銑削加工為機械加工技術的核心。在機械加工過程中一旦刀具崩刃或磨損嚴重，即須立即檢知並加以處理，否則工件因加工尺寸產生誤差或表面品質不良，將導致產品品質低落。過去的研究大多利用力量計量測銑削力以判斷刀具磨耗的程度，然而因為其價格昂貴且受限於加工工法以及架設位置，並不具有商業應用價值。因此本研究在CNC銑床上架設麥克風，量測刀具磨耗時產生的音頻振動訊號。
此外過去研究只在固定切削條件下進行切削訊號分析，也沒有針對所選取之特徵指標進行篩選及驗證。本研究則探討在不同切削條件下音頻訊號與刀具磨耗間的關係，首先透過小波包分解產生數個時頻特徵指標，並使用共線性診斷及逐步迴歸分析進行特徵指標篩選。這些篩選後的特徵指標被用來建置統計迴歸模型及類神經網路模型，以預測刀具磨耗的程度。最後可以從預測模型之結果得知，本研究方法所提取之特徵指標並不會因為切削參數的改變而造成預測失準，且與過去研究使用的方均根指標相比有著較佳的準確率。

Milling plays a core role in the manufacturing industry. When a milling tool has severe wear or breakage during the manufacturing process, it requires immediate attention to prevent the precision error or unqualified surface quality which undermines the quality of the product. Most previous literatures applied the dynamometer to estimate the degree of tool wear, but this methodology has limited commercial application due to the expensive instrument and the inconvenient installation. Therefore, this research installed the microphone on the CNC machine to measure the acoustic signal of the tool wear generated in the milling operation.
In addition, previous studies only conducted the experiment under the specific cutting condition, and it neither selected nor validated the features of the signal to determine the tool wear. This research hence discussed the relationship between the acoustic signal and the tool wear under multiple cutting conditions. Firstly, the time-frequency statistic features generated from wavelet packet decomposition were evaluated by collinearity diagnostics and stepwise regression procedure. The selected features then were used to develop regression and artificial neural network models to predict the degree of tool wear. It could be concluded that the features used in this research would maintain the prediction accuracy regardless of different cutting parameters, and they have a better accuracy than the commonly-used root mean square value.

[1] Ghosh, N., et al. "Estimation of tool wear during CNC milling using neural network-based sensor fusion." Mechanical Systems and Signal Processing 21.1 (2007): 466-479.
[2] Palanisamy, P., I. Rajendran, and S. Shanmugasundaram. "Prediction of tool wear using regression and ANN models in end-milling operation." The International Journal of Advanced Manufacturing Technology 37.1-2 (2008): 29-41.
[3] Kaya, Bulent, Cuneyt Oysu, and Huseyin M. Ertunc. "Force-torque based on-line tool wear estimation system for CNC milling of Inconel 718 using neural networks." Advances in Engineering Software 42.3 (2011): 76-84.
[4] Nouri, Mehdi, et al. "Real-time tool wear monitoring in milling using a cutting condition independent method." International Journal of Machine Tools and Manufacture 89 (2015): 1-13.
[5] Li, H. Z., H. Zeng, and X. Q. Chen. "An experimental study of tool wear and cutting force variation in the end milling of Inconel 718 with coated carbide inserts." Journal of Materials Processing Technology 180.1 (2006): 296-304.
[6] Rivero, A., LN López de Lacalle, and Ma Luz Penalva. "Tool wear detection in dry high-speed milling based upon the analysis of machine internal signals." Mechatronics 18.10 (2008): 627-633.
[7] Orhan, Sadettin, et al. "Tool wear evaluation by vibration analysis during end milling of AISI D3 cold work tool steel with 35 HRC hardness." NDT & E International 40.2 (2007): 121-126.
[8] Haber, Rodolfo E., et al. "An investigation of tool-wear monitoring in a high-speed machining process." Sensors and Actuators A: Physical 116.3 (2004): 539-545.
[9] Kopač, Janez, and Samo Šali. "Tool wear monitoring during the turning process." Journal of Materials Processing Technology 113.1 (2001): 312-316.
[10] Duspara, Miroslav, Kristian Sabo, and Antun Stoic. "Acoustic emission as tool wear monitoring." Tehnicki Vjesnik-Technical Gazette 21.5 (2014): 1097-1101.
[11] Seemuang, N., T. McLeay, and T. Slatter. "Using spindle noise to monitor tool wear in a turning process." The International Journal of Advanced Manufacturing Technology 86.9-12 (2016): 2781-2790.
[12] Rafezi, Hamed, Javad Akbari, and Mehdi Behzad. "Tool Condition Monitoring based on sound and vibration analysis and wavelet packet decomposition." Mechatronics and its Applications (ISMA), 2012 8th International Symposium on. IEEE, 2012.
[13] Wang, G. F., et al. "Vibration sensor based tool condition monitoring using ν support vector machine and locality preserving projection." Sensors and Actuators A: Physical 209 (2014): 24-32.
[14] Chen, Baojia, et al. "Reliability estimation for cutting tools based on logistic regression model using vibration signals." Mechanical Systems and Signal Processing 25.7 (2011): 2526-2537.
[15] Aliustaoglu, Cuneyt, H. Metin Ertunc, and Hasan Ocak. "Tool wear condition monitoring using a sensor fusion model based on fuzzy inference system." Mechanical Systems and Signal Processing 23.2 (2009): 539-546.
[16] Zhang, Kai-feng, Hui-qun Yuan, and Peng Nie. "A method for tool condition monitoring based on sensor fusion." Journal of Intelligent Manufacturing 26.5 (2015): 1011-1026.
[17] Asiltürk, Ilhan, and Mehmet Çunkaş. "Modeling and prediction of surface roughness in turning operations using artificial neural network and multiple regression method." Expert systems with applications 38.5 (2011): 5826-5832.
[18] Özel, Tuğrul, and Yiğit Karpat. "Predictive modeling of surface roughness and tool wear in hard turning using regression and neural networks." International Journal of Machine Tools and Manufacture 45.4-5 (2005): 467-479.
[19] Kaya, Bulent, Cuneyt Oysu, and Huseyin M. Ertunc. "Force-torque based on-line tool wear estimation system for CNC milling of Inconel 718 using neural networks." Advances in Engineering Software 42.3 (2011): 76-84.
[20] Siddhpura, A., and R. Paurobally. "A review of flank wear prediction methods for tool condition monitoring in a turning process." The International Journal of Advanced Manufacturing Technology 65.1-4 (2013): 371-393.
[21] Snr, Dimla E. Dimla. "Sensor signals for tool-wear monitoring in metal cutting operations—a review of methods." International Journal of Machine Tools and Manufacture 40.8 (2000): 1073-1098.
[22] ISO 8688-2：1989, Tool life testing in milling. Part 2-End milling, International Standard.
[23] Machai, C., Biermann, D., 2011, “Machining of β-titanium-alloy Ti–10V–2Fe–3Al under cryogenic conditions: Cooling with carbon dioxide snow.” Journal of Materials Processing Technology, vol. 211, pp. 1175-1183.
[24] Merry, R. J. E., & Steinbuch, M. (2005). Wavelet theory and applications. literature study, Eindhoven university of technology, Department of mechanical engineering, Control systems technology group.
[25] O’brien, Robert M. "A caution regarding rules of thumb for variance inflation factors." Quality & Quantity 41.5 (2007): 673-690.
[26] Bendel, Robert B., and Abdelmonem A. Afifi. "Comparison of stopping rules in forward “stepwise” regression." Journal of the American Statistical association 72.357 (1977): 46-53.
[27] Wilamowski, Bogdan M., and Hao Yu. "Improved computation for Levenberg–Marquardt training." IEEE transactions on neural networks 21.6 (2010): 930-937.
[28] Heiple, Clinton R., et al. "Acoustic emission from single point machining: source mechanisms and signal changes with tool wear." Materials Evaluation;(United States) 52.5 (1994).
[29] Dimla, Dimla Eric. "The correlation of vibration signal features to cutting tool wear in a metal turning operation." The International Journal of Advanced Manufacturing Technology 19.10 (2002): 705-713.