Simulation of Breaking Waves Using Particle Level Set Method

• Main Authors: Chun-Yuan Lin, 林俊遠
• Other Authors: Ching-Jer Huang
• Format: Others
• Language: en_US
• Published: 2007
• Online Access: http://ndltd.ncl.edu.tw/handle/03609055381094367485

博士 === 國立成功大學 === 水利及海洋工程學系碩博士班 === 95 === In present study a two-dimensional numerical wave tank in viscous fluid was developed and applied to simulate the propagation of water waves over a submerged breakwater. The unsteady, two-dimensional Reynolds Averaged Navier-Stokes (RANS) equations and the turbulence model ( model) were solved for simulating the realistic fluid. A hybrid particle level set method was adopted to capture the evolution of the highly complex free surface. A piston-type wavemaker was also set up in the computational domain to produce the desired incident waves. In this model the governing equations were discretized by means of a finite-analytical scheme. The SIMPLER algorithm was used to calculate the coupled velocity and pressure fields. The evolution of level set method was solved using forth-order TVD Runge-Kutta method and fifth-order WENO scheme, and the accuracy was confirmed by solving the Zalesak’s problem. Two major subjects were discussed in present study. First, to understand the characteristics of the complicated wave fields as a periodic wave train propagated over a submerged breakwater without breaking, present research proposed a new technique to separate incident and reflected higher harmonic waves using four or more spatially separated probes. Both the free and locked modes in the higher harmonics of the regular waves can be isolated. The complex waves are decomposed into individual frequency components using the Fourier transform. The least squares method is applied to minimize the error caused by possible signal noise and to obtain the equations for solving the unknown parameters related to the wave amplitudes. Probe spacing condition for preventing singularity in the calculation is provided. The accuracy of this method is verified by applying it to resolve artificial waves with arbitrarily selected amplitudes. The sensitivity of this method to the noise inevitably associated with the experiments is also tested. The free surface elevation data collected from probes located upstream and downstream of a submerged breakwater in a numerical wave tank are analyzed to demonstrate the applicability of this method. The results are compared with those obtained using the methods of Goda and Suzuki (1976) and Mansard and Funke (1980). The comparison indicated that this method gives exactly the same first harmonic incident and reflected wave amplitudes as the other two methods. The full modes in the higher harmonics can be determined using this method, while the higher harmonics are presumed to be free waves in the other two methods. Second, as the wave breaking, the internal kinetics of fluid flow became complicate and was difficult to measure in laboratory. In order to investigate the kinematics properties of overturning waves, present numerical model was performed to simulate the internal velocity and its corresponding surface evolution as the water wave passed over the submerged breakwater. The evolution of a solitary wave overturning on a submerged breakwater was in good agreement with the experimental data of Yasuda et al. (1997). According to the numerical results, the maximum flow velocity increased gradually during the wave breaking process until the third times of reattachment occurred. To realize the energy loss during wave breaking, the energy translation between kinetic energy and potential energy was discussed systematically with difference scales of submerged breakwater at several significant stages including the shoaling process, re-attachment and splash-up stages. The numerical results showed that the energy loss increased intensively at the stage of re-attachment. Besides, the location of the breaking point moves from the downstream of the breakwater to be above the breakwater, as the breakwater height increases; and the wave breaking type changes from the spilling breaker to plunging breaker and then collapsing breaker.