在精密金屬加工產業中，若可透過刀具診斷模型以確保刀具的可用性，將可維護加工工件之品質。然而，在工件昂貴且品質嚴格要求的情況下，於實際場域中實難以獲得足夠且具代表性之樣本以建立刀具診斷模型。
因此，本研究提出一種漸進式更新機制(Gradual Refreshing Scheme, GRS)，其可用於建模、運行和更新刀具診斷模型。藉由本研究所提出之樣本擴增法(Sample Extension Method, SEM)，在少量且狀態侷限的實際樣本情況下，可於建模初期產生擴充的虛擬樣本，以先行建立可用模型，如此除可縮短蒐集建模所需樣本的時間外，亦可達到足夠的刀具診斷準確性。然而，在模型使用期，更可藉由評估指標的指示，以實際樣本更新已有模型，如此即可達到刀具診斷模型之適性化目標。
在研究成果上，以航太產業之發動機機匣加工為例，在少數可用樣本下，應用樣本擴增法所建立之診斷模型，針對刀具異常狀態分類的F1值達到0.947。而基於此足夠的刀具狀態辨別能力，將可避免不必要的更換刀具。如此，即可在確保加工之品質的條件下，可將現行刀具利用率提升10-30%，以有效降低加工成本。

In the precision metal manufacturing industry, machining quality can be ensured by evaluating and maintaining tool availability based on a tool diagnosis model. However, it is hard to collect sufficient and representative samples for modeling while facing expensive workpieces with strict quality requirement.
Hence, this dissertation proposes a Gradual Refreshing Scheme (GRS) for modeling, running, and refreshing tool diagnosis models to solve this problem. In the modeling phase, a sample extension method (SEM) is presented to generate dummy samples for building model early, so that the time for modeling can be reduced while enhancing the accuracy of tool diagnosis. In the running phase, the existing model can be refreshed with actual samples according to indication of the evaluating indices for achieving the goal of adaptive diagnosis model.
The examples of engine-case machining are adopted in this research to illustrate that the F1 value of classifying tool abnormal state is 0.947 when adopting the diagnosis model using SEM with few available samples. Unnecessary tool changes may be avoided based on the adequate capability of tool state classification. By ensuring machining quality, tool utilization can be further improved by 10-30%, so as to reduce machining cost efficiently.

[1] D. E. Dimla, “Sensor signals for tool-wear monitoring in metal cutting operations—a review of methods,” International Journal of Machine Tools and Manufacture, vol. 40, no. 8, pp. 1073–1098, 2000.
[2] T. Mikołajczyk, K. Nowicki, A. Bustillo and D. Yu Pimenov, “Predicting tool life in turning operations using neural networks and image processing,” Mechanical Systems and Signal Processing, vol. 104, pp. 503-513, 2018.
[3] M. Bhuiyan, I. Choudhury, M. Dahari, Y. Nukman and S. Dawal, “Application of acoustic emission sensor to investigate the frequency of tool-wear and plastic deformation in tool condition monitoring,” Measurement, vol. 92, pp. 208-217, 2016.
[4] E.-Y. Heo, H. Lee, C.-S. Lee, D.-W. Kim, and D. Y. Lee, “Process monitoring technology based on virtual machining,” Procedia Manufacturing, vol. 11, pp. 982–988, 2017.
[5] H. S. Cho, J.-H. Han, S.-Y. Chi, and K.-H. Yoo, “A tool breakage detection system using load signals of spindle motors in CNC machines,” in Proc. 2016 Eighth International Conference on Ubiquitous and Future Networks (ICUFN), 2016.
[6] E. Ahearne et al. “Tool-wear in milling of medical grade cobalt chromium alloy-requirements for advanced process monitoring and data analytics,” The Machine Tool Technologies Research Foundation (MTTRF) and iAM-CNC Annual Meeting 2016, San Fransisco, California, USA, 5-7 July 2016. Machine Tool Technologies Research Foundation, 2016.
[7] L. C. Lee, K. S. Lee, and C. S. Gan, “On the correlation between dynamic cutting force and tool-wear,” International Journal of Machine Tools and Manufacture, vol. 29, no. 3, pp. 295-303, 1989.
[8] S. S. Rangwala, “Machining process characterization and intelligent tool condition monitoring using acoustic emission signal analysis,” 1990.
[9] D. Yan, T. I.El-Wardany, and M. A. Elbestawi, “A multi-sensor strategy for tool failure detection in milling,” International Journal of Machine Tools and Manufacture, vol. 35, no. 3, pp. 383-398, 1995.
[10] D. Dornfeld, “Application of acoustic emission techniques in manufacturing,” Ndt & E International, vol. 25, no. 6, pp. 259-269, 1992.
[11] A, D. Dornfeld, “In process recognition of cutting states,” JSME international journal. Ser. C, Dynamics, control, robotics, design and manufacturing, vol. 37, no. 4, pp. 638-650, 1994.
[12] P. Kulandaivelu and P. S. Kumar, “Investigate the Crater Wear Monitoring of Single Point Cutting Tool Using Adaptive Neuro Fuzzy Inference System,” Journal of Applied Science and Engineering, vol. 15, no. 3, pp. 265-274, 2012.
[13] K. Zhu, Y. S. Wong, and G. S. Hong, “Wavelet analysis of sensor signals for tool condition monitoring: A review and some new results,” International Journal of Machine Tools and Manufacture, vol. 49, no. 7-8, pp. 537-553, 2009.
[14] S. E. Oraby and D. R. Hayhurst, “Tool life determination based on the measurement of wear and tool force ratio variation,” International Journal of Machine Tools & Manufacture, vol. 44, pp. 1261-1269, 2004.
[15] N. Constantinides and S. Bennett, “An investigation of methods for the on-line estimation of tool-wear,” International Journal of Machine Tools and Manufacture, vol. 27, no. 2, pp. 225-237, 1987.
[16] D. Shi and N. N Gindy, “Tool-wear predictive model based on least squares support vector machines,” Mechanical Systems and Signal Processing, vol. 21, no. 4, pp. 1799-1814, 2007.
[17] T. H. Miguel and R. D. Carlos, “FPGA-Based Fused Smart- Sensor for Tool- Wear Area Quantitative Estimation in CNC Machine Inserts” Sensors, vol. 10, pp. 3373-3388, 2010.
[18] G. F. Wang, Z. W. Guo and L. Qian, “Tool-wear prediction considering uncovered data based on partial least square regression,” Journal of Mechanical Science and Technology, vol. 28 , pp. 317-322, 2014.
[19] M. E. Nakai, P. R. Aguiar, H. Guillardi, E. C. Bianchi, D. H. Spatti, and D. M. D’Addona, “Evaluation of neural models applied to the estimation of tool-wear in the grinding of advanced ceramics,” Expert Systems with Applications, vol. 42, no. 20, pp. 7026–7035, 2015.
[20] S. Wold, M. Sjöström, and L. Eriksson, “PLS-regression: a basic tool of chemometrics,” Chemometrics and Intelligent Laboratory Systems, vol. 58, no. 2, pp. 109–130, 2001.
[21] H.-C. Yang, Y.-Y. Li, M.-H. Hung, and F.-T. Cheng, “A cyber-physical scheme for predicting tool-wear based on a hybrid dynamic neural network,” Journal of the Chinese Institute of Engineers, vol. 40, no. 7, pp. 614–625, Mar. 2017.
[22] L. Guo, N. Li, F. Jia, Y. Lei and J. Lin, “A recurrent neural network based health indicator for remaining useful life prediction of bearings,” Neurocomputing, vol. 240, pp. 98-109, 2017.
[23] C. Drouillet, J. Karandikar, C. Nath, A. Journeaux, M. El Mansori and T. Kurfess, “Tool life predictions in milling using spindle power with the neural network technique,” Journal of Manufacturing Processes, vol. 22, pp. 161-168, 2016.
[24] J. Gokulachandran and K. Mohandas, “Comparative study of two soft computing techniques for the prediction of remaining useful life of cutting tools,” Journal of Intelligent Manufacturing, vol. 26, no. 2, pp. 255–268, April. 2015.
[25] Z. Khan, M. Hayat, and M. Khan, “Discrimination of acidic and alkaline enzyme using Chou’s pseudo amino acid composition in conjunction with probabilistic neural network model,” Journal of Theoretical Biology, vol. 365, pp. 197-203, 2015.
[26] T. Saito and M. Rehmsmeier, “The precision-recall plot is more informative than the ROC plot when evaluating binary classifiers on imbalanced datasets,” Plos One, vol. 10, no. 3, Apr. 2015.
[27] A. Bustillo and J. J. Rodríguez, “Online breakage detection of multitooth tools using classifier ensembles for imbalanced data,” International Journal of Systems Science, vol. 45, no. 12, pp. 2590–2602, May 2013.
[28] J. G. Lei, M. Sell, A. Rinaldo, R. J. Tibshirani, and L. Wasserman, “Distribution-free predictive inference for regression,” Journal of the American Statistical Association, pp. 1-18, 2018.
[29] Y. Bengio, et al., “Out-of-sample extensions for lle, isomap, mds, eigenmaps, and spectral clustering,” Advances in neural information processing systems, 2004.
[30] B. Pan, W. Chen, B. Chen, C. Xu and J. Lai, “Out-of-Sample Extensions for Non-Parametric Kernel Methods,” IEEE Transactions on Neural Networks and Learning Systems, vol. 28, no. 2, pp. 334-345, 2017.
[31] L. H. Vanegas and G. A. Paula, “An extension of log-symmetric regression models: R codes and applications,” Journal of Statistical Computation and Simulation, vol. 86, no. 9, pp. 1709–1735, Nov. 2015.
[32] J. J. Faraway, “Extending the linear model with R: generalized linear, mixed effects and nonparametric regression models,” vol. 124. CRC press, 2016.
[33] F.-T. Cheng, Y.-T. Chen, Y.-C. Su, and D. L. Zeng, “Evaluating reliance level of a virtual metrology system,” IEEE Trans. Semicond. Manuf., vol. 21, no. 1, pp. 92–103, Feb. 2008
[34] F.-T. Cheng, H.-C. Huang, and C.-A. Kao, “Developing an Automatic Virtual Metrology System,” IEEE Transactions on Automation Science and Engineering, vol. 9, no. 1, pp. 181–188, 2012.
[35] D. J. Hand and R. J. Till, “A simple generalisation of the area under the ROC curve for multiple class classification problems,” Machine learning, vol. 45, Issue 2, pp. 171-186, 2001.
[36] ISO 8688-2, “Tool life testing in milling — Part 2 End milling”, 1989.
[37] H. Abdi, “Partial Least Squares Regression and Projection on Latent Structure Regression (PLS Regression),” Wiley Interdisciplinary Rev.: Computational Statistics, vol. 2, no. 1, pp. 97-106, 2010.