On-line Ball Screw Preload Degradation Monitoring Based on Vibration Signals

基於振動訊號的線上滾珠螺桿預壓力衰減監測

蔡秉均(Ping-Chun Tsai) ； 鄭志鈞(Chih-Chun Cheng) ； 黃逸群(Yih-Chyun Hwang)

《中國機械工程學刊》 38卷5期 (2017/10) Pp. 489-497

摘要

A novel technique for on-line monitoring ball screw preload degradation for ball screw feed drive systems is proposed. Simply attaching an accelerometer on the ball screw nut of a ball screw feed drive system in operation, the ball screw preload degradation is found to be correlated well with the actual ball pass order (ABPO), the vibration RMS amplitude at ABPO, the variance of vibration at ABPO and the dispersion of order spectra from the extracted vibration signal after processing using Angular Velocity Vold-Kalman Filtering Order Tracking (AV VKF-OT). Then a methodology in monitoring automatically the preload degradation for a ball screw feed drive system based on principal component analysis with these four parameters as the process variables is proposed and demonstrated experimentally.

關鍵字

Ball screw preload ； Vold-Kalman filtering ； Principal components analysis

並列摘要

本研究針對工具機之進給系統提出一創新的線上監測方法，用以監測滾珠螺桿預壓力之衰減情形。透過安裝於滾珠螺桿之螺帽上的加速規量測螺桿運轉之振動訊號，並以角速度馮卡曼階次追蹤法(Angular Velocity Vold-Kalman Filtering Order Tracking)對其進行訊號處理以獲取滾珠振動訊號，並根據本研究所提出之演算法計算四項特徵參數值如真實球通階次(Actual ball pass order)、球通振動RMS值(Vibration RMS amplitude at ABPO)、球通振動變異數(Variance of vibration at ABPO)及階次譜分散度(Dispersion of order spectra)，此四項特徵參數經實驗證實與滾珠螺桿之預壓力衰減情形息息相關。據此，本研究透過結合該四項特徵參數與主成份分析法(principal component analysis)技術進而提出一線上螺桿預壓力衰減情形自動監測方法，經實驗驗證此法可有效且準確地辨識滾珠螺桿之預壓力是否消失。