請用此 Handle URI 來引用此文件:
http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/89111
標題: | 應用計算流體力學探討NA-706事故潛在天氣危險因子 Application of computational fluid dynamics to investigate potential weather risk factors of NA-706 accident |
作者: | 廖昱豪 Yu-Hao Liao |
指導教授: | 黃美嬌 Mei-Jiau Huang |
關鍵字: | 計算流體力學,大氣邊界層,蘭嶼島環境風場,數據同化方法, Computational fluid dynamics (CFD),Atmospheric boundary layer (ABL),Orchid Island (Lanyu) environmental wind field,Data assimilation method, |
出版年 : | 2023 |
學位: | 碩士 |
摘要: | 2018年2月5日,中華民國空中勤務總隊編號NA-706的UH-60M型直升機於台北時間23:48:36從蘭嶼機場起飛前往台東機場,在起飛81秒後(台北時間23:49:57)墜落在蘭嶼西南側外海約3.5公里處。本研究目標為重建事發當時蘭嶼島周遭環境風場,並觀察直升機飛行路線上的氣流是否有危害飛行安全的可能性。研究使用ANSYS FLUENT 21,採用Reynolds-Averaged Navier-Stokes (RANS)搭配三種紊流模型(Standard k-ε、Realizable k-ε、SST k-ω),以中性大氣邊界層作為入口條件,模擬大氣流過蘭嶼島的流動情形。
本研究將測量數據規劃成三種用途:(1)校正入口條件、(2)實施數據同化方法、(3)結果驗證。研究採取數種方法對模擬進行優化,例如將地形模糊化以減少計算成本,根據大氣流動特性調整Standard k-ε模型的參數,加入科氏力,以及在壁面函數中加入地面粗糙度的影響等。最後,本論文引入以模擬機測試出的飛行風速限制值與紊流動能安全標準值來評估直升機飛行時受氣流影響的程度。模擬結果顯示,三種紊流模型對於整體流場定性上差異不大,其中SST k-ω模型對於漩渦的再現程度較佳,調整參數後的Standard k-ε模型在飛行軌跡上的紊流動能較大。在模擬已能夠重現蘭嶼島周遭流場大致流動情況下,引入數據同化方法對流場的影響不明顯,然而,該方法可以改善測站位置的模擬誤差,提升模擬結果的準確度。 根據事發當時的測量結果,氣流以風向20度吹入模擬區域。位於山坡背風側的蘭嶼機場周圍出現漩渦,影響直升機起飛作業,漩渦的尾流延伸至海面上,影響直升機飛行軌跡中後段的部分。此外,飛行軌跡上的平均模擬風向與直升機航向僅相差約30度,顯示直昇機為順風飛行,可能不易操控。檢視沿飛行軌跡上的模擬風速,均沒有超出飛行風速限制值,但紊流動能在飛行軌跡部分範圍會超出顯著紊流動能標準,超出的範圍與飛行員在對話紀錄中所提到遭遇亂流的範圍相符。 On February 5, 2018, a UH-60M helicopter No. NA-706 of the Taiwan National Airborne Service Corps took off from Orchid Island (Lanyu) Airport to Taitung Airport at 23:48:36 Taipei time. However, 81 seconds after takeoff, at 23:49:57 Taipei time, the helicopter crashed into the sea approximately 3.5 kilometers southwest of Orchid Island. The objective of this study is to reconstruct the environmental wind field around Orchid Island at the time of the accident, aims to observe if there was any potential risk to flight safety due to the airflow along the helicopter's flight route. This study uses ANSYS FLUENT 21, employing the Reynolds-Averaged Navier-Stokes (RANS) equations coupled with three turbulent models (Standard k-ε, Realizable k-ε, SST k-ω). The neutral Atmospheric Boundary Layer (ABL) is used as the inlet condition to simulate the flow of the atmosphere over Orchid Island. The measured data are planned for three purposes: (1) calibration of the inlet boundary condition, (2) implementation of data assimilation methods, and (3) validation of the results. This study takes some measures to optimize the simulation, such as blurring the terrain surface to reduce computational costs, modifying the parameters of the Standard k-ε turbulence model based on atmospheric flow characteristics, incorporating the Coriolis Force, and considering the influence of surface roughness in the wall functions. Finally, this study introduces wind speed limits measured by the flight simulator and turbulence kinetic energy safety criteria to evaluate helicopter flight conditions. According to the simulation result, the three turbulence models show little qualitative difference in the overall flow field. The SST k-ω turbulence model reproduces the vortices better, while the modified Standard k-ε turbulence model exhibits higher turbulence kinetic energy along the flight path. When the simulation is able to reproduce the general flow conditions around Orchid Island, the effect of the data assimilation method on the flow field is not significant. Nevertheless, the method can reduce the error from the local measurement, improving the reliability of the simulation results. According to the measurements at the time of the incident, the inflow enters the simulation area with a wind direction of 20 degrees. Some vortices are observed around Orchid Island Airport on the lee side of the hill, affects helicopter takeoff operations. When these vortices extend to the sea, the later part of the helicopter's flight path affected by strong turbulence. In addition, there is a difference of approximately 30 degrees between the average simulated wind direction along the flight path and the helicopter's heading, indicating a headwind flight condition that is unfavorable for control. The simulated wind speed along the flight path does not exceed the wind speed limit, but the turbulence kinetic energy exceeds the noticeable turbulence kinetic energy criteria in some parts of the flight path. The range where the noticeable turbulence phenomenon occurs matches the range mentioned by the pilot in the conversation records regarding encountering turbulence. |
URI: | http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/89111 |
DOI: | 10.6342/NTU202303047 |
全文授權: | 未授權 |
顯示於系所單位: | 機械工程學系 |
文件中的檔案:
檔案 | 大小 | 格式 | |
---|---|---|---|
ntu-111-2.pdf 目前未授權公開取用 | 17.58 MB | Adobe PDF |
系統中的文件,除了特別指名其著作權條款之外,均受到著作權保護,並且保留所有的權利。