Research On Numerical Wave Tank Based On Pressure Fluctuate Reducing MPS Method

In the Computational Fluid Dynamics, the flows including large deforming free surface have been studied for a long time. The main numerical simulation methods in CFD are grid method and meshless method and meshless method have an advantage on resolving flow problem with free surface. Moving-Particle Semi-Implicit Method (MPS) is a new meshless numerical method which bases on Lagrange particle, which trace the changing feature of dynamic and kinematic. Due to the advantage in free surface simulation, the MPS method plays more and more important role in numerical research.?Although MPS have a good result in simulation with free surface, the pressure fluctuation restricts the application in engineering. People had some study on reduce the pressure fluctuation, but they did not get some good results. This paper focuses on the reason on the pressure oscillation, therefore combine the pressure Poisson equation with divergence-free condition which is usually used to ensure the incompressibility of flows. There are two kinds of cases being numerical simulated which are water collapse and ship roll sloshing. Comparing to the original MPS method, the result from the new method is real, so that the new method has good practicability and reliability.Wave is one of the most important dynamic factors in the Naval architecture and Ocean Engineering and Coastal Engineering. It is used to establish model to solve such problem. Numerical wave tank had become a main way to study waves. It can not only save huge amout of time than physical model, but also have more expansibility on scale and controllability. A numerical wave tank has been build successfully based on ship model test towing tank in SJTU.Large Eddy Simulation(LES) is a new method proposed to solve turbulent flows which is common in the world and it has good effect. LES is used in the new numerical wave tank to generate wave numerically. The constract has been made based on different kernel function and KF1 is suitable for numerical wave generation. This paper combines LES with new modified MPS to generate wave numerically to reduce the pressure oscillation successfully. At last, the influence of relaxation factor to the numerical wave has been studied based on KF1 kernel function.