隨著鐵路運輸的快速發展，台灣鐵路逐漸長軌化，提供列車行駛的穩定性與舒適度，但隨之而來的問題是，軌溫增減所產生的位移被長軌束制住，溫度力轉為鋼軌內應力，導致軌溫過高容易發生挫屈，軌溫過低容易發生拉斷。軌道穩定性除了與軌道結構狀況、列車動態作用力等因素有關，最直接的因素是軌溫，如何即時掌握軌溫數據並發送警示是一項重要的課題。
本文研究軌溫量測方法，開發無線監測軌溫系統並以不同感測儀器進行驗證。無線監測系統可以24小時不間斷量測數據，並以無線方式將資料傳回電腦及監測網站，提供完整且即時的軌溫資訊，當軌溫超出設定門檻時，使用者可以立即得知警示訊息。
由於許多區域仍缺乏當地軌溫歷時資料，本文旨在利用氣溫、日照量等氣象條件預測軌溫變化趨勢，並分析不同環境條件對軌溫的影響。最後，將現地試驗量測數據與氣象資料建立預測模型，並以台鐵大甲車站之軌溫資料做為驗證。結果顯示預測與實際最高軌溫的平均誤差為在1℃以內。模型可用氣溫、日照量歷時數據，預測最高與最低軌溫以及軌溫變化趨勢。

With the rapid development of rail transport, Continuously Welded Rail (CWR) has been used in Taiwan by degrees, providing the train with stability and comfort. But, the displacement caused by the changes of rail temperature would be fixed by CWR with temperature force transforming into internal stress. Therefore, the rail can distort causing buckling at higher temperature and tension cracks at lower temperature. Besides the condition of rail structure and dynamic force caused by train, the key element for the stability of rail track is rail temperature. It is a major issue to control the instant rail temperature data and raise the alarm.
The study discussed the methods of measuring temperature and developed the wireless monitoring system of rail temperature which is verified with different sensors. The wireless monitoring system can detect 24 hours a day, and send the data to PC or website. It can provide complete and real-time rail temperature data. Users can learn the alarm messages instantly at certain temperature thresholds.
Due to lack of rail temperature data in some areas, the study purposed to predict rail temperature with air temperature and amount of insolation, and analyze the effects of different ambient conditions. Finally, the prediction model is developed with measuring data and meteorological data. The model performance can be verified with the rail temperature data at Dajia Train Station. The results show that the mean error of predict maximum temperature and actual maximum temperature is in 1℃. The model can predict maximum/minimum temperature and the trend of rail temperature with air temperature data and amount of insolation data.

1.Huifang Gao, Kaisheng Ma, and Wenchao Zhang, The real-time temperature measuring system for the jointless rail, 2011.
2.Massimo Leonardo Filograno, Pedro Corredera Guillén, Alberto Rodríguez-Barrios, Sonia Martín-López, Miguel Rodríguez-Plaza, Álvaro Andrés-Alguacil and Miguel González-Herráez, Real-Time Monitoring of Railway Traffic Using Fiber Bragg Grating Sensors, 2012.
3.Federal Railway Administration, In-motion, non-contact rail temperature measurement sensor, 2012.
4.Federal Railway Administration, “Development of rail temperature prediction model”, 2008.
5.L. Chapman, J. E. Thornes, Y. Huang, X. Cai, V. L. Sanderson, and S. P. White, Modelling of rail surface temperatures: a preliminary study, 2007.
6.L. Girardi, D. Boulanger, J.Colibri, M. Fumey, C. Mangone, P. Pouligny, J.M. Balabaud, J.M. Pissot, J.Plu, Long and short-term rail temperature forecasts for track works planning and rail stress control, 2009.
7.1067公厘軌距鐵路長焊鋼軌鋪設及養護規範，交通部，2002年11月
8.朱旭，軌道工程，2010年5月
9.Andrew Kish, Gopal Samavedam, Improved Destressing of Continuous Welded Rail for Better Management of Rail Neutral Temperature, 2005.
10.Esveld C., Modern Railway Track, 2nd ed., MRT Productions, Zaltbommel, ISBN:90-800324-3-3, 2001.
11.Ventry D, “Rail track company specification: continuous welded rail (CWR) track” RT=CE=S=011, Issue 4, February 2002, 55 pp, (2002)
12.Hunt GA, “An analysis of track buckling risk”, British Railways Internal Report RRTM013, 31pp, (1994)
13.RailTempMATE, from
http://railtempmate.com/index.htm
14.evopro eRTM, from
http://www.evopro.hu/eng/page/ertm
15.Kapsch CarrierCom, Rail Switch Temperature Monitoring Solution, from
https://www.kapsch.net/kcc/downloads/DocumentItems/Kapsch-KCC-Rail_Switch_Temperature_Monitoring_Solu?lang=en-US
16.DataTraks, SafeTraks Rail Stress Monitoring System, from http://datatraks.com/PDFs/rail_stress_monitoring_system.pdf
17.LBFoster, Rail Stress Monitor, from
http://www.lbfoster-salientsystems.com/Rail_Stress_Module.asp