運用多目標模型於車削加工參數最佳化之研究

Research on Turning Process Parameters Optimization Using Multi-Objective Model

呂念庭(Nien-Ting Lu)

指導教授 ： 黃博滄

中原大學/電機資訊學院/工業與系統工程研究所/碩士(2021年)

https://doi.org/10.6840/cycu202100971

摘要

CNC是目前廣泛應用於機械製造系統的一種加工方法，合理的選擇加工參數可顯著的降低加工成本，實現高品質與高效率至關重要，為了達到表面粗糙度越低越好與材料去除率越高越好作為目標，因此本研究應用CNC數值控制精密車床對6061鋁合金材料進行外徑車削加工製程的參數最佳化，選訂主軸轉速、進給速率、切削深度為切削之參數，並探討多目標與加工參數之關係。
依照實驗設計所規劃完成數據蒐集，並運用多目標最佳化中的非優勢排序遺傳演算法(Non-dominate Sorting Genetic Algorithm , NSGA-II)進行建模，而目標要具備對應之加工參數的關係式，因此藉由關聯性分析來了解加工參數的重要影響因子，並使用迴歸分析(Regression Analysis)，建立表面粗糙度與材料去除率的迴歸模型，投入NSGA-II進行運算。
本研究得出參數最佳化下多目標的結果為表面粗糙度0.861(um)、材料去除率8,120.2(mm3/min)，參數設定為主軸轉速2482rpm、進給速率0.1mm/rev、切削深度0.3mm，最後本運用不同刀具進行驗證，證明其模型的可行性。

關鍵字

CNC車削加工 ； 表面粗糙度 ； 材料去除率 ； 多目標最佳化 ； 非優勢排序遺傳演算法

並列摘要

CNC is currently a processing method widely used in machinery manufacturing systems. A good selection of processing parameters can significantly reduce processing costs. It is essential to achieve high quality and high efficiency. To achieve the lower the surface roughness, the better and the higher the material removal rate. This research aims to find the optimal parameters of the 6061 aluminum alloy material outside the diameter turning process. The higher, the better, so this research uses CNC numerical control precision lathe to set the spindle speed, feed rate, and depth of cut as the cutting parameters and discuss the relationship between surface roughness, material removal rate, and processing parameters.
Complete data were collected according to the experiment's design and used the Non-dominate Sorting Genetic Algorithm (NSGA-II) in multi-object optimization for modeling. The target must have the relationship of the corresponding processing parameters. Therefore, the correlation analysis is used to understand the important influencing factors of the processing parameters, and regression analysis is used to establish the regression model of surface roughness and material removal rate and input NSGA-II for calculation.
In this study, the results of multi-objective under parameter optimization were surface roughness 0.861 (um), material removal rate 8,120.2 (mm3/min), and the parameters were set to spindle speed 2482 rpm and feed rate 0.1 mm/rev. The depth of the cut was 0.3 mm. Finally, this study used different tools to verify and prove the feasibility of the model.

並列關鍵字

CNC Turning ； Surface Roughness ； Material Removal Rate ； Multi-objective ； NSGA-II

參考文獻

參考文獻

Amouzgar, K., Ng, A. H., Ljustina, G. (2020). Optimizing index positions on CNC tool magazines considering cutting tool life and duplicates. Procedia CIRP, 93, 1508-1513.

Arunnath, A., Masooth, P. H. S. (2021). Optimization of process parameters in CNC turning process on machining SCM440 steel by uncoated carbide and TiCN/Al2O3/TiN coated carbide tool under dry conditions. Materials Today: Proceedings, 45, 6253-6269.

Bagaber, S. A., Yusoff, A. R. (2019). Energy and cost integration for multi-objective optimisation in a sustainable turning process. Measurement, 136, 795-810.

Bensmaine, A., Dahane, M., Benyoucef, L. (2013). A non-dominated sorting genetic algorithm based approach for optimal machines selection in reconfigurable manufacturing environment. Computers Industrial Engineering, 66(3), 519-524.