臺灣博碩士論文加值系統

本研究以水冷片模組結合多重物理耦合分析以及能估算主軸頭切削點非線性(偏斜)熱變形的類神經網路分析模型，模擬調控一台國產車铣複合機的主軸頭結構溫度分布，目標為減少主軸頭結構因熱變形量不對稱導致熱彎曲而產生的偏斜誤差。在工具機運轉時，機台上的多個熱源會造成機台的結構溫度產生動態變化；同時，持續的加工條件改變也會造成機台熱源的發熱量產生變化，進而導致動態的機台熱變形，造成複雜的機台熱誤差，影響加工精度及穩定性。為了更有效地調控工具機台的熱誤差，本研究擬透過已經過基礎實機驗證的數值模擬分析模型，獲得機台結構關鍵溫度點的溫度分布及變化與切削點熱變位之間關聯性的模擬數據，並將觀測點溫度值以及所對應的切削點熱變位輸入類神經網路進行學習訓練，以建立一個可以快速且有效準確估測機台結構溫升 - 熱變形的模型。接著，透過此類神經網路模型的估測功能，於此機台主軸頭結構區域模擬配置多個小型的水冷片，經由調整水冷片的位置及個數，調控機台關鍵結構位置的溫度分布，進而控制切削點的熱變位量。
欲建立有效的類神經網路模型，需要透過實驗以及模擬分析取得足夠數量的學習樣本。經過大量有效、具代表性的適當樣本學習，類神經網路便有機會具備透過關鍵溫度點正確判斷機台結構熱變形的能力。本研究主要的創新性為結合人工智慧與機台熱平衡技術，達到智慧化的熱平衡系統優化設計，進而提升工具機台熱精度。本研究透過基礎實驗以及多重物理耦合模擬分析累積的溫升-熱變位數據，建立類神經網路模型。此類神經網路分析模型所估測的溫升 - 熱變位數據與訓練參數以外的個案模擬結果進行比對及驗證，以測試本研究建立的類神經網路模型應用於工具機主軸頭結構溫升 - 熱變位分析應用上的有效性。本研究並且運用此類神經網路模型尋找與切削點熱變位相關性最明顯的結構溫度關鍵點，以作為未來監控機台溫升-熱變位的溫度參考點設計配置的重要依據，並且提高溫升-熱變位以及熱平衡參數優化類神經網路模型的估測效率。
接著，本研究使用粒子群最佳化演算法結合類神經網路，找出最適當的水冷片配置。透過粒子群最佳化演算法計算得出改善熱變形響應效果最佳的水冷片配置位置，同時也嘗試透過粒子群最佳化演算法決定水冷片冷卻點的調控溫度，以達成適應式的機台溫度分布調控，有效且即時地改善工具機機台在動態運轉下的熱誤差。

關鍵字：工具機、熱變形、熱平衡、類神經網路、粒子群最佳化演算法

This study use a water-cooled film module with multiple physical coupling analysis and a neural network model that can estimate the nonlinear thermal distortion at the cutting point of the spindle head to estimate and control temperature distribution of a CNC lathe spindle head. In order to more effectively regulate the thermal errors of the machine tool, this study intends to obtain the simulation data of the key temperature points and thermal deformation of cutting points of the machine structure through numerical simulation analysis verified by the experiment of the real machine tool. Then, the data are input into the neural network for learning and training to establish a model that can quickly and effectively predict the temperature rise - heat deformation of the machine tool structure. In addition, this study uses the particle swarm optimization algorithm combined with the neural network to find the most appropriate water cooling film configuration.
Key words：Machine tool, Thermal deformation, Artificial neural network, Particle swarm optimization algorithm, Adaptive thermal balance technique

摘要 i
誌謝 vi
目錄 vii
表目錄 ix
圖目錄 x
符號表 xiii
第1章 緒論 1
1.1 前言 1
1.2 文獻回顧 2
1.3 研究動機與目的 4
第2章 理論簡介 6
2.1 工具機熱平衡技術概念 6
2.2 工具機熱、流、固耦合分析原理 6
2.3 熱流耦合統御方程式 6
2.4熱彈性力學方程式 8
第三章 數值模擬及最佳化分析 9
3.1 數值分析參數規劃 9
3.2 間接耦合模擬分析流程 10
3.3 邊界條件設定 12
3.4 類神經網路系統原理 15
3.5 粒子群最佳化演算法原理 19
第四章 結果與討論 23
4.1 局部溫度調控對機台熱變形效應模擬結果分析 23
4.2 類神經模型學習訓練 28
4.3 以粒子群最佳化演算法分析最佳溫控參數 92
第五章 結論 106
Reference 108

[1]J.Bryan, International status of thermal error research(1990), Annals of the CIRP, 1990.
[2]J. Zhang, P. Feng, C. Chen, D. Yu, and Z. Wu, A method for thermal performance modeling and simulation of machine tools, Int J Adv Manuf Technol, vol. 68, pp. 1517-1527, 2013.
[3]E. Creighton, A. Honegger, A. Tulsian, and D. Mukhopadhyay, Analysis of thermal errors in a high-speed micro-milling spindle, International Journal of Machine Tools ＆ Manufacture, vol. 50, pp. 386-393, 2010.
[4]P. Vanherck, J. Dehaes, and M. Nuttin, Compensation of thermal deformations in machine tools with neural nets, Computers in Industry, vol. 33, pp. 119-125, 1997.
[5]A. M. Abdulshahed, A. P. Longstaff, S. Fletcher, and A. Potdar, Thermal error modelling of a gantry-type 5-axis machine tool using a Grey Neural Network Model, Journal of Manufacturing Systems, vol. 41, pp. 130-142, 2016.
[6]J. Mayr, J. Jedrzejewski, E. Uhlmann, M. A. Donmez, W. Knapp, F. Hartig, et al., Thermal issues in machine tools, CIRP Annals - Manufacturing Technology, vol. 61, pp. 771-791, 2012.
[7]J.-J. Kim, Y. H. Jeong, and D.-W. C. , Thermal behavior of a machine tool equipped with linear motors, International Journal of Machine Tools ＆ Manufacture, vol. 44, pp. 749-758, 2004.
[8]V.-T. Than and J. H. Huang, Nonlinear thermal effects on high-speed spindle bearings subjected to preload, Tribology International, vol. 96, pp. 361-372, 2016.
[9]N. S. Mian, S. Fletcher, A. P. Longstaff, and A. Myers, Efficient estimation by FEA of machine tool distortion due to environmental temperature perturbations, Precision Engineering, vol. 37, pp. 372-379, 2013.
[10]H. Shi, C. Ma, J. Yang, L. Zhao, X. Mei, and G. Gong, Investigation into effect of thermal expansion on thermally induced error of ball screw feed drive system of precision machine tools, International Journal of Machine Tools and Manufacture, vol. 97, pp. 60-71, 2015.
[11]Z. Haitao, Y. Jianguo, and S. Jinhua, Simulation of thermal behavior of a CNC machine tool spindle, International Journal of Machine Tools and Manufacture, vol. 47, pp. 1003-1010, 2007.
[12]B. Tan, X. Mao, H. Liu, B. Li, S. He, F. Peng, et al., A thermal error model for large machine tools that considers environmental thermal hysteresis effects, International Journal of Machine Tools and Manufacture, vol. 82-83, pp. 11-20, 2014.
[13]M. Gebhardt, M. Ess, S. Weikert, W. Knapp, and K. Wegener, Phenomenological compensation of thermally caused position and orientation errors of rotary axes, Journal of Manufacturing Processes, vol. 15, pp. 452-459, 2013.
[14]M. Gebhardt, J. Mayr, N. Furrer, T. Widmer, S. Weikert, and W. Knapp, High precision grey-box model for compensation of thermal errors on five-axis machines, CIRP Annals - Manufacturing Technology, vol. 63, pp. 509-512, 2014.
[15]J. Mayr, M. Egeter, S. Weikert, and K. Wegener, Thermal error compensation of rotary axes and main spindles using cooling power as input parameter, Journal of Manufacturing Systems, vol. 37, pp. 542-549, 2015.
[16]P. Blaser, F. Pavliček, K. Mori, J. Mayr, S. Weikert, and K. Wegener, Adaptive learning control for thermal error compensation of 5-axis machine tools, Journal of Manufacturing Systems, vol. 44, pp. 302-309, 2017.
[17]E. Bitar-Nehme and J. R. R. Mayer, Modelling and compensation of dominant thermally induced geometric errors using rotary axes’ power consumption, CIRP Annals, 2018.
[18]M. Zbiec, Application of Neural Network in Simple Tool Wear Monitoring and Identification System in MDF Milling, Drvna Industrija, vol. 62(1), pp. 43-54, 2011.
[19]J.Yang, H.Shi, B.Feng, C.Ma, and X.Mei, Applying neural network based on fuzzy cluster pre-processing to thermal error modeling for coordinate boring machine Procedia CIRP vol. 17, pp. 698-703, 2014.
[20]S. Eskandari, B. Arezoo, and A. Abdullah, Positional, geometrical, and thermal errors compensation by tool path modification using three methods of regression, neural networks, and fuzzy logic, Int J Adv Manuf Technol, vol. 65, pp. 1635-1649, 2013.
[21]J. S. Chen, J. Yuan, and J. Ni, Thermal Error Modelling for Real-Time Error Compensation, Int J Adv Manuf Technol, vol. 12, pp. 266-275, 1996.
[22]N. Harsha, I. A. Kumar, K. S. R. Raju, and S. Rajesh, Prediction of Machinability characteristics of Ti6Al4V alloy using Neural Networks and Neuro-Fuzzy techniques, Materials Today: Proceedings, vol. 5, pp. 8454-8463, 2018.
[23]S.-L. Chen and Y. W. Jen, Data fusion neural network for tool condition monitoring in CNC milling machining, International Journal of Machine Tools ＆ Manufacture, vol. 40, pp. 381-400, 2000.
[24]R. L. Malghan, K. R. M. C, A. K. Shettigar, S. S. Rao, and R. J. D’Souza, Forward and reverse mapping for milling process using artificial neural networks, Data in Brief, vol. 16, pp. 114-121, 2018.
[25]ANSYS, ANSYS Fluent Theory Guide, 2013.
[26]Harris and T. A, Rolling Bearing Analysis, 2001.
[27]陳美惠, 車銑複合機主軸頭熱平衡優化設計分析與實機驗證, 機械工程學系研究所, 國立中興大學, 2017.
[28]F. P. INCROPERA, D. P. DEWITT, T. L. BERGMAN, and A. S. LAVINE, Fundamentals of heat and mass transfer: John Wiley ＆ Sons, 2005.
[29]W. S. Mcculloch and W. Pitts, a logical calculus of the ideas immanent in nervous activity, bulletin of mathematical biophysics, vol. 5, p. 19, 1943.
[30]J. Kennedy and R. Eberhart, Partical Swarm Optimization, Neural Networks, 1995. Proceedings.,IEEE International Conference, pp. 1942-1948, 1995.