高強度汽車保險桿沖壓成形特性之研究

Abstract

在環保意識高漲與車身輕量化之需求下，目前各大車廠開發時應用高強度鋼板已經是明確之趨勢。但由於鋼板強度的提高，高強度鋼板成形性低於一般低強度鋼板，故高強度汽車沖壓件，例如保險桿於深引伸時，其破裂情況更加嚴重，使過去低強度保險桿之模具設計法則已無法完全套用於高強度保險桿之沖壓成形。又因鋼板強度提升，成形後之回彈缺陷也較一般低強度鋼板明顯，造成回彈模面補償的設計更加困難。 針對沖壓成形之有限元素法分析，在過去研究文獻中，雖已針對模擬參數進行最佳化之收斂性測試，但對於提升預測回彈之準確度仍有改善空間。因此，本研究將透過包辛格實驗與雙軸拉伸實驗建立完整之高強度鋼板材料模型，藉以提高有限元素法模擬回彈之準確性。本研究同時針對不同材料模型對於模擬分析之準確度影響進行探討，驗證了Yoshida-Uemori材料模型不僅因考慮包辛格效應使有限元素法模擬回彈預測之準確度提升，且發現該材料模型於沖壓成形之卸載(unloading)過程中對於板材楊氏係數之修正，亦會對回彈預測之準確度具有明顯之影響性。本研究亦選擇具代表性之基礎載具進行沖壓成形回彈現象之有限元素法模擬分析並與實際試驗結果比較，得知 Barlat 91+Yoshida-Uemori材料模型與Hill 90+Yoshida-Uemori材料模型較為接近實驗值。 應用上述建立之材料模型，本研究針對高強度保險桿沖壓成形特性進行探討。首先透過不同保險桿成形模面之特徵分析，歸納出解決成形缺陷所對應之餘肉造型，進而針對本研究之高強度保險桿進行餘肉造型設計，解決其成形缺陷。並於板材沖壓成形過程中，觀察板材受力情形，發現板材於成形中受到拉伸與壓縮之變形歷程，產生完整的包辛格效應，故使用將包辛格效應加入塑性變形分析之Yoshida-Uemori材料模型確實有其必要性。最後透過試模驗證得知， 採用Barlat 91+Yoshida-Uemori材料模型或Hill 90+Yoshida-Uemori材料模型之有限元素法模擬分析，其結果仍較採用其他材料模型較為接近實驗值。

Due to the environmental consciousness and the demand of light-weight vehicles, high strength steel has been widely used in automotive parts. However, because of the increased strength of steel, the formability of high strength steel is inferior to traditional low strength steels. Therefore, the die design concept for stamping low strength steel sheets is no longer applicable to high strength steel sheets. Concerning the finite element analysis on the stamping of high strength steel sheets, the efforts have been endeavored to establish the optimum simulation parameters. However, the accuracy in the prediction of springback is yet to be improved even with the optimum simulation parameters adopted. Therefore, biaxial stretching experiments and cyclic tension-compression experiments were implemented in the present study to establish a complete material model for the high strength steels so as to enhance the accuracy of the finite element simulations in predicting the springback phenomenon. The finite element simulations with different material models adopted were also performed in the present study to evaluate their efficiency on predicting springback occurred in the stamping process. The simulation results were compared with the experimental data and the outcome reveals that the Yoshida-Uemori material model which considers the Bauschinger effect lends itself to the most efficient model in improving the accuracy of springback prediction. It is also found that the change of Young's modulus that is taken into account in the Yoshida-Uemori model during the unloading process affects the springback prediction significantly. By comparing the simulation and experiment results of the stamping of simple bent parts, it indicates that the simulation results with Barlat 91 yield criterion and Yoshida-Uemori model adopted, or the Hill 90 yield criterion and Yoshida-Uemori model employed, are much closer to the experimental values. The forming characteristics of stamping a front bumper with high strength steel sheet was then examined with the material constants established from the cyclic tension-compression tests and biaxial stretching tests conducted in the present study. The die addendum design was investigated by categorizing the features of different bumper die surfaces, and the optimum addendums that could eliminate specific defects such as wrinkle and fracture were identified. Since some portions of the sheet blank were subjected to tension and compression deformation during stamping, the Bauschinger effect predominated in the deformation process. Therefore, using the Yoshida-Uemori material model is necessary. Through the simulation validation, we confirm that the Barlat 91 yield criterion and Yoshida-Uemori model, or the Hill 90 yield criterion and Yoshida-Uemori model are still the most efficient material models to be used in the finite element simulations for stamping of high strength steel sheets.