本研究使用顯性動態有限元素法結合異向性降伏理論和連體損傷模式，進行金屬板材擴孔製程之裂縫成長分析，並探討不同初始孔徑對於沖頭負荷、初始破裂衝程、初始破裂位置和裂縫成長的影響，並與實驗做比較。 擴孔製程之裂縫成長數值模擬是運用單軸拉伸試驗之應力-應變關係式，配合連體損傷模式計算該材料損傷特性之損傷曲線，作為模擬擴孔製程料片破裂判斷之依據。擴孔製程數值分析隨著衝程增加，成形負荷會持續增加，於是材料內部產生了損傷，並開始有頸縮而導致微裂縫的產生，且利用應變最大值決定裂縫的成長方向，並以刪除元素法進行裂縫成長模擬分析。 為驗證數值模擬分析之正確性，本研究設計一組無引伸扣環模具進行初始孔徑 15.0mm、 9.0mm和 3.0mm料片之擴孔製程實驗。實驗結果顯示初始孔徑 3.0mm料片會於實驗中流入模穴，導致工件收縮過大，無法獲得正確實驗結果，因此另設計一組具引伸扣環模具以防止工件收縮，並重新進行此三種不同初始孔徑料片之實驗，且執行相對應邊界設定之數值模擬分析。二組不同模具之實驗結果皆顯示，由於料片材料不易均質和料片定位不易無誤差，導致初始破裂位置不易與數值分析一致，但趨勢相同；此外，沖頭負荷、初始破裂衝 程和裂縫成長皆與數值分析一致。 The objective of this thesis was to analyze the crack growth of workpieces with different initial holes diameters in bore-expanding process by using explicit dynamic finite element method based on anisotropic yield criterion and continuum damage model. The punch load, initial crack stroke, initial crack position and crack growth of workpieces were discussed and compared with the experimental results. The material damage curve was calculated based on stress-strain relationship of uniaxial tensile test and continuum damage model, and was used as the criterion of fracture in the crack growth simulation of bore-expanding process. It was seen that forming load increased with the increase of punch stroke more and more. Then the necking and micro crack occurred due to the interior damage of material. The crack growth direction was determined by maximum strain value, and the crack growth simulation was carried out by eroding element. In order to prove the reliability of numerical simulation analysis, this study designed a die set without draw bead to carry out the experiments of bore-expending process for the blanks with initial hole diameters of 15.0mm, 9.0mm and 3.0mm. The experimental results showed that the workpiece flowed into the die cavity when the initial hole diameter of blank was 3.0 mm. It caused the workpiece to shrink largely and resulted in improper experimental data. Therefore, another die set with draw bead was designed to avoid the shrink of workpiece, and was used to perform the same experiments for the blank with the same initial hole diameters again. The numerical simulation analysis was also proceeded for the corresponding boundary conditions. Because blank material was not uniform and blank orientation was not perfectly accurate, the comparisons between experiments using above two die sets and numerical simulation analysis of initial crack position were not very close, but the tendency of those results was agreed. Furthermore, other comparisons of initial crack stroke and crack growth in bore-expanding process showed good agreement between numerical and experimental analysis.
