透過您的圖書館登入
IP:3.14.253.221
  • 學位論文

渦電流致動轉軸動平衡系統之精度與速度改善研究

A Study Of The Accuracy And Speed Characteristics Of An Eddy Current Actuated Balancing System

指導教授 : 陳政雄 黃以文
若您是本文的作者,可授權文章由華藝線上圖書館中協助推廣。

摘要


工具機主軸轉動時因本身存在著不平衡量,進而造成振動,目前最有效率的解決方案為在轉軸上加裝一套線上動平衡系統。過去實驗室已完成建立渦電流動平衡系統,而線上動平衡技術所需要的重要資訊為加速規量取之原始不平衡量之相位資訊,但過去在實機測試誤差約三度,故本論文將針對相位穩定性以及精度兩項作探討改善。一但能將加速規量測之振動訊號穩定,可研判往後之運算過程錯誤能夠減少進而增加平衡之速度進而改善致動系統運作時間。 為了使加速規量測之振動訊號資訊穩定且精確,在電路上特別加上防止訊號漂移電路,此電路包含了一組積分器及兩組高通濾波器。由積分電路將訊號取平均值。藉由moving average的方法有效的將訊號穩定度提高。由實驗驗證moving average有效地將加速規所量測之振動訊號穩定,經由防止訊號漂移之電路後將穩定之振動訊號透過施密特觸發電路將振動訊號之弦波轉換成方波訊號。施密特觸發可經由設計之上觸發點以及下觸發點,此特性可改善零準位觸發電路於零準位附近雜訊之干擾雜訊造成之誤差狀況。 本論文首先將針對惡劣環境下,加速規之訊號在受到干擾之情況下能夠穩定,接著對相位偵測系統做重複性之測試。過去濾波電路僅靠一組追隨調適(self-tracking)帶通濾波(band-pass filter)技術來過濾雜訊,但為求加速規所量測之訊號能夠穩定使相位偵測系統之誤差減小,本論文在帶通濾波器前加入高通濾波器(high-pass filter)以及低通濾波器(low-pass filter),分別將100Hz以上以及10Hz以下之干擾去除。接著將加速規以及電路部分利用防護隔離(shielding)設計來隔離工廠中因電力線與空氣電磁波所造成的高頻電磁干擾;經過完整的濾波電路以及防護隔離後,將加速規之弦波訊號送入防漂移電路以及施密特觸發電路,避免零準位附近之雜訊干擾系統之相位計算。接著利用微分器將其方波訊號轉換成脈衝訊號,接著送入CPLD以及Matlab做相位偵測計算原始不平衡量之相位。 本論文首先針對加速規原始訊號之頻率及波形做改善,接著進行主軸原始不平衡量之相位偵測。目前加速規訊號頻率經由Matlab計算出之頻率已和主軸轉速相同,且在角度計算上誤差值約1度較過去系統(約3度)穩定。

並列摘要


The required stability and processing quality machine industry where the machine vibration became a major project to overcome. Machine tool having a rotary shaft of the source of vibration problem whether it's the existence of residual unbalance, non-symmetrical shape of work piece or centrifugal force of rotation from spindle will has to produced vibration, in this situation not only affects the processing efficiency not only affects the processing efficiency, work quality, accuracy of the product and also make the operator's into a dangerous working place. Currently the most efficient solution was installation the On Line Balancing System. This system was available to supervised the information of spindle rotates, location and size vibration, while the phase unbalance the efficiency and accuracy is an important factor in balancing system. Laboratory has successfully developed outlet on balancing system, but in phase detection there still have much better for improvement. Therefore this thesis will focus on harsh environments, the vibration signal in accelerometer detect from the environments and machine vibration noise will have much of interference situation and we hope to solved and stabilize the unnecessary interference noise. When accelerometer vibration signal was stabilized, calculation will reduced and will improvement the system operation time. This paper proposed a circuit to prevent signal drift; this contains a set of integrator and two high-pass filter. The moving average can improve the stability signal and phase information. From our experiment verify moving average effective to stabilized vibration signal of accelerometer. In this experiment, first we try to make the experiment working in a harsh environment and interference an accelerometer vibration signal can stabilized. In the past filter circuit only had self-tracking band-pass filter techniques to filter noise, but in order to measured accelerometer signals to stabilize the phase detection system error is reduced, therefore the at the band-pass filter before adding the high-pass filter and a low-pass filter, respectively filter 10Hz-100Hz unnecessary low and high frequency generated by the floor and the machine vibration. since to prevent electromagnetic noise we designed a protective isolation circuit part (shielding) to isolate plan , due to power line and the air waves caused by high-frequency electromagnetic interference accelerometer ; through the complete filter circuit and protective isolation, it will accelerate the regulation of the sinusoidal signal into Schmitt special trigger circuit, avoiding the zero level in the vicinity of the phase calculation system noise interference. Then using the differential square-wave signal will converted into a pulse signal, and then fed back into the phase detector CPLD and Matlab to calculations of original unbalance phase. In this thesis, a 12-inch lathe spindle on experiments, first for the accelerometer frequency of the original signal and the waveform to make an improvement and followed by the original unbalance of the spindle phase detected. Currently accelerometer signal frequency of the frequency calculated by Matlab and spindle speed has been the same, and the error in the calculation of the value of the angle about 1 degree over the last system (about 3 degrees) stability.

參考文獻


【2】Fan Hongwei, Jing Minqing, Liu Heng“ Program Design of An Online Dynamic Balancing System for Grinding-wheel and Spindle ” , 2011 IEEE ,Vol 1,P173 - 177 ,Conference Publications
【3】Balancing Rotating Machines ,Journal I.E.E., June 1955
【4】M. Hecquet, P. Brochet, Lee Sang Jin and P. Delsalle,“A Linear Eddy Current Braking System Defined by Finite Element Method”,IEEE Trans. on Magnetics, Vol.35, No.3, May, 1999, pp.1841-1844.
【5】P. J. Wang and S. J. Chiueh, “Analysis of Eddy-Current Brakes for High Speed Railway, ” IEEE Transactions on magnetics, Vol.34, No.4, July, 1998, pp. 1237-1239.
【6】Guoxin Li ,Zongli Lin ,Costin Untaroiu, Paul E. Allaire, “Balancing of High-speed Rotating Machinery Using Convex Optimization” , P4351 - 4356 vol.4 Conference Publications

延伸閱讀