「超音波刀具(Ultrasonic Cutting Tools)」全名為「蜂巢狀複合材料超音波輔助切割刀具」，是為航太領域中專門切割蜂巢結構(Honeycomb Structure)所研發之工具。此種刀具主要由碳化鎢所組成且價格昂貴，其加工方式為將刀具藉由螺紋鎖緊於超音波變幅桿末端，利用高頻超音波振動切割紙蜂巢結構。一般將超音波主軸安裝在CNC工具機上以加工複雜曲面。然而超音波刀具的可靠度參差不齊，造成換刀時機不易預測，常有刀具尚可切削卻遭淘汰，造成經濟效益低落。
本研究透過高靈敏度麥克風，擷取超音波刀具在無負載時所發出的音壓訊號，藉希爾伯特-黃轉換(Hilbert-Huang Transform)找出刀具之瞬時頻率，進而診斷超音波刀具潛在之可靠度。其中音壓擷取實驗使用自行設計之變幅桿(horn)作為量測的基準，藉由ANSYS的模態分析(Modal Analysis)與簡諧響應分析(Harmonic Response Analysis)做為其前導設計。利用希爾伯特邊際頻譜可擷取特徵找出其特徵參數，並使用支持向量機(Support Vector Machine, SVM)之機器監督式學習技術做為分類器，可判斷刀具的可靠度，除了可以當作交貨時的驗刀機制之外，並可預測刀具的使用期限，達到降低成本之目的。

Ultrasonic cutting tool is specifically used in the manufacturing of honeycomb structure in aerospace engineering. This tool made of expensive tungsten carbine is locked at the end of ultrasonic horn by screw thread, and it generates high frequency ultrasonic vibrations to cut the honeycomb structure. In general, the ultrasonic spindle is mounted on the CNC machine to work on complex surface structure. However, the reliability of the ultrasonic tool is unsteady and its life cycle is unpredictable. It renders low economic efficiency when an eligible tool is replaced in advance to prevent premature failure.
In this research, a high sensitive microphone is used to capture the sound pressure generated by the ultrasonic cutting tool without loading. The time-frequency spectrum generated by Hilbert-Huang transform (HHT) is applied to evaluate the potential reliability of the tool. To conduct the experiment, a customized horn is designed by using modal analysis and harmonic response analysis in ANSYS. The vibration features can be captured by marginal spectrum, and it uses support vector machine (SVM), a supervised learning algorithm, as the classifier to identify the reliability of the tool. The proposed mechanism not only predicts the reliably of the cutting tool, but also can be used as an evaluation standard for the new purchase. Hence the manufacturing cost can be reduced.

[1]Wood, Robert Williams; Loomis, Alfred L. xxxviii. the physical and biological effects of high-frequency sound-waves of great intensity. the london, edinburgh, and dublin philosophical magazine and journal of science, 1927.
[2]Balamuth, Lewis. ultrasonic surgical methods. u.s. patent no 3,862,630, 1975.
[3]Dam, Henrik; Quist, Per; Schreiber, Mads Peter. productivity, surface quality and tolerances in ultrasonic machining of ceramics. journal of materials processing technology, 1995.
[4]Thoe, T. B.; Aspinwall, D. K.; Wise, M.L.H. review on ultrasonic machining. international journal of machine tools and manufacture, 1998.
[5]Kang, Feng. a difference formulation based on the variational principle. appl. mathematics and comp. mathematics, 1965, 2.4: 238-262.
[6]Seah, K. H. W.; Wong, Y. S.; Lee, L. C. design of tool holders for ultrasonic machining using fem. journal of materials processing technology, 1993.
[7]Amin, S. G.; Ahmed, M. H. M.; Youssef, H. A. computer-aided design of acoustic horns for ultrasonic machining using finite-element analysis. journal of materials processing technology, 1995.
[8]Abdullah, Amir; Pak, Abbas. correct prediction of the vibration behavior of a high power ultrasonic transducer by fem simulation. the international journal of advanced manufacturing technology, 2008.

[9]Kuo, Kei-Lin,. ultrasonic vibrating system design and tool analysis. transactions of nonferrous metals society of china, 2009, 19: s225-s231.
[10]Naď, Milan. ultrasonic horn design for ultrasonic machining technologies.applied and computational mechanics, 2010.
[11]喻, et al. ANSYS 辅助楔形超声波聚能器的优化设计和实验研究. 声学技术, 2010.
[12]Choi, Young-Jae, et al. effect of ultrasonic vibration in grinding; horn design and experiment. international journal of precision engineering and manufacturing, 2013.
[13]吴欣, et al. 蜂窝复合材料超声辅助切割工具设计. 中国机械工程, 2015.
[14]Che,Jiangtao, et al. experimental study on horizontal ultrasonic electrical discharge machining. journal of materials processing technology, 2016.
[15]Cooley, James W.; Tukey,John W. an algorithm for the machine calculation of complex fourier series. mathematics of computation, 1965.
[16]Welch, Peter D. the use of fast fourier transform for the estimation of power spectra: a method based on time averaging over short, modified periodograms. ieee transactions on audio and electroacoustics, 1967.
[17]Lei, Yaguo; He, Zhengjia; Zi, Yanyang. application of the eemd method to rotor fault diagnosis of rotating machinery. mechanical systems and signal processing, 2009.
[18]Akiniwa, Yoshiaki; Tanaka, Keisuke; Nakatsu, Akira. evaluation of fatigue strength in very-long life regime of sncm439 steels. nippon kikai gakkai ronbunshu, a hen/transactions of the japan society of mechanical engineers, part a, 2004.
[19]林高田,超音波變幅桿設計於脂溶性成分萃取之研究，台灣科技大學碩士論文.2015.
[20]Daubechies, ingrid. the wavelet transform, time-frequency localization and signal analysis. information theory, ieee transactions on, 1990.
[21]Huang, Norden E., et al. the empirical mode decomposition and the hilbert spectrum for nonlinear and non-stationary time series analysis. in: proceedings of the royal society of london a: mathematical, physical and engineering sciences. the royal society, 1998.
[22]Malakooti, Behnam; Wang, Jun; Tandler, Evan C. a sensor-based accelerated approach for multi-attribute machinability and tool life evaluation. the international journal of production research, 1990.
[23]Rawson, F. F., High power ultrasonic resonant horns, part 1 – basic design concepts. velocity of ultrasound at 20 khz; effects of material and horn dimensions, ultrasonics int. 87 conf. proc., london, 1987.
[24]Cortes, Corinna; Vapnik, Vladimir. support-vector networks. machine learning, 1995.
[25]Cottle, Richard W. William Karush and the kkt theorem. doc. math., extra volume: optimization stories, 2012,
[26]Vapnik, Vladimir. the nature of statistical learning theory. springer science & business media, 2013.
[27]Boser, B. E.; Guyon, I. M.; Vapnik, V. N. (1992). "a training algorithm for optimal margin classifiers". proceedings of the fifth annual workshop on computational learning theory – colt '92.
[28]Jack, L. B.; Nandi, A. K. fault detection using support vector machines and artificial neural networks, augmented by genetic algorithms. mechanical systems and signal processing, 2002.
[29]Samanta, B.; al-balushi, K. R.; al-araimi, S. A. artificial neural networks and support vector machines with genetic algorithm for bearing fault detection. engineering applications of artificial intelligence, 2003,
[30]Konar, P.; Chattopadhyay, P. bearing fault detection of induction motor using wavelet and support vector machines (svms). applied soft computing, 2011.
[31]Wu, Tian Yau; Chen, J. C.; Wang, C. C. characterization of gear faults in variable rotating speed using hilbert-huang transform and instantaneous dimensionless frequency normalization. mechanical systems and signal processing, 2012.
[32]Wang, Y. S., et al. an intelligent approach for engine fault diagnosis based on hilbert–huang transform and support vector machine. applied acoustics, 2014.
[33]Chang, Chih-Chung; Lin, Chih-Jen. libsvm: a library for support vector machines. acm transactions on intelligent systems and technology (tist), 2011.
[34]Huang, Norden eh. hilbert-huang transform and its applications. world scientific, 2014.