由於大型結構實驗費用昂貴，研究人員經常利用實驗數據建立數值模型，用以模擬結構實驗反應，在數值模型建立的過程中，非線性材料模型參數通常需要使用試誤法找尋最適合的參數組合，在這過程中往往耗費大量時間，且試誤法找尋出的參數組合並非為擬合實驗數據的最佳解。為減少研究人員建立數值模型所耗費的時間，本研究以MATLAB做為開發平台，搭配Abaqus有限元素分析軟體作為分析工具，建立一套自動化分析的流程，並利用粒子群最佳化（Particle Swarm Optimization, PSO）以及生物共生演算法(Symbiotic Organisms Search, SOS)之最佳化搜尋方法擬合實驗數據，找尋最佳的非線性材料模型參數，以利研究人員建立可靠的數值模型。本研究以自動化流程建立非線性材料模型參數最佳化搜尋方法，以兩組實驗進行參數擬合，驗證本方法的可行性。

Because of the high cost of structural experiments, researchers usually use real experimental data to build numerical models to simulate the reactions. To build a reliable numerical model, researchers will use the nonlinear material model to make the simulation as much as possible to be real. It will spend a lot of time to find the most suitable parameters of the nonlinear material model by trial and error. In this study, the optimization methods are used to estimate the parameters of the combined hardening nonlinear material model in order to reduce the time it takes for researchers.
This study used two methods of optimization, Particle Swarm Optimization (PSO) and Symbiotic Organisms Search (SOS), to find the most suitable parameters of the combined hardening model in Abaqus by MATLAB as the programming platform, and to verify the feasibility by the real experiments.

參考文獻

[1] 莊明介、謝尚賢、吳安傑, “實驗數據自動化參數擬合方法之研究-以BRB實驗為例,” 結構工程, 第三十卷, 第四期, pp. 79-106, 12 2015.
[2] 吳紫瑩, 冷作對BCP柱耐震性能之影響, 國立台灣科技大學碩士論文, 2018.
[3] J. S. Charles, 耐震用含側封板H鋼梁之設計方法, 國立台灣科技大學碩士論文, 2018.
[4] Erwin, H-型鋼梁之挫屈行為研究, 國立台灣科技大學碩士論文, 2011.
[5] 楊裕豪, 含防挫屈裝置H型鋼梁之反復載重試驗, 國立台灣科技大學碩士論文, 2016.
[6] D. Pachoumis, E. Galoussis, C. Kalfas and I. Efthimiou, "Cyclic performance of steel moment-resisting connections with reduced beam sections—experimental analysis and finite element model simulation," Engineering Structures, vol. 32, no. 9, pp. 2683-2692, 2010.
[7] C.-C. Chen, Erwin, C. J. Salim and R. Tiara, "Seismic performance of wide flange steel beams partly covered by side plates," Journal of Constructional Steel Research, pp. 275-286, 2018.
[8] Y. M. M. Palomino, Model fitting of a bilinear material with Genetic Algorithm, master thesis, University of California, Irvine, 2011.
[9] J. Lemaitre and J. Chaboche, "Mechanics of solid materials," Cambridge University, Cambridge, United Kingdom, pp. 161-241, 1990.
[10] 石永久、王佼姣、王元清、潘鹏、牧野俊雄、齊雪, “循环荷载下低屈服点钢材LYP225的力学性能,” 东南大学学报:自然科学版, vol.44, no. 6, pp. 1260-1265, 2014.
[11] Y. Gorash and D. MacKenzie, "On cyclic yield strength in definition of limits for characterisation of fatigue and creep behavior," Open Engineering, vol. 7.1, pp. 126-140, 2017.
[12] "Abaqus 6.13 Documentation," Dassault Systèmes, 2013.
[13] R. E. a. J. Kennedy, "A new optimizer using particle swarm theory," MHS'95. Proceedings of the Sixth International Symposium on Micro Machine and Human Science. , 1995.
[14] M.-Y. Cheng and D. Prayogo, "Symbiotic organisms search: a new metaheuristic optimization algorithm," Computers & Structures, vol. 139, pp. 98-112, 2014.
[15] 鄭明淵、邱建國、邱永芳、吳育偉、徐梓隆、吳宏興、林正軒, “創新生物共生演算法 (SOS) 在橋梁生命週期風險評估與維護策略之研究,” 中國土木水利工程學刊, 26(4), pp. 293-308, 2014.
[16] R. Tiara, 含封板之H型鋼樑在反覆載重下之側向扭轉挫屈行為, 國立台灣科技大學碩士論文, 2016.
[17] F. Sabourin and P. Simon, "Identification of a Combined Hardening and Application to a “One-Step” Inverse Approach," in VIII International Conference on Computational Plasticity, 2005.
[18] M. Wolff, B. Suhr and C. Simsir, "Parameter identification for an Armstrong--Frederick hardening law for supercooled austenite of SAE 52100 steel," Computational Materials Science, vol. 50, no. 2, pp. 487-495, 2010.