
Many effective methods have been derived for the numerical simulation of the interaction between waves and structures. Presently, the hydrodynamic analysis of ships with well deck is based on the viscous flow theory, which can realize the numerical simulation of the coupled response of ship motion and flow field generated in well deck.
Hopman et al. [15] realized the test of two ship models (with good performance and newly designed) through a typical model test program, made the model in different modal time periods, and simulated the ship model test in a series of wave directions and speeds under two irregular spectral conditions. Subsequently, Bass et al. [16] used the commercial CFD integration code with ship motion code (MOTSIM, a 3D nonlinear time domain program) to simulate the wave interaction in the wetted well deck and verified the numerical simulation results with the model test. Cartwight et al. [17] used SPH (Smoothed Particle Hydrodynamics) method to simulate 2D landing craft containing well deck and explore the influence of welldeck design on landing craft movement. Yoon and Seo [18] conducted model tests on landing craft with different welldeck positions and studied their seakeeping and safety, respectively.
In this work, a simplified ship model is used to verify the effectiveness of the numerical tank. Based on the published WigleyIII ship model data, the numerical simulation of ship diffraction in waves is conducted based on the CFD method. Firstly, the heave and pitch motions of the ship and the modified ship are simulated and compared with the test data of DUT (Delft University of Technology) [19, 20]. Secondly, the heave and pitch motion of the WigleyIII ship model are simulated. Finally, an attempt is made to simulate the motion of well deck with floating structures, so as to provide an effective technical scheme for relevant study.
2. Numerical Method
2.1. Governing Equation and Calculation Algorithm
Continuous equation and Navier–Stokes equation were applied as the governing equation of an incompressible flow as follows:where U is the velocity, which has three components in x, y, and z direction, respectively, is the density, t is the time, is the coordinate, is the pressure, is the kinetic viscosity, is the turbulent stress, and is the acceleration due to gravity. ICEM CFD and ANSYS FLUENT are used for grid generation and for the numerical simulation, respectively. FLUENT is based on the Finite Volume Method (FVM), in which governing equation has been discretized by the FVM.
The Computational Fluid Dynamics (CFD) package, FLUENT, has been successfully utilized to solve the fluid and fluidstructure interactions, and it is used as a basic solver with User Defined Functions (UDF) in this work. To ensure accurate results, we adopt the following settings. (1) The pressure staggering option (PRESTO) discretization scheme is used to solve the pressure. (2) The quadratic upwind interpolation of convective kinematics (QUICK) algorithm is used to discretize the momentum. (3) The pressure implicit with splitting of operators (PISO) algorithm is used for the pressure velocity coupling. (4) The volume of fluid (VOF) method is adopted to capture the water surface. (5) The RNG kε model (2eqn) is used to simulate the flow of the incompressible and viscous fluid.
To deal with the wave and structure interaction in the Numerical Wave Tank (NWT), the work adopts the C programming language into the basic solver through UDF. The UDF is used to deal with the wavegenerating boundary conditions in the inlet boundary, the wave absorbing zone in the end of the Numerical Wave Tank.
2.2. Boundary Conditions
Generally, the top boundary of the computation domain is taken as a pressure inlet, the bottom boundary of the computation domain is the wall, which depends on the depth, and the left and rightside boundaries of the computation domain are set as symmetry. The setting of the left inlet wave boundary of the wavegenerating area and the rightend boundary of the waveeliminating area needs to be adjusted according to the calculation, as presented in Figure 1. The other boundaries are also adjusted accordingly to the calculation which can be performed stably.
3. Numerical Wave Tank and Its Validation
3.1. Numerical Model and Mesh Discretization for NWT
The calculation domain is designed as a length of 25 m, a width of 2 m, and a depth of 3 m. The distance between calm water surface and pressure inlet boundary is 1 m and the water depth is 2 m. The end of tank is the wave elimination area, which is 10 m, as shown in Figure 1.
In Figure 1, G1, G2, G3, and G4 are the wave elevation probe, which is located at x = 2.5 m, x = 5 m, x = 10 m, and x = 25 m to monitor the corresponding wave height, respectively.
The model is meshed by ICEM CFD module of ANSYS. In the longitudinal direction of the main computational domain, the meshes are distributed equally along the wave propagation direction; however, in the waveeliminating region, the meshes gradually become bigger, which can reduce the total number of grids, the operation time, and can improve the calculating efficiency, meanwhile which can result in numerical dissipation. In the vertical direction, according to the water wave theory, the grid is encrypted near the water surface, in order to capture the accurate position of the free surface and improve the numerical accuracy. The place, which is far away from the free surface, will become sparse gradually. In the transverse direction, the grid is divided equally, as shown in Figure 2.
3.2. Convergent Study for NWT
In the numerical solution, mesh discretization and time interval are the most important factors which affects the simulation accuracy, and the numerical solutions have to be stable with the improvement of mesh encryption skill and time interval. Thus, the convergence analysis of the numerical wave tank is carried out and the wave parameter for the simulation is described in Table 1.

The convergence study of time interval has been carried out firstly and has been presented in Figure 3. It can be obtained that convergent results of wave height with different incident wavelength can be found while time interval is less than 0.01 s.
We also consider the study of the mesh density, which is represented by the number of cells Nx in the longitudinal direction within each wavelength and Nz in vertical direction per wave height. We can find that good and convergent results can be obtained when Nx is great than 50 and Nz is great than 25, as shown in Figures 4 and 5. Nx = 50 and Nz = 30 will be applied for the following simulations, and the snapshot of free surface can be found in Figure 6.
We can also find the time history results of wave elevation at different probes (x = 2.5, x = 5, and x = 10) and good agreements can be obtained by comparing the present CFD results and theoretical value, as shown in Figures 7–9, respectively. We can conclude that the present tank can be used for long time simulations. From comparison, we can also find the nonlinear phenomenon because of the nonlinear governing equation. It has typical nonlinear effect that the peaks and troughs of waves show asymmetric.
The efficiency of wave elimination can be found in Figures 10 and 11. High efficiency waveeliminating result can be obtained in the present model as the waveeliminating technology has been involved in the end of the tank.
4. Wave Diffraction of WigleyIII Ship Model
4.1. Numerical Model for Wave Diffraction of WigleyIII Ship Model
4.1.1. Calculation Model of WigleyIII
The model for this calculation is WigleyIII ship model [19, 20]. Journée evaluated the motion of the four kinds of Wigley through the experimental analysis. This paper will use the WigleyIII model, and the main dimensions are shown in Table 2.

In Cartesian coordinate system , the origin of coordinates is located on the steady water in the middle of the WigleyIII ship and it can be described mathematically by equation (2). Figures 12 and 13 display the exact shape of the ship in 2D and 3D, respectively:
4.1.2. Boundary Conditions
The inlet boundary is a velocity inlet, which is defined by UDF. When the wave passes through the wave elimination area, the right end should be stationary; thus, the right end boundary can be set as pressureoutlet. The top of the tank is set as pressureoutlet boundary condition. The bottom and the surface of WigleyIII are set as wall boundary conditions. Both sides of the numerical wave tank are set as symmetrical boundary conditions.
4.2. Numerical Simulation and Discussion of Wave Diffraction Case
The test data can be used to verify the simulation results in the numerical wave tank. Journée [20] provided the data of the Wigely model in regular wave, which will be adopted for the comparison in this work.
4.2.1. Numerical Model Setting
In this paper, numerical simulation is carried out by using the WigleyIII ship model under the incident wave and with forward speed. The calculation parameters are shown in Table 3.

4.2.2. Numerical Results with Incident Wave Direction β = 180°
Figures 14 and 15 show the comparison of the hydrodynamic force amplitude in longitudinal direction () and the coefficient of the corresponding force (). and can be calculated by the following equation:
(a)
(b)
(a)
(b)
By comparing the experimental data, it can be found that the CFD simulation results in this work are in good agreement with the experimental data. Figures 16 and 17 show the heave force and its coefficient and their comparison with the test data.
(a)
(b)
(a)
(b)
The numerical results are in good agreement with the experimental results and the heave force of the ship model in the case of heading wave increases with the wavelength, and the relative error is very small.
Figures 18 and 19 show the pitch force and its coefficient under the incident heading wave condition. It can be seen that the calculated results in this work are in good agreement with the experimental data.
(a)
(b)
(a)
(b)
4.2.3. Numerical Results at Incident Wave Direction β = 150°
From the numerical results of wave force in the previous section, the CFD results in this paper agree well with the test data. The method adopted in this paper can simulate the ship sailing in heading waves. However, the sea state is random including wave direction and wave period, and we need to consider the typical wave angle such as β = 150° and 3D BEM [21] results will be used for comparison.
Figure 20 shows the coefficient of exciting force on the WigleyIII ship model under six degrees of freedom. As we can see that the dimensionless coefficient of force in waves are in good agreement with the results of the potential flow theory. For the long wave simulation, the CFD results in this paper are slightly larger than that of potential flow theory. The reason [22, 23] is that it has a serious inaccuracy in the flow field near the wave surface when the potential flow is used to solve the velocity problem. It has been solved by Duan et al. and the technology has been used in China Numerical Tank.
5. Hydrodynamic Study of WigleyIII with Floating Body in the Well Deck
5.1. Numerical Model for Case with Floating Body in the Well Deck
5.1.1. Modified WigleyIII Ship Model
The public data for the ship with well deck cannot be found in the database, so we set a ship with well deck artificially. In order to analyze the effect of floating body, the model is based on the original WigleyIII ship model. A well deck of 0.15 m width and 0.2 m height was built at the stern of the ship, named the modified WigleyIII ship model. Details can be found in Figure 21. Note that the shape of the present ship model is quite different from that of the actual ship with well deck, and further study should be executed by a more appropriate ship model.
5.1.2. The Model with Small Floating Body in the Well Deck
Based on the modified WigleyIII ship, a small floating structure has been put into the well deck to analyze its effect and multibody interaction. The floating body in the well deck is simplified with a rectangular floating object for the convenient calculation. The model is shown in Figure 22 and the parameters can be found in Table 4.

5.2. Numerical Results and Discussion
5.2.1. Numerical Model Setting
When the well deck is opened and the floating body exits in the well deck, the hull is mostly at zero forward speed; thus, this part focuses on the motion of the hull without forward speed. In the numerical wave tank, the WigleyIII ship model, the modified WigleyIII ship model, and the modified WigleyIII ship model containing the floating body in the well deck has been calculated and compared. The parameters for the simulation are given in Table 5.

5.2.2. Numerical Results with Heading Waves β = 180°
The hull, encountering different incident heading waves, has been calculated. The obtained results are shown in Figures 23 and 24. It can be seen that the numerical results are consistent with the experimental data.
In Figure 23, the results of the modified WigleyIII ship model has a similar trend with the original WigleyIII ship model, and it can be clearly seen that the heave coefficient of the modified WigleyIII ship model is reduced when the wavelength is greater than the length of the ship. It means that the well deck is good for the heave motion.
For the pitch, as shown in Figure 24, the trend of the ship model with the well deck changes significantly with the short waves. The amplitude of pitch motion increases significantly when λ/L = 0.5. We can also find that the well deck has disadvantage for the pitch motion. Through comparing the motions, it is found that the floating body in the well deck has small effect on the overall motion of the ship.
5.2.3. Numerical Results at Incident Wave Direction β = 45°
Figures 25 and 26 show the heave and pitch of the ships at incident wave direction β = 45°. The coefficient of them has been presented in Figures 27 and 28. It can be seen that the motions of ships tend to be consistent and have same trend. Figure 27 shows that the heave motion of the ships increases with the wavelength, and the amplitude of the heave motion increases, and becomes closer to a stable value. Compared to the original WigleyIII ship, the pitch motion of the modified WigleyIII ship model becomes more obvious when the wavelength is greater than the length of the ship and the amplitude of pitch motion is reduced, as shown in Figure 28. We may conclude that the floating body does not have serious effect on the modified WigleyIII ship. Later, we may focus on the motion of the floating body in the well deck in the high sea state.
6. Conclusion
Based on CFD (Computational Fluid Dynamics) tools, this paper uses FLUENT to simulate the WigleyIII ship model in threedimensional linear numerical wave tank and the modified WigleyIII ship with well deck also has been investigated. Through the calculation we have conclusions as below:(1)Based on the Navier–Stoke governing equation, the VOF method was used to capture the water surface, and the RNG turbulence model was used to solve the numerical wave tank. The effects of mesh discretization and time interval have been presented and discussed. We found that the time interval has much greater effect on the accuracy of numerical simulation when the mesh discretization is fixed. We can obtain accurate results when regular wave propagates in the numerical wave tank by using the present model.(2)As the public data for ships with well deck is not available, we modified the stern of the original WigleyIII ship model. The numerical simulation for the original WigleyIII ship and modified WigleyIII ship with the well deck has been carried out, and the numerical results are in good agreement with the experimental data for wave diffraction problem and the motion of the modified model is similar to that of the original ship.(3)We also found that the floating body did not have significant effect on the modified WigleyIII ship with the well deck. For the further study, we should consider the ship under the high sea state and predict the motion of the floating body in the well deck.
Data Availability
The data used to support the findings of the study are available from the corresponding author upon reasonable request.
Conflicts of Interest
The authors declare that they have no conflicts of interest.
Acknowledgments
This work was supported by the National Natural Science Foundation of China (Grant nos. 51709137 and 51879125), Key University Science Research Project of Jiangsu (Grant no. 18KJA130001), Marine Equipment and Technology Institute of Jiangsu University of Science and Technology Funding Project (Grant no. HZ20190922), and Postgraduate Research and Practice Innovation Program of Jiangsu Province (Grant nos. KYCX18_2343 and SJCX19_1187).
References
 M. S. LonguetHiggins and E. D. Cokelet, “The deformation of steep surface waves on water i. a numerical method of computation,” Proceedings of the Royal Society A: Mathematical, Physical and Engineering Sciences, vol. 350, no. 1660, pp. 1–26, 1976. View at: Publisher Site  Google Scholar
 G. R. Baker, D. I. Meiron, and S. A. Orszag, “Application of a generalized vortex method to nonlinear free surface flows,” in The proceedings, Third International Conference on Numerical Ship Hydrodynamics, pp. 179–191, Paris, France, June 1981. View at: Google Scholar
 T. Vinje and P. Brevig, “Nonlinear ship motions,” in Proceedings of the International Conference on Numerical Ship Hydrodynamics, Paris, France, June 1981. View at: Google Scholar
 H. Jin and Z. Zou, “Hyperbolic mild slope equations with inclusion of amplitude dispersion effect: regular waves,” China Ocean Engineering, vol. 22, no. 3, pp. 431–444, 2008. View at: Google Scholar
 B. Yan, H. Lin, and D. Ning, “Nonlinear numerical simulation of the effect of waves on submerged horizontal cylinders,” Marine Sciences Progressive, vol. 31, no. 3, pp. 319–325, 2013. View at: Google Scholar
 K. Z. Fang, L. Wang, Z. B. Liu, and Z. L. Zou, “A fully nonlinear boussinesq model for wave propagation based on MUSTA scheme,” Chinese Journal of Theoretical & Applied Mechanics, vol. 46, no. 5, pp. 647–654, 2014. View at: Google Scholar
 D. H. Zhang, J. D. Wang, and G. H. He, “Numerical simulation of strongly nonlinear waves,” Proceeding of National Symposium on Hydrodynamics, vol. 1, 2015, in Chinese. View at: Google Scholar
 X Li and D. Ning, “Simulation study on propagation characteristics of long wave train,” Hydrodynamics Research and Progress (Series A), vol. 33, no. 6, pp. 714–719, 2018. View at: Google Scholar
 W. Bai and R. E. Taylor, “Higherorder boundary element simulation of fully nonlinear wave radiation by oscillating vertical cylinders,” Applied Ocean Research, vol. 28, no. 4, pp. 247–265, 2006. View at: Publisher Site  Google Scholar
 W. Bai and R. E. Taylor, “Numerical simulation of fully nonlinear regular and focused wave diffraction around a vertical cylinder using domain decomposition,” Applied Ocean Research, vol. 29, no. 12, pp. 55–71, 2007. View at: Publisher Site  Google Scholar
 W. Bai and R. E. Taylor, “Fully nonlinear simulation of wave interaction with fixed and floating flared structures,” Ocean Engineering, vol. 36, no. 34, pp. 223–236, 2009. View at: Publisher Site  Google Scholar
 J. X. Li, S. X. Liu, and K. Y. Hong, “Numerical simulation of nonlinear waves,” Journal of Dalian University of Technology, vol. 48, no. 3, pp. 430–435, 2008. View at: Google Scholar
 Y. Chen, Z. H. Fang, and X. Z. Zhao, “Numerical simulation of offshore wave evolution,” in Proceedings of the 17th China Ocean (Coastal) Engineering Symposium, vol. 1, Nanning, China, 2015, in Chinese. View at: Google Scholar
 X. W. Wu, W. B. Liu, and T. C. Liu, “Numerical study of wave propagation and deformation characteristics in offshore waters,” Science Technology and Engineering, vol. 18, no. 33, pp. 126–131, 2018. View at: Google Scholar
 H. Hopman, G. K. Kapsenberg, and E. M. Krikke, “Design and hydromechanic aspects of the amphibious transport vessel for the royal Netherlands navy,” Naval Engineers Journal, vol. 106, no. 3, pp. 163–174, 1994. View at: Publisher Site  Google Scholar
 D. Bass, D. Molyneux, and K. McTaggart, Simulating Wave Action in the Well Deck of Landing Platform Dock Ships Using Computational Fluid dynamics, Warship, Toledo, OH, USA, 2004.
 B. Cartwright, M. Renilson, G. MacFarlane, D. McGuckin, and S. Cannon, “Motions of a landing craft in a flooded well dock—effect of well dock design,” in Proceedings of the International Conference on Military Support Ships, London UK, November 2007. View at: Google Scholar
 S. H. Yoon and K. H. Seo, “Well dock design and assessment of relative motions during the operation of the landing crafts within well dock,” Journal of the Society of Naval Architects of Korea, vol. 49, no. 2, 2012. View at: Publisher Site  Google Scholar
 J. M. J. Journée, “Experiments and calculations on four wigley hullforms,” in Society of Naval Architects and Marine Engineers, Society of Naval Architects and Marine Engineers Headquarters, Jersey City, NJ, USA, 1992. View at: Google Scholar
 J. M. J. Journée, Experiments and Calculations on 4 Wigley Hull Forms in Head waves, Delft University of Technology, Delft, Netherlands, 1992.
 Z. Z. Fang, R. C. Zhu, G. P. Miao, and C. Gong, “Numerical simulation of diffraction problems of moving vessels in numerical wave tank,” Journal of Shanghai Jiaotong University, vol. 46, no. 8, pp. 1203–1217, 2012. View at: Google Scholar
 G. Xu, B. C. Khoo, and A. M. S. Hamouda, “Timedomain simulation of secondorder irregular wave diffraction based on hybrid water wave radiation condition,” Applied Mathematical Modelling, vol. 40, no. 78, pp. 4451–4467, 2016. View at: Publisher Site  Google Scholar
 W. Duan, J. Chen, and B. Zhao, “Secondorder taylor expansion boundary element method for the secondorder wave diffraction problem,” Engineering Analysis with Boundary Elements, vol. 58, pp. 140–150, 2015. View at: Publisher Site  Google Scholar
Copyright
Copyright © 2020 Gang Xu et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.