基於零階與包絡法進行刀具銑削穩定性預測及顫振分析

Chatter Analysis and Stability Prediction of Milling Tool Based on Zero-order and Envelope Methods

陳忠誠(Chung-Cheng Chen)

指導教授 ： 張文陽

國立虎尾科技大學/工程學院/機械與電腦輔助工程系碩士班/碩士(2017年)

摘要

近年來隨著工業技術進步迅速，高速加工已逐漸成為主流趨勢，然而加工時的穩定性會隨機台加工速度上升而下降，以往只依靠經驗進行加工的方式已無法因應目前多變的加工條件。顫振預測技術是以數值計算的方式避開不穩定的轉速區間，藉此獲得穩定的切削條件。此項技術目前仍屬國外發展較成熟，國內應用上仍屬少數，其原因在於技術門檻較高，演算法複雜、切削參數取得不易、計算時間較長等因素，因此多半無法廣泛地應用於實際製程中。有鑑於此，本研究擬開發一套簡易且快速之銑削顫振預測系統，提供加工條件選擇上的一項參考依據。研究主要分為三個部分，其一是以再生性顫振理論與系統動態響應方程式建立出主軸轉速與切削深度之間的關係；其二是以曲線擬合的方式來估算刀具系統之模態參數；其三是以商用軟體LabVIEW開發一套銑削穩定性預測系統，其功能包含多模態參數辨識以及銑削顫振穩定曲線的計算。理論分析方面，本研究採用Altintas等人所提出之零階分析法(ZOA)來預測每一模態下之穩定性，再將多組模態所建立之穩定曲線以最低包絡的方式來獲得顫振穩定區域之包絡曲線，可快速、簡易地獲得預測結果。第二部分研究著重在模態參數的提取，本研究利用脈衝敲擊試驗(Impact test)來獲得分析所需的刀具系統頻率響應特性，並使用全域有理數多項式(GRFP)進行響應函數之曲線擬合，同時辨識出其阻尼值。最後本研究將預測出的顫振結果與實驗結果進行比較，以驗證銑削穩定性預測系統之準確度，並以相位頻譜、振動頻譜對加工時顫振訊號進行分析。實驗顯示，在徑向進給100%的條件下，十一個測試點中有三筆是與預測結果不相符的，初估預測準確率約72%。在相位差量測部分，在顫振發生時，其主導頻率之相位差大多大於60度且變動幅度也較大；而在穩定的情況下，其主導頻率大多落在刀具通過頻率及其倍頻上且異常頻率之相位差也大多不超過30度。

關鍵字

刀具顫振 ； 穩定耳垂圖 ； 再生性顫振 ； 零階分析法 ； 曲線包絡法

並列摘要

In recent years, with the rapid progress of industrial technology, high-speed machining has gradually become a trend, but for high-speed machining, the stability of the machine would decrease with spindle speed rising. The traditional experience method has not been able to deal with the variable processing conditions. The chatter prediction technique can be used to avoid the unstable speed area by numerical means, and obtaining the stable cutting conditions. Compared to domestic, this technology is more mature in foreign countries. The reason is that the technical standard is high, complex algorithm, the cutting parameters are not easy to obtain, and the calculation time is longer and so on, so most can not be widely used in the actual process. Therefore, this study intends to develop a simple and fast milling chatter prediction system that provides a basis for selection of machining conditions.This paper is divided into three parts. First, construction a relationship between the spindle speed and cutting depth by regenerative chatter theory and frequency respond function. Second, the modal parameters of the cutting tool system are estimated by curve fitting.Furthermore, we develop a processing stability prediction system with commercial software LabVIEW, The system include multi-modal parameter identification and chatter stability curve drawing function. In this study, uses the zero-order analysis method by professor Altintas and Lowest envelop method (LEM) to obtain the chatter stability curve diagram. Which can calculate by separately considering different dominant modes from the experiment of dynamic stiffness. The second part focuses on the identification of modal parameters, the experiment uses the hammer impact test to obtain the tool frequency response characteristics required for the analysis, and uses the global rational polynomial (GRFP) to perform the curve fitting of the response function to identify the damping ratio. Finally, the predicted chatter boundaries are compared to the experimental results in order to validate the modal and the stability analysis. The experiments show that under the condition of 100% radial feed, three of the eleven test points are not consistent with the predicted results, and the initial prediction accuracy is about 72%. In the phase difference measurement section, when the chatter occurs, the major frequency of the phase difference is greater than 60 degrees and the fluctuation range is also larger；In the case of stability, most of its major frequency located on the frequency of the tool passing frequency and its multiplier and the phase difference of the abnormal frequency is not more than 30 degrees.

並列關鍵字

Machine chatter ； Stability lobes diagram ； Regenerative chatter ； Zero-Oder Analysis method ； Envelop method

參考文獻

[37] 王柏村,梁秀瑋,周嘉莉,”單自由度與多自由度系統模型於刀具顫震穩定圖分析”,中華民國振動與噪音工程學會第18屆學術研討會, 2010年6月12日.

[3] R.P.H Faassen, “Chatter prediction and control for high-speed milling: modelling and experiments”, Technische Universiteit Eindhoven, 2007.

[8] Jianping Yue,” Creating a Stability Lobe Diagram”, Division of Engineering Technologies and Computer Sciences, Essex County College, Session IT 301-050.

[9] R.P.H Faassen, “Chatter prediction and control for high-speed milling: modelling and experiments”, Technische Universiteit Eindhoven, 2007, page 15.

[10] S. A. Tobias and W. Fishwick, “A theory of Regenerative chatter,” The Engineer-London, 1958.