本文利用壓電材料-聚偏氟乙烯（Po1yvinylidene fluoride, PVDF）作為跑步機應力量測裝置，並利用系統識別的方法求得跑步機動態模式(Ｍodel)，透過跑步機模式與跑步機量測應力裝置可得到人在跑步機上的地面反作用力，並透過商品化產品荷重元(Load Cell)驗證PVDF在量測應力上的可行性、可靠度和精確度。為驗證此跑步機模式的正確性，本文規劃量測十位受測者在三種不同運動類型下(原地踏步、走路和跑步)的訊號。其中踏步訊號擁有真實的輸入訊號，可用於直接判斷跑步機模式的正確性，其真實輸入和預測輸入之平均波形相關係數(Correlation Coefficient)大多高達0.98以上，而走路和跑步訊號雖沒有真實量測輸入，但預測輸入和實際輸出訊號的相關係數也可達到0.96以上，這代表著跑步機模式預測準確度高且擁有高線性度。
最後透過平均波形的比較，可觀察出該十位受測者不同測試訊號波形上的差異性。在原地踏步上地面反作用力(Ground Reaction Force, GRF)波形幾乎較為類似，而走路則因個人習慣的不同造成波形型態較廣。至於跑步GRF波形受腳掌落於地面的方式不同，而有些許的差異。

To identify the dynamic model of the treadmill, this study uses the piezoelectricity material (Po1yviny-lidene fluoride, PVDF) to measure the force under treadmill. With this dynamic model, the ground reaction force (GRF) can be derived from the PVDF sensors. The reliability and precision of the GRF results are verified by replacing the PVDF with the traditional load cell (Force measurement devices).
To verify the accuracy of treadmill model, this study acquires three different types of GRF signals (marking time, walking and running) from ten subjects. For the marking time case, the correlation coefficients between the actual and predicated GRF signals are approximately 0.98. This study also demonstrates that the proposed model can provide sufficient bandwidth for the walking and running GRF signals.
Finally, via comparing the average GRF profile, inter-personal differences of the GRF signal can be observed. Among the three tested locomotion patterns, the marking time GRF has the highest similarity whereas the running GRF signals has the largest variability.

論文審定書 i
誌謝 ii
表目錄 vi
圖目錄 vii
摘要 ix
Abstract x
第一章 緒論 1
1.1前言 1
1.2研究動機與目的 2
1.3文獻回顧 2
1.4論文架構 3
第二章 跑步機量測平台之建立 4
2.1訊號擷取 4
2.2感測元件之特性介紹 5
2.2.1壓電材料 5
2.2.1.1壓電材料種類 5
2.2.1.2 PVDF 6
2.2.2 荷重元(Load Cell) 8
2.2.2.1 應變規 8
2.2.2.2 惠斯通電橋 9
2.3架設跑步機量測平台 10
2.3.1踏步用量測平台 10
2.3.2跑步機下方之裝置 11
第三章 實驗方法與跑步機模式建立 15
3.1建立跑步機模式 15
3.1.1實驗理論 15
3.1.2實驗步驟 17
3.1.3模式建立 18
3.2驗證跑步機模式 20
3.2.1平均步伐波形 20
3.2.2踏步測試 26
3.2.3走路和跑步測試 32
第四章 結果與討論 37
4.1荷重元與高分子壓電材料之預測GRF比較 37
4.2使用PVDF預測各種測試訊號之GRF準確度 38
4.3人與人之間GRF的差異性 41
4.3.1踏步測試 41
4.3.2走路測試 44
4.3.3跑步測試 49
第五章 結論與建議 53
參考文獻 55

Alex,　J.Y. Lee, Jia-Hao Chou1, Ying-Fang Liu, Wei-Hsiu Lin, Tzyy-Yuang Shiang “Correlation Between Treadmill Acceleration, Plantar Pressure, and Ground Reaction Force During Running,” The Engineering of Sport 7, 2008; 1:282-290.
Darren J. Dutto and Gerald A. Smith “Changes in spring-mass characteristics during treadmill running to exhaustion,” Medicine and Science in Sports and Exercise, 2002; 34(8).
Frédéric Dierick, Massimo Penta, David Renaut, Christine Detrembleur “A force measuring treadmill in clinical gait analysis,” Gait and Posture, 2004; 20:299-303.
Gary D. Heise, Philip E. Martin “Are variations in running economy in humans associated with ground reaction force characteristics,” J Appl Physiol, 2001; 84:438-442.
Jean L. McCrory, Scott C. White, Robert M. Lifeso “Vertical ground reaction forces: objective measures of gait following hip arthroplasty,” Gait and Posture, 2001; 14:104-209.
Kei Masani, Motoki Kouzaki and Tetsuo Fukunaga “Variability of ground reaction forces during treadmill walking,” J Appl Physiol, 2000; 92:1885-1890.
Patrick O. Riley , Gabriele Paolini , Ugo Della Croce , Kate W. Paylo , D. Casey Kerrigan “A kinematic and kinetic comparison of overground and treadmill walking in healthy subjects,” Gait and Posture, 2007; 26: 17-24.
Peter M. Mills , Rod S. Barrett , Steven Morrison “Agreement between footswitch and ground reaction force techniquesfor identifying gait events: Inter-session repeatability and theeffect of walking speed,” Gait and Posture, 2007; 26:323-326.
Roder Kram, Timothy M. Griffin, J. Maxwell Donelan, And Young Hui Chang “Force treadmill for measuring vertical and horizontal ground reaction forces,” J Appl Physiol, 1998; 85:764-769.
Vitomir Racic , James Mark William Brownjohn　“Stochastic model of near-periodic vertical loads due to humans walking,” Advanced Engineering Informatics, 2011; 25:259–275.