本研究透過ANSYS Mechanical 和LS-DYNA兩種求解器，對擬靜態拉伸試驗進行隱性與顯性求解，並使用兩種材料模型Bi-linear (BISO)與Multi-linear (MISO)應力-應變曲線來模擬降伏點後的力學行為，最後根據實驗值與模擬值誤差和求解效率，比較兩者方法之優缺點。 首先利用MTS810針對熱浸鍍鋅鋼(SGCC)試片進行拉伸試驗，並將工程量測數據轉換為真應力應變，獲得其機械性質。接著於ANSYS中建立有限元模型，為了降低模擬誤差與提升分析準確度，在實際模擬進行前先藉由模型設定(積分設定方式、元素形狀與沙漏控制)與收斂性分析，篩選出適合本實驗之模型設定與網格。最後依據實驗所獲得之應力-應變曲線建立兩種材料模型，並分別設定邊界條件與此一有限元模型進行拉伸模擬。 由模擬至大約抗拉強度之結果表明，隱性求解器的BISO模型得到之應力誤差為1.017%，顯性求解器的BISO所得到之應力誤差為1.070%，兩者對照實驗數據皆非常接近。而隱性求解中額外使用了MISO模型，此一應力誤差僅有0.334%，分析更精準。由兩種求解器模擬過程與結果顯示，隱性與顯性求解皆能適用於此一擬靜態拉伸模擬：隱性算法求解快且CAE精準度高；顯性則有更多參數提供個別設定，面對複雜模型或不同受力情況時，能幫助穩定計算過程，減少模擬之誤差。 ;In this study, the implicit solver of ANSYS Mechanical and the explicit solver of LS-DYNA were used to perform quasi-static tensile tests. Two material models, Bi-linear (BISO) and Multi-linear (MISO) stress-strain curves, were employed to simulate the mechanical behavior beyond the yield point. Finally, the pros and cons of both methods were compared to each other based on the validation of simulation against experimental data and computational efficiency. First of all, the MTS810 was used to perform tensile tests on hot-dip galvanized steel (SGCC) specimens, and the engineering measurement was converted into the true stress-strain to obtain their mechanical properties. Subsequently, a finite element model was established using ANSYS. In order to reduce simulation errors and improve analysis accuracy, the parameters (integration settings, element shapes, and hourglass control) and convergence analysis were performed to select suitable parameters and meshes for this experiment. Finally, two material models were established based on the stress-strain curves obtained from the experiment. They were respectively incorporated into the finite element model with boundary conditions for tensile simulations. The results of simulations around the ultimate tensile strength indicate that the implicit solver with the BISO model obtained a stress error of 1.017%, while the explicit solver with the BISO model obtained a stress error of 1.070%. Both results were close to the experimental data. Additionally, the implicit solver with the MISO model was used to achieve a stress error of only 0.334% which provides even more accurate analysis. The simulation results of the two solvers demonstrate that both implicit and explicit solvers are suitable for the quasi-static tensile test. The implicit algorithm offers fast solving speed and high CAE accuracy. In contrast, the explicit solver allows for individual settings for the model also stabilizes the calculation process and reduces simulation errors, especially for complex models or different loading conditions.
