台灣的氣候變化較大，當颱風及暴雨帶來大量降雨時，上游處經常會遭大雨不斷侵蝕導致土石鬆落進而掉入河道中形成渾水潭，渾水運移時會因懸浮物質造成流體呈一垂直壓力梯度的非均質分佈，此現象稱之為層化異重流。本研究為了解異重流的各種層化形貌，以利減緩異重流沉降於水庫，導致水庫泥砂淤積日益嚴重的問題，透過狹長型的定界交換水槽實驗模擬實際河道的形貌，並利用染劑區別不同密度的釋放源並依照密度順序進行疊加，使影像顯示不同初始條件下的運動形貌及頭端位置，而初始條件則會隨著兩個主要配比參數而有所變化，其一為重力驅動比(B*)，B*值越大時為上層釋放源驅動，其二為初始釋放源密度差異比(R*)，R*值越大時，表示越趨近於均質，因此兩層重流體混合較為顯著。由於整體釋放源與環境流體之密度差異較小，因此數值模擬中引入布氏假設(Boussinesq approximation)於納維爾-史托克斯方程組(Navier-Stokes equations)，而為了使運算時間可大幅縮減，因此加入大渦流紊流模式(Large-eddy Simulations)進行演算。藉由演算獲致的流場資訊可進而推估層化異重流運移過程中的能量收支(Energy Budget)，搭配實驗影像資料判斷產生混合的機制及範圍。本研究將採用Winters (1995)所提出的有效位能(Available Potential Energy)及背景位能(Background Potential Energy)的概念量化其混合程度。於每一暫態時間重新排列層化異重流的網格的密度值分佈，從中得出混合最劇烈的時段並與能量圖上的起伏變化與實驗進行比對。此外，該分析也可探討層化異重流所衍生出Kelvin–Helmholtz的流體不穩定混合機制。研究中，依照演化過程及混合機制大致可分為四種不同的層化流況，從影像中發現流體躍過、侵入以及混合等現象，並利用該分析預測異重流於河道中的沉積之過程。

A study of the magnetic effect on heat and mass transfer by mixed convection flow from a linearly stretching surface is reported considering thermophoretic effect as well. In this study, the flow field was modeled as two dimensional, incompressible, and steady-state laminar flow. The similarity transformation and numerical techniques are used to solve governing equations include mass, momentum, energy and concentration. The solutions are obtained by fourth-order Runge-Kutta integration method with shooting scheme. The results include Hydrogen (H2) , Oxygen (O2) , and Carbon dioxide (CO2) gases velocity, temperature and concentration profiles under different variable Hartmann number (Ha), Richardson number (Ri), Eckert number (Ec), and Schmidt number (Sc).
The conclusions could be summarized as: 1. The larger magnetic effect Ha decrease the velocity profile, whereas increase the thermal and concentration profile due to the enhancing Lorentz force. 2. The larger Ri increase the velocity profile but decrease the thermal and concentration profile due to larger ratio of buoyancy compared to stretching velocity. 3. The larger Ec increase the thermal profile due to larger ratio of stretching velocity compared to temperature difference. 4. The concentration profile decrease with increasing Sc correspond to the lower gas mass diffusivity.

本研究主要是利用計算流體力學方法探討位於艦艇上層結構之煙囪形狀、各種不同風向角度之相對速度、排煙溫度情況下，空氣流場流經上層結構對排煙擴散分佈之數值模擬與分析。本研究係以艦艇於風洞試驗中所觀察之不同煙囪幾何結構之流場變化與物理現象機制，利用計算流體方式，分析在不同風向與角度之相對速度所造成之煙流、擴散路徑及煙流溫度分佈情形，探討煙囪高度、幾何構型、不同之風向角度、排煙溫度、排煙速度比等各參數對行進間之艦艇運動變化。本研究將建置針對艦艇因上層結構之煙囪形狀、各種不同風向角度之相對速度、排煙溫度之數值分析模式，並藉此模式探討未來因煙囪形狀、位置及各風向速度流場變化，以提供設計與改善減少阻力之可行性。
本研究工作重點為分析煙囪高度、幾何構型、不同之風向角度、排煙溫度、排煙速度比等各參數對行進間之艦艇運動的流場變化，比較數值模擬與實驗數據之分佈差異。最後本研究計畫將綜合研究成果，評估該數值模式對艦艇上層結構之煙囪形狀、各種不同風向角度之相對速度、排煙溫度模擬能力與未來提供概念設計及初步設計可行性，以作為煙囪與結構物規劃及設計參考之用。研究成果預期瞭解艦艇結構與煙囪形狀、位置對於大氣風壓、擴散擾動以及噪音的產生與衝擊，以及空氣流經上層結構物之擴散模擬的範本。

A study of aerosol particle deposition rate affected by magnetic effect and variable viscosity onto a stretching surface is proposed considering the thermophoresis as well. The interactive effects among different transport mechanisms are proved to be very important for particle deposition rate. In this study, the air flow is modeled as two dimensional, incompressible, and steady-state laminar flow. The similarity transformation and numerical techniques are used to solve governing equations include mass, momentum, energy and concentration fields. The solutions are obtained by fourth-order Runge-Kutta integration method with shooting scheme. The particle deposition velocity (Vd) could be calculated after solving the concentration equations. The magnetic effect parameter (M), variable viscosity parameter (θr), and Richardson number (Ri) are selected to discuss the results among the effects. As the coupled effects are considered, we can find that particle deposition velocity increase with the increasing magnetic effect, M as dp<0.1μm; but decrease with increasing, M as dp>0.1μm. One the other hand, the deposition velocity increase with the decreasing θr in the range of selected particle size. Furthermore, the deposition velocity increase with the increasing Ri in the range of selected particle size as well due to the enhancement of compared ratio of buoyancy force versus to stretching velocity.

在廣闊的大海中，身為頂級掠食者的鯊魚，能透過極度敏銳的嗅覺能力，快速辨別出獵物所在的位置。而鯊魚是藉著游向前的速度引導水流通過鼻腔，便可感測水中的氣味分子，所以不論是外部水流的流動狀況或者鼻腔內部的流動方式，皆可能影響氣味分子的接收情況。本研究將以鋸峰齒鯊為研究對象，運用計算流體力學（CFD）模擬鯊魚在行進時嗅覺區域的水流狀況，對外部的流場以及鼻腔內部的流動情況進行分析與討論，以流體力學的角度進行鯊魚在水中移動時對嗅覺影響的先期研究。
由於最初之模型僅有魚頭部分，因此於鼻腔中加入層狀結構通道。進行數值模擬時，網格的品質與粗細將影響模擬結果，且本研究所使用的鯊魚模型具有複雜的曲面構造，容易在模型壁面處產生品質較差的網格，因此會以Building Cube Method(BCM)進行網格建構。完成模型與網格的建立後，運用三種不同外觀之模型進行 CFD 模擬並比較其流場，得到柱狀加長之模型較能呈現真實流場。再通過對鼻腔區域建置不同大小之網格的測試結果，發現網格大小為0.092mm 時，即可呈現鼻腔內完整的水流流動狀況。最後由柱狀加長搭配最小尺寸為0.092mm之網格所進行的CFD模擬結果顯示，鼻腔內之主要通道的流速是由前端往末端遞減，而層狀結構通道中的流速是由入口端往末端遞減。
本研究已得到由CFD模擬出鋸峰齒鯊鼻腔中的水流流況，並且所使用的外觀模型、網格配置與模擬之水流結果，可以作為未來研究發展之參考基礎。

The transport of human expiratory droplets plays a critical role in the transmission of respiratory infectious diseases. Although the transport mechanism has been extensively studied in the past, there has been little investigation into the ambient relative humidity effect on droplet travelling distance. Motivated by this consideration, an analytical approach is proposed to study the transport and dispersion of human expiratory droplets. In this study, the human violent exhaled activity was modelled by turbulence round jet model. The effect of turbulence fluctuation on droplet dispersion was modelled by a discrete random walk model. The particle reach probability is analyzed to evaluate the droplet travelling distance. The droplet transport distribution is unveiled in a series of scenarios of different exhaled activities. The environmental conditions showed the sensitive impact on droplet transport. The large droplets (>100 μm) fall to the ground before evaporating to droplet nuclei whereas the small droplets (<10 μm) dry out before hitting the ground. Increasing the ambient relative humidity will significantly decrease the droplet reach distance. The present model could be applied to determine the indoor social distance during the Covid-19 pandemic.