Real-Time Cloud Simulation Based On Physical Models

Clouds are a ubiquitous, ever-changing feature of outdoors. The simulation of natural phenomena such as clouds, smoke, fire and water is one of the most important research areas in computer graphics. Physically based simulation of natural phenomena has been a hotspot and one of the most difficult tasks in Compute Graphics. In particular, clouds play an important role in creating images of outdoor scenes.With the development of computer graphics, recently for the simulation of clouds, people more focus on the dynamic modeling and rendering based on the physics reality, and the whole results considering the interactive effects between the clouds and the environment.This paper focuses on the study in four aspects. First, derived physical models of clouds simulation accurately based on cloud physics and CFD. Second, I realized the physically-based cumulus simulation on GPU (Graphics Processing Unit). GPU takes advantages of the parallelism and the programmability of graphics hardware to accelerate the calculation. Third, I improved the control method of clouds animation for more realistic. Last, a roamed 3D sky background was realized. The results showed that the visual effect of 3D cloud scene is true to life and real-time.1,Derived physical models of clouds simulation.Clouds have an amorphous structure without well defined surfaces and boundaries. The complexity of the physical computation often leads to serious performance degradation. There is no uniform method to render clouds. So it is important to investigate on the physical models of clouds, find its moving discipline.Firstly, this paper presents a survey on the development of physically based clouds simulation in computer animation. Then studies and analyzes the fluid motion, thermodynamic processes, buoyant forces, and water phase transitions, based by which, I derived the clouds simulation equation. Then I presented the computation of the cloud simulation equation which is physically-based, visually-realistic. The process to solve this equation likes the arithmetic to simulate fluids of which the core methods are semi-Lagrangian and partial differential equations (PDE). The difference between the two computation methods is that some physical quantity about clouds is commutated in cloud simulation equations.I also simulate the interaction of clouds with light, including self-shadowing and single forward light scattering. Cloud is one kind of participating media. Light transported through a cloud is affected by emission, in-scattering, absorption, and out- scattering. The scattering in clouds is too complex to model as well as losing computation performance, so I simplify the radiance equation, only take forward single scattering into consideration. In this radiance equation, incident ray is computed by integer on-pass voxel traversal algorithm as it is very fast for computing. Before the ray which is sent from the light source reaching the destination voxel, it has traveled through some other voxels. So we should take account of the light attenuation along the light path.The radiance of a voxel can beDescribed as this:2. Simulation clouds on GPUThe development of GPU various applications associated with computer graphics advance greatly. At the same time, the highly processing power, parallelism and programmability available nowadays on the contemporary GPU provide an ideal platform on which the general-purpose computation could be made. Realistic fluid real time rendering is a technique which accomplisher realistic rendering in a definite period of time under the constraint of current hardware and graphic algorithms. This paper realizes an application program which simulates clouds on GPU with realistic fluid real time rendering method. According to the flow theory in GPU, I packing some flows in different channels of texture for improving the operation speed. The experimental results show that our clouds correspond to real cumulus and do not have a cross look or a cylindrical look as predecessors did. The simulation program runs steadily with several examples.3. Improved control strategies of simulationTo make the simulation more realistic, I improve the control strategies. Set upward velocity vector as well as the color value in source, to simulate cumulus'growing. Set periodic horizontal velocity vector in source. This work can ensure the simulation visually-realistic in every period. I simulate cloud flying in the wind by move simulation space in world space as well as add external force in simulation process for detail realistic of cloud flying.4. Realized a roamed 3D sky backgroundI render a terrain in GPU pipeline with a HightMap, which realized a roamed 3D sky background. As several clouds simulate in it, it looks as a cloud layer. The results show that the visual effect of 3D cloud scene is true to life and real-time.The innovative points in this paper are derivation cloud simulation equation all by atmospheric physics. Every step of the derivation is based of cumulus characteristics in physics. There are two improvements:1 Add the diffusion term in simulation equation. 2Derivate a new buoyancy force which is different from Mark.J.Harris's cloud simulation.With the improved control strategies, this model shows a good performance.