高黏滯性物體之融化模擬與流動控制

Abstract

高黏滯彈性物體之融化在真實世界中視一相當複雜的現象，包涵融化物質之流動、熱量的傳遞轉移，以及物質之相態變化等。我們將利用光滑粒子流體動力學法(Smoothed-Particle Hydrodynamics)來模擬物體融化，並且使用一個具有回饋機制的控制演算法來操控融化的流體。在納維-斯托克斯方程式(Navier-Stokes equation)中加入計算彈性壓力的力量，讓我們可以模擬出同時具有固態與液態特性的高黏滯彈性流體。我們並在粒子間模擬熱能傳導的行為，利用模擬後的溫度自動來決定黏滯性與彈力的模擬係數。此外，我們將設計了一套直覺化的控制機制來幫助使用者設計融化的過程。使用者可以直接在場景表面上繪製融化的路徑與融化後流體的形狀。本計畫為期兩年，第一年著力於高黏滯彈性物體融化模擬系統之發展與實作；第二年將設計一流體控制系統，並發展方法加速熱量傳遞之計算以及GPU加速技術。

The melting process of viscoelastic objects is a complicated phenomenon in real world, which includes fluid flowing, thermal radiation and heat transfer, and change between solid and fluid . We will develop a SPH-based simulation framework for object melting simulation and a feedback control algorithm to manipulate the melting flow. By adding an elastic stress term to the Navier-Stokes equation, we will simulate the melting behavior which has both fluid and solid features. We will develop a thermals transfer system to simulate the heat flow between particles, and based on the temperature of particles to automatically determine the viscosity and elastic stress coefficient. Moreover, we will introduce a feedback control algorithm which can be used to intuitively design the flowing behavior. Users can draw flowing path directly on the surface to design the shape of melting fluid. This is a two-year long project. In the first year, we will develop and implement the SPH-based simulation system for object melting. In the second year, we will develop a feedback control scheme that allows artists to design the melting flow and its final shape, and speed up the thermal exchange computation and develop a GPU-based computing framework.