本研究主要是透過電腦有限元素分析與實際應變量測進行跑步機車身 結構輕量化之最佳化設計分析，首先使用有限元素分析與實際應變量測進 行原始設計模擬分析並確認有限元素分析模型的有效性。之後搭配田口實 驗方法與灰關聯分析，探討跑步機車架於最大使用者重量負載時，在車架 避免發生塑性變形的前題下，進行同時具有最小等效應力值與最輕重量的 二個階段最佳化設計參數分析。再以第二階段獲得最佳化的結果進行實體 樣機製作，比較其有限元素模擬與實驗量測應力的差異。 研究結果顯示，經二階段設計參數最佳化的結果與原始設計比較，跑 步機車架重量由原始設計的68.7 kg 下降至60.2 kg，下降了12.4 %；雖然跑 步機車架結構的最大等效應力由原始設計的124 MPa 上升至203 MPa，上 升了63.7 %，並未超過材料的降伏強度 (245 MPa)，所以結構不會產生塑性 變形，顯示車架結構是安全無虞的。最後，依第二階段最佳化設計參數製 作實體樣機，並進行實驗應力量測，經實際應力量測結果與有限元素分析 結果比較，各量測點的誤差值皆小於11 %，表示經過二階段設計參數最佳 化的有限元素模擬分析模型是有效的。;The aim of this study is to optimize the design for reducing weight of frame structure in a treadmill through finite element analysis (FEA) and experimental strain measurement. Firstly, FEA simulation and strain measurement are conducted for the original design to confirm the effectiveness of the FEA model. After that, Taguchi method and gray relational analysis are applied in a two-stage optimization process to determine the optimal combination of design parameters for reducing the frame weight without plastic deformation. Finally, based on the results of the second-stage optimization, a prototype frame is made to validate the FEM modeling in optimization process by performing an experimental strain measurement under maximum allowable load. The results show that after the two-stage optimization of design parameters, the frame weight of the treadmill is reduced from the original design of 68.7 kg to 60.2 kg, namely a decrease of 12.4%. The maximum von-Mises equivalent stress in the frame structure of new design is increased from 124 MPa to 203 MPa, which is an increase of 63.7% but does not exceed the yield stress (245 MPa). Therefore, no plastic deformation is predicted for the new design, indicating that the frame structure is safe. A prototype frame is made according to the optimized design parameters for comparing the FEA simulation with the experimental results. It reveals that the difference of stress at each selected measurement point is less than 11% between simulation and experimental result. It is thus confirmed that the FEA model developed for the two-stage optimization process is effective.
