隨著智慧製造的發展，自動化及智慧化成為製造產業升級重要的階段。數值控制(CNC)工具機亦為主要升級的目標之一，而系統的優化製程參數與準確預測製程反應可以使整體的製造效率大幅提升，並延長機台及刀具壽命與減少重複加工所造成的成本。雖然現在已有許多電腦輔助設計與製造(CAD、CAM)系統有自動優化製程參數的功能，然而，其結果無法因不同機台的性能差異而調整。因此，如何以加裝感測器與利用人工智慧的技術預測並優化製程參數為許多公司與研究欲解決的問題。
基於感測器的預測控制方法雖然能有效地解決優化製程參數以提升生產效率，但是其預測模型的準確度卻是影響此方法有效性的關鍵因素，除此之外，如何有效地評估模型的好壞亦是重要的一環。
本研究結合集成學習與神經網路發展電流預測模型，並將其作為預測控制的依據，其中集成學習包含嶺回歸、支援向量機、隨機森林與梯度提升決策樹等模型，並以結合注意力集中機制的神經網路進行非線性組合以提升預測準確率。此外，並以基於產品型號的交叉驗證方式，作為模型是否可進入下一階段的依據。本研究架構包含問題定義、資料準備、模型建構與評估、優化CNC 工具機的參數、專家評估與決策。研究結果可提供有效且穩健的電流預測模型及CNC 工具機參數優化的架構以提升CNC 工具機的生產效率。此外，本研究以一航太製造業為實證案例驗證提出架構之效度。驗證結果顯示本研究提出的參數設定能有效降低銑削時間並減少發生斷刀之次數。

Automation and intellectualization had become a vital stage in the upgrading of manufacturing industry with the development of intelligent manufacturing. Computer Numerical Control (CNC) is one of the subjects that people commit to working on. It is crucial for optimizing process parameters systematically and predicting process situation precisely that can enhance productivity make tool life longer and reduce the rework cost. CAD and CAM can help automatically optimize process parameters. However, existing techniques cannot adaptively optimize NC code under different machine condition. Consequently, how to use the additional sensor and artificial intelligence to predict and optimize process parameters is an important issue in many companies and studies.
This research aims to develop a current prediction model by ensemble learning with neural network and it is a reference model in predictive-control method. Ridge regression, SVR, random forest, and gradient boosting tree are base learner model in ensemble learning. Then, in order to increase accuracy, attention-based neural network is a meta-learner to make a non-linear combination with base learner. Additionally, this study applies k-fold cross validation with product ID to evaluate models’ accuracy. This research framework includes problem definition, data preparation, model construction and evaluation, parameters of CNC milling machine optimization, and tradeoff and decision. Sensor-based productivity-control method can solve the issue of process parameter optimization, but the accuracy of prediction model is a key factor to impact the effectiveness of this approach. Besides, model evaluation is also an important step for this approach.
The result of this study provides an effective and robust current prediction model and a process parameter optimization framework for CNC milling machine to improve productivity. Furthermore, we validate the proposed framework with the an aerospace manufacturing company as an empirical study. The result shows that the parameters setting by this framework can improve milling time and decrease a risk of broken tool significantly.

Table of Contents i
List of Tables iii
List of Figures iv
Chapter 1 Introduction 1
1.1 Background and Motivation 1
1.2 Research Objectives 2
1.3 Thesis Organization 3
Chapter 2 Literature Review 4
2.1 CNC Milling Machine and Process Parameter Optimization 4
2.2 Prediction-Based Optimization Model 6
2.3 Adaptive Control Theory 9
2.4 Ensemble Learning 9
Chapter 3 Research Framework 11
3.1 Problem Definition 13
3.2 Data Preparation 13
3.2.1 Data Collection 13
3.2.2 Data Cleaning 14
3.2.3 Feature Extraction 15
3.3 Model Construction and Evaluation 15
3.3.1 Ridge, SVR, Random Forest, and Gradient Boosting Tree 17
3.3.2 Neural Network 18
3.3.3 Hyper-Parameter Tuning 19
3.3.4 Model Evaluation 20
3.4 Parameters of CNC Milling Machine Optimization 21
3.5 Tradeoff and Decision 21
Chapter 4 Empirical Study 23
4.1 Problem Definition and Background 23
4.2 Data Preparation 24
4.3 Hyper-Parameter Setting 26
4.4 Current Prediction Model Construction and Validation Results 27
4.5 Evaluation and Interpretation 29
Chapter 5 Conclusion and Future Research 31
5.1 Conclusion 31
5.2 Future Research 31
Reference 33

Akyuz, A. O., Uysal, M., Bulbul, B. A., and Uysal, M. O. (2017), "Ensemble approach for time series analysis in demand forecasting: Ensemble learning," Proceedings of 2017 IEEE International Conference on INnovations in Intelligent SysTems and Applications (INISTA).
Anifowose, F. A., Labadin, J., and Abdulraheem, A. (2017), "Ensemble machine learning: An untapped modeling paradigm for petroleum reservoir characterization," Journal of Petroleum Science and Engineering, Vol. 151, No., pp. 480-487.
Asiltürk, I., Neşeli, S., and Ince, M. A. (2016), "Optimisation of parameters affecting surface roughness of Co28Cr6Mo medical material during CNC lathe machining by using the Taguchi and RSM methods," Measurement, Vol. 78, No., pp. 120-128.
Beudaert, X., Lavernhe, S., and Tournier, C. (2012), "Feedrate interpolation with axis jerk constraints on 5-axis NURBS and G1 tool path," International Journal of Machine Tools and Manufacture, Vol. 57, No., pp. 73-82.
Bort, C. M. G., Leonesio, M., and Bosetti, P. (2016), "A model-based adaptive controller for chatter mitigation and productivity enhancement in CNC milling machines," Robotics and Computer-Integrated Manufacturing, Vol. 40, No., pp. 34-43.
Breiman, L. (1996), "Bagging predictors," Machine learning, Vol. 24, No. 2, pp. 123-140.
Budak, E., Lazoglu, I., and Guzel, B. (2004), "Improving cycle time in sculptured surface machining through force modeling," CIRP Annals, Vol. 53, No. 1, pp. 103-106.
Chien, C.-F., Chou, C.-W., and Yu, H.-C. (2016), "A novel route selection and resource allocation approach to improve the efficiency of manual material handling system in 200-mm wafer fabs for industry 3.5," IEEE Transactions on Automation Science and Engineering, Vol. 13, No. 4, pp. 1567-1580.
Chien, C.-F., Hong, T.-Y., and Guo, H.-Z. (2017), "An empirical study for smart production for TFT-LCD to empower Industry 3.5," Journal of the Chinese Institute of Engineers, Vol. 40, No. 7, pp. 552-561.
Chien, C. F., Hong, T. Y., and Guo, H. Z. (2017). "A conceptual framework for “Industry 3.5” to empower intelligent manufacturing and case studies. "Procedia Manufacturing, 11, 2009-2017.
Chien, C.-F., Lin, Y.-S., and Lin, S.-K. (2020), "Deep reinforcement learning for selecting demand forecast models to empower Industry 3.5 and an empirical study for a semiconductor component distributor," International Journal of Production Research, Vol., No., pp. 1-21.
Chon, Y., Talipov, E., Shin, H., and Cha, H. (2011), "Mobility prediction-based smartphone energy optimization for everyday location monitoring," Proceedings of Proceedings of the 9th ACM conference on embedded networked sensor systems.
Cook, D. F., Ragsdale, C. T., and Major, R. (2000), "Combining a neural network with a genetic algorithm for process parameter optimization," Engineering applications of artificial intelligence, Vol. 13, No. 4, pp. 391-396.
Dhiman, H. S., Deb, D., and Guerrero, J. M. (2019), "Hybrid machine intelligent SVR variants for wind forecasting and ramp events," Renewable and Sustainable Energy Reviews, Vol. 108, No., pp. 369-379.
Dietterich, T. G. (2000), "Ensemble methods in machine learning," Proceedings of International workshop on multiple classifier systems.
Dietterich, T. G. (2002), "Ensemble learning," The handbook of brain theory and neural networks, Vol. 2, No., pp. 110-125.
Erdim, H., Lazoglu, I., and Ozturk, B. (2006), "Feedrate scheduling strategies for free-form surfaces," International Journal of Machine Tools and Manufacture, Vol. 46, No. 7-8, pp. 747-757.
Erkorkmaz, K., Layegh, S. E., Lazoglu, I., and Erdim, H. (2013), "Feedrate optimization for freeform milling considering constraints from the feed drive system and process mechanics," CIRP Annals, Vol. 62, No. 1, pp. 395-398.
Freitas, F. D., De Souza, A. F., and de Almeida, A. R. (2009), "Prediction-based portfolio optimization model using neural networks," Neurocomputing, Vol. 72, No. 10-12, pp. 2155-2170.
Freund, Y. (1995), "Boosting a weak learning algorithm by majority," Information and computation, Vol. 121, No. 2, pp. 256-285.
Hossain, M. S. J. and Liao, T. W. (2017), "Cutting Parameter Optimization for End Milling Operation Using Advanced Metaheuristic Algorithms," Vol., No., pp.
Ip, R. W., Lau, H. C., and Chan, F. T. (2003), "An economical sculptured surface machining approach using fuzzy models and ball-nosed cutters," Journal of materials processing technology, Vol. 138, No. 1-3, pp. 579-585.
Joshi, A. and Kothiyal, P. (2013), "Investigating effect of machining parameters of CNC milling on surface finish by taguchi method," International Journal on Theoretical and Applied Research in Mechanical Engineering, Vol. 2, No. 2, pp. 113-119.
Karuppanan, B. R. C. and Saravanan, M. (2019), "Optimized sequencing of CNC milling toolpath segments using metaheuristic algorithms," Journal of Mechanical Science and Technology, Vol. 33, No. 2, pp. 791-800.
Khanghah, S. P., Boozarpoor, M., Lotfi, M., and Teimouri, R. (2015), "Optimization of micro-milling parameters regarding burr size minimization via RSM and simulated annealing algorithm," Transactions of the Indian Institute of Metals, Vol. 68, No. 5, pp. 897-910.
Ku, C.-C., Chien, C.-F., and Ma, K.-T. (2020), "Digital transformation to empower smart production for Industry 3.5 and an empirical study for textile dyeing," Computers & Industrial Engineering, Vol. 142, No., pp. 106297.
Kumbhar, A., Bhosale, R., Modi, A., Jadhav, S., Nipanikar, S., and Kulkarni, A. (2015), "Multi-objective optimization of machining parameters in CNC end milling of stainless steel 304," International Journal of Innovative Research in Science, Engineering and Technology, Vol. 4, No. 9, pp. 8419-8426.
Lazoglu, I., Boz, Y., and Erdim, H. (2011), "Five-axis milling mechanics for complex free form surfaces," CIRP annals, Vol. 60, No. 1, pp. 117-120.
Lee, W., Kim, S. H., Park, J., and Min, B.-K. (2017), "Simulation-based machining condition optimization for machine tool energy consumption reduction," Journal of cleaner production, Vol. 150, No., pp. 352-360.
Li, K., He, S., Li, B., Liu, H., Mao, X., and Shi, C. (2020), "A novel online chatter detection method in milling process based on multiscale entropy and gradient tree boosting," Mechanical Systems and Signal Processing, Vol. 135, No., pp. 106385.
Liang, H., Song, L., and Li, X. (2019), "The Rotate Stress of Steam Turbine Prediction Method Based on Stacking Ensemble Learning," Proceedings of 2019 IEEE 19th International Symposium on High Assurance Systems Engineering (HASE).
Lughofer, E., Zavoianu, A.-C., Pratama, M., and Radauer, T. (2019), "Automated Process Optimization in Manufacturing Systems Based on Static and Dynamic Prediction Models," in: (eds.), Predictive Maintenance in Dynamic Systems, Springer, pp. 485-531.
Madić, M., Radovanović, M., Manić, M., and Trajanović, M. (2014), "Optimization of ANN models using different optimization methods for improving CO 2 laser cut quality characteristics," Journal of the Brazilian Society of Mechanical Sciences and Engineering, Vol. 36, No. 1, pp. 91-99.
Mahdavi, I. and Shirazi, B. (2010), "Decision Support System Architecture for the Adaptive," Journal of Artificial Intelligence, Vol. 3, No. 4, pp. 201-219.
Mahesh, T. P. and Rajesh, R. (2014), "Optimal selection of process parameters in CNC end milling of Al 7075-T6 aluminium alloy using a Taguchi-fuzzy approach," Procedia Materials Science, Vol. 5, No., pp. 2493-2502.
Ndiaye, E., Le, T., Fercoq, O., Salmon, J., and Takeuchi, I. (2018), "Safe grid search with optimal complexity," arXiv preprint arXiv:1810.05471, Vol., No., pp.
Pare, V., Agnihotri, G., and Krishna, C. (2015), "Selection of optimum process parameters in high speed CNC end-milling of composite materials using meta heuristic techniques--a comparative study/Izbira optimalnih parametrov procesa visokohitrostnega CNC-rezkanja kompozitnih materialov z metahevristicnimi tehnikami--primerjalna studija," Strojniski Vestnik-Journal of Mechanical Engineering, Vol. 61, No. 3, pp. 176-187.
Peng, X. (2010), "TSVR: an efficient twin support vector machine for regression," Neural Networks, Vol. 23, No. 3, pp. 365-372.
Radzevich, S. P. (2012), Dudley's handbook of practical gear design and manufacture. CRC Press.
Rao, H. P., Dave, R., Thakore, R., Rao, H. P., Dave, R., and Thakore, R. (2017), "Optimization of CNC End Milling Process Parameters for Aluminium 6061 Alloy using Carbide Tool Material by Design of Experiments," International Journal, Vol. 3, No., pp. 84-90.
Seni, G. and Elder, J. F. (2010), "Ensemble methods in data mining: improving accuracy through combining predictions," Synthesis lectures on data mining and knowledge discovery, Vol. 2, No. 1, pp. 1-126.
Shmueli, G. (2010), "To explain or to predict?," Statistical science, Vol. 25, No. 3, pp. 289-310.
Silva, J. A., Abellán-Nebot, J. V., Siller, H. R., and Guedea-Elizalde, F. (2014), "Adaptive control optimisation system for minimising production cost in hard milling operations," International Journal of Computer Integrated Manufacturing, Vol. 27, No. 4, pp. 348-360.
Soleimani, R., Mahmood, T., and Bahadori, A. (2016), "Assessment of compressor power and condenser duty per refrigeration duty in three-stage propane refrigerant systems using a new ensemble learning tool," Chemeca 2016: Chemical Engineering-Regeneration, Recovery and Reinvention, Vol., No., pp. 23.
Theja, K. D., Gowd, G. H., and Kareemulla, S. (2013), "Prediction & optimization of end milling process parameters using artificial neural networks," International Journal of Emerging Technology and Advanced Engineering, Vol. 3, No. 9, pp. 117-122.
Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A. N., Kaiser, Ł., and Polosukhin, I. (2017), "Attention is all you need," Proceedings of Advances in neural information processing systems.
Wang, A., Sha, M., Liu, L., and Chu, M. (2015), "A new process industry fault diagnosis algorithm based on ensemble improved binary-tree SVM," Chinese Journal of Electronics, Vol. 24, No. 2, pp. 258-262.
Wang, W. P. (1988), "Solid modeling for optimizing metal removal of three-dimensional NC end milling," Journal of Manufacturing Systems, Vol. 7, No. 1, pp. 57-65.
Weber, J., Boxnick, S., and Dangelmaier, W. (2014), "Experiments using meta-heuristics to shape experimental design for a simulation-based optimization system: intelligent configuration and setup of virtual tooling," Proceedings of Asia-Pacific World Congress on Computer Science and Engineering.
Weber, J., Mueß, A., and Dangelmaier, W. (2017), "A simulation based optimization approach for setting-up CNC machines," in: (eds.), Operations Research Proceedings 2015, Springer, pp. 445-451.
Willmott, C. J. and Matsuura, K. (2005), "Advantages of the mean absolute error (MAE) over the root mean square error (RMSE) in assessing average model performance," Climate research, Vol. 30, No. 1, pp. 79-82.
Wolpert, D. H. (1992), "Stacked generalization," Neural networks, Vol. 5, No. 2, pp. 241-259.
Yang, Z. M., Djurdjanovic, D., and Ni, J. (2008), "Maintenance scheduling in manufacturing systems based on predicted machine degradation," Journal of intelligent manufacturing, Vol. 19, No. 1, pp. 87-98.
Yousefian, O., Balabokhin, A., and Tarbutton, J. (2020), "Point-by-point prediction of cutting force in 3-axis CNC milling machines through voxel framework in digital manufacturing," Journal of Intelligent Manufacturing, Vol. 31, No. 1, pp. 215-226.
Zenko, B., Todorovski, L., and Dzeroski, S. (2001), "A comparison of stacking with meta decision trees to bagging, boosting, and stacking with other methods," Proceedings of Proceedings 2001 IEEE International Conference on Data Mining.
Zhang, C. and Jiang, P. (2019), "Sustainability evaluation of process planning for single CNC machine tool under the consideration of energy-efficient control strategies using random forests," Sustainability, Vol. 11, No. 11, pp. 3060.
Zhang, D. and Wei, B. (2017), "A review on model reference adaptive control of robotic manipulators," Annual Reviews in Control, Vol. 43, No., pp. 188-198.
Zhao, D., Li, X., and Yang, J. (2018), "An improved wolf pack algorithm for sustainable machining parameter optimization," Proceedings of 2018 IEEE 3rd Advanced Information Technology, Electronic and Automation Control Conference (IAEAC).
Zhou, Q., Rong, Y., Shao, X., Jiang, P., Gao, Z., and Cao, L. (2018), "Optimization of laser brazing onto galvanized steel based on ensemble of metamodels," Journal of Intelligent Manufacturing, Vol. 29, No. 7, pp. 1417-1431.