根據波音公司97年度之全世界民用航空器失事統計報告，約50﹪飛航事故發生於降落階段。若班機在降落過程中遭遇到陣風，會使班機穩定度產生變化，此研究便著力於分析某機型遭遇陣風對於其飛行穩定度所造成的影響。並且另外分析該班機在跑道濕滑情況下地面與輪胎之間摩擦係數大小與其班機最後發生滑出跑道事故之關連性。
本研究使用飛航記錄器所記錄的飛航資料作為資料來源，一開始先將記錄器的飛航資料經由擴展式卡爾曼濾波器將其雜訊以及誤差修正，再利用EKF的估算結果去計算飛航記錄器所未記載之飛航參數以及三維風場。將上述結果與FDR資料作為系統輸入，使用在幅狀基底函數類神經網路來獲得縱向以及橫向之氣動力導數。在最後一部份將FDR資料作為系統輸入，利用落地煞車動力模型以及最小平方法來算出落地煞車階段輪胎與地面間之摩擦係數。
根據RBFNN所計算出來的結果，橫向氣動力導數在某時間區段內超出穩定範圍，表示該陣風確實使班機在落地前的橫向運動方面呈現不穩定狀態。而跑道摩擦係數估算結果顯示此班機由於是在小陣雨天氣降落，因此其煞車係數值便落在合理的濕滑跑道應有的煞車係數範圍中。

According to one of published statistics from Boeing Commercial Airplane, there are 50 % of flight accidents that are occurred in landing phase. The weather hazard becomes much more severe when airplanes encounter a strong cross wind. The research aims to infer flight information especially in cross wind effect.
The study presents data processing approaches to obtain flight information from actual flight data recorder (FDR) data. Firstly, the Extended Kalman Filter (EKF) is applied to reduce the effect of the data noise and bias of the recorded flight data. Secondly, the EKF results are used to compute the un-recorded useful parameters and three dimensional winds alone the flight track. Thirdly, the radial based function neural network (RBFNN) method is used to compute both longitudinal and lateral aerodynamic derivatives. In addition, a smoothed method is applied to confirm the derivative trends. In the last approach, the rolling resistance is estimated with a least-squares method while holding the other force terms (e.g. aerodynamic force) to their last estimates.
The results from processing a FDR data affected by a strong cross wind show that the overall processing approach is capable of inferring the flight safety information. In compute solution, the aerodynamic derivatives rapidly changed to a unstable condition show the cross wind temporally threatens the lateral stability while less affect on the longitudinal stability. In addition, the results of braking coefficient are reliable.

[1] 行政院飛航安全委員會.98年度工作報告.
[2] Boeing Commercial Airplane, Airplane Safety,“Statistical Summary of Commercial Jet Airplane Accidents Worldwide Operation,” 2008.
[3] Dr. James C. Wambold and J. J. Henry… “International Runway Friction Index (IRFI) versus Aircraft Braking Coefficient (Mu) ” CDRM, Inc.,
[4] M. Bastian, J.B. Croll, J.C.T. “Martin,Falcon 20 Aircraft Performance Testing on Contaminated Runway Surfaces During the Winter of 1999/2000”. September 2001
[5] 翁錦堂. (2004). 從飛行記錄資料進行降落時遭遇風切航機之氣動力分析，博士論文，國立成功大學航空太空工程研究所. 台南.
[6] 楊明浩、何慶雄、蕭飛賓. (2006). 擴展式卡爾曼濾波器於航機遭遇亂流之飛航資料估算. 中國航太學會/中華民航學會聯合學術研討會.
[7] 官文霖、梁群. (2004). 飛航記錄器於飛航事故調查之應用. 2004 海峽兩岸智慧型運輸系統學術研討會.
[8] 王亭介.(2008). 應用擴展式卡爾曼濾波器於估算積冰事故飛機之氣動力導數，碩士論文，國立成功大學航空太空工程研究所. 台南.
[9] Louis, V. S. (1998). Introduction to aircraft flight dynamics. Reston, VA: AIAA.
[10] 蔣宏仁. (1997). GPS輔助慣性導航分散濾波之設計及驗證，碩士論文，國立成功大學航空太空工程研究所. 台南.
[11] 廖俊安. (2007). 應用GPS/INS於超輕航機氣動力係數之估算，碩士論文，國立成功大學航空太空工程研究所. 台南.
[12] Gelb, A. (1974). Applied Optimal Estimation. Cambridge: M.I.T. Press
[13] Minkler, G., & Minkler, J. (1993). Theory and Application of Kalman Filtering. Palm Bay, FL: Magellan Book Company.
[14] 楊明浩、何慶雄、蕭飛賓.(2008). 民航機記錄飛航資料應用類神經網路識別縱向氣動力導數. 航太學會學術研討會.
[15] 行政院飛航安全委員會. 飛航事故調查報告 台北縣新店市
[16] Boeing website: http://www.boeing.com/commercial
[17] O'Callaghan, J. (2002). Aircraft Performance. Washington , D.C: NTSB.
[18] Roskam, J., “Airplane Design-Part V Component Weight Estimation,”
DARcorporation, 1999,p.p.17-24.

[19] Wang, Z., Li, J., C., E. L., Brandon, M. J. (2001). “Estimation of Unsteady Aerodynamic Models from Flight Test Data”. AIAA Paper .
[20] Brian, L. S., Frank,L. L., “Aircraft Control and Simulation,”John Wiley & Sons,Inc., 1992.
[21] Nelson, R. C., “Flight stability and automatic control, ” 2nd ed. Boston, Mass.: WCB/McGraw Hill, 1998.
[22] AIRBUS INDUSTRIE “Flight Operations Support ,Chapter 8:Landing Techniques” AIRBUS INDUSTRIE
[23] 官文霖. (2008).民用航空器飛航事故於溼滑跑道之落地性能研究.第三屆 海峽兩岸航空氣象與飛行安全研討會.