闕志鴻Chiueh, Tzihong臺灣大學:天文物理研究所蔡御之Tsai, Yu-ChihYu-ChihTsai2010-05-042018-06-282010-05-042018-06-282008U0001-2507200822415000http://ntur.lib.ntu.edu.tw//handle/246246/178565自適應網格精緻化是一個當計算誤差變大時,將網格切得更細密以達到更高精確度的一個模擬方法。本論文發展一個使用高速圖形計算核心的自適應網格精緻化的模擬,提供可達一般自適應網格流體模擬的十倍效能。Adaptive Mesh Refinement is an idea of saving computing resources. In many cases the most important flow features occupy only a small region in the computation domain. AMR will recursively refine grids when errors grow and remove the fine grids where the coarser grids still give an enough accuracy. In my thesis, I present a fluid simulation program using the graphic processing units which are much powerful than the central processing unit in common personal computers. The program offers ten times computing efficiency than general fluid simulations before.致謝………………………………………………………………… i文摘要…………………………………………………………… ii文摘要…………………………………………………………… iii錄………………………………………………………………… iv目錄……………………………………………………………… v目錄……………………………………………………………… vi一章 緒論…………………………………………………… 1二章 程式內容介紹………………………………………… 3 2.1 程式簡介…………………………………………… 3 2.2 程式流程…………………………………………… 7 2.3 模擬初始條件的賦值……………………………… 9 2.4 網格調變…………………………………………… 10 2.5 模擬時距的決定…………………………………… 14 2.6 演化………………………………………………… 16 2.7 資料輸出…………………………………………… 19三章 程式的效能分析與測試……………………………… 20 3.1 震波管測試.………………………………………… 20 3.2 爆炸波測試………………………………………… 24四章 紊流的模擬分析……………………………………… 28 4.1 紊流………………………………………………… 28 4.2 模擬分析…………………………………………… 29五章 結論…………………………………………………… 36考文獻………………………………………………………… 37application/pdf2086632 bytesapplication/pdfen-US自適應網格流體模擬爆炸波切變流震波顯示卡AMRfluid simulationGPUshockshear flow使用高速圖形計算核心的自適應網格精緻化的流體模擬Adaptive Mesh Refinement Fluid Simulation with Ultra-fast Graphic Processing Unitshttp://ntur.lib.ntu.edu.tw/bitstream/246246/178565/1/ntu-97-R94244008-1.pdf