依1993年AASHTO柔性鋪面加鋪設計所述，由校估後撓度值回算求得之回彈模數，並不等於在試驗室利用重覆載重試驗量測而得之回彈模數，建議以一係數校正之，方能獲得鋪面結構強度SNeff以進行柔性鋪面加鋪設計。本研究之目的即利用中央大學發展之本土化撓度值校估模式及中央大學試驗道路之相關試驗結果，進行校估後撓度值與回算值之相關性分析及回算值校正係數建立之本土化研究，最後進行柔性鋪面加鋪設計應用之分析及探討，藉以提供公路養護單位對於FWD在柔性鋪面加鋪設計上的有效運用。 對於校估後撓度值與回算值之相關性分析，本研究採用人工神經網路以進行分析，經利用倒傳遞神經網路訓練分析結果顯示，應用人工神經網路進行之相關性分析結果頗佳。對於鋪面各結構層回算值校正係數之建立，因瀝青混凝土之回彈模數係隨溫度而改變之溫感性材料，且本研究採用校估至瀝青混凝土有效溫度為25℃之撓度值進行後續分析，故將瀝青混凝土層之試驗室25℃回彈模數值，與FWD之回算值結果進行比較，以建立瀝青混凝土層有效溫度為25℃之回算值校正係數；而粒料級配及路基土壤之回彈模數係隨應力而改變之非線性材料，故本研究利用有限元素程式ABAQUS建立一2D軸對稱柔性鋪面有限元素模型，分析有限元素模型在承受FWD衝擊式動態載重下之粒料級配基層及路基土壤應力值，再依此應力值求解試驗室之非線性回彈模數值，與FWD之回算值結果進行比較後，即可建立粒料級配基層及路基土壤回算值之校正係數。由回算值校正係數分析結果得知，瀝青混凝土層為0.67、粒料級配基層為1.48，而路基土壤為1.10。 According to the 1993 AASHTO Guide for Design of Pavement Structure, the moduli calculated by adjusted deflection basins are not equal to the resilient moduli measured in the laboratory. It’s suggested that the adjusted moduli should be modified by the factor that is a function of structure number for overlay design in the Guide. The principal intention of this study is to develop the adjustment factors of backcalculated resilient moduli. Besides, the correction between adjusted deflection basins and backcalculated resilient moduli is another emphasis. The above-mentioned locally deflection adjusted model was developed from Nation Central University Test Road. According to the modified model, it provides effective application of Falling Weight Deflectometer (FWD) on overlay design in Taiwan. Artificial Neural Network is used as a methodological tool to develop the correction between adjusted deflection basins and resilient moduli, and the trained results illustrate the method considerably well. In order to develop adjustment factors of backcalculated resilient moduli of pavement structure layers, this study compares resilient moduli measured in the laboratory with backcalculated resilient moduli of FWD. Besides, this research analyzed the stress state of the granular subbase material and subgrade soil by Finite Element Method, in accordance with FWD dynamic loads to obtain the resilient moduli of laboratory. Comparing with backcalculated resilient modulus of FWD, this research develops adjustment factors of granular subbase and subgrade soil. The result of the adjustment factors show the factor is 0.67 for asphalt concrete mixture, 1.48 for granular subbase, and 1.10 for subgrade soil.
