Real-time Water Wave Simulation Based On Grid-particle

Recently, with the development of computer technoloy, it is possible to real-time simulation of complicated scene, which is one of hot research areas in computer graphics. Water wave simulation is an active topic in the field.Up to date, there are three main types of algorithms for simulating water wave. One is grid-based geometry modeling method, which uses trigonometric functions or quadratic surfaces, and tunes parameters for surface modeling. This method is sensitive to parametric settings and not suitable for interaction. The second method is physically-based modeling on method, which describes movements of all internal particles by solving the Navier-Stokes equation. The simulation results obtained by this method are ideal due to using particle system. However, owing to dealing with numerous particles, the computation cost is so high that it is unable to render wave in real time.The last one is wave spectrum inversion method based on observation results of ocean. This method is very similar to the first method, but the parameter setting totally depends on long-time ocean observation result. It can generate effects is like thereal environment when only applied to simulate the calm ocean.Based on current research achievements and methods, This paper presents a new simulation method, named height-map-based single lay-region particle modeling, by combining grid with particle system and, which simulates water wave in two phases: First, generates a basic fluid surface with the height field; Then, according to the location of the waves, self-adaptively produces particles and maps the height of them into the corresponding grids. Our algorithm can be used to efficiently simulate water waves in many fluid circumstances, including waves formed by single click, by sweeping something over the water surface, by ship motion and wind driven water wave. The experimental results demonstrate that our method is able to simulate water waves in real time and fulfils the requirement of human-computer interaction.Since our algorithm can efficiently reduce the computation time through avoiding non-interaction field, and it's easy to implement and has flexible control ability. Thus, it can be applied to many simulating systems.