以DSP為基礎獨立式監測系統硬體製作

Abstract

本研究在探討轉動機械上之線上即時故障診斷技術。整個故障診斷系統結構分為兩個部份。第一、在特徵值方面，採用了以模型為基礎的參數判別法來產生特徵值。第二、在判斷方面，使用閾值檢查（limit checking）的方式來判斷。實驗中所需要量測的訊號為轉子的X及Y軸的側向振動訊號和轉子的轉速訊號。整個FDI（faul tdetection and isolation）系統是透過數位訊號處理器 （DSP）來實現。此外，我們建立了一套轉子測試平臺，並驗證了這理論的可行性。最後，我們亦成功地發展出一套應用於轉動機械的獨立式故障診斷系統。

This research concerns the on-line fault detection and isolation(FDI) system applied to the rotating machinery. The architecture of the system is mainly comprised of two parts. The first part is the feature generation. We generate the feature by using the parameter identification. The second part is the fault inference. The limit checking is exploited for inference. The signals of the rotating velocity and lateral vibration on X -axis and Y -axis are measured in this approach. The FDI system is carried out by the digital signal processor. Besides, the rotor kit is built for examining if the theory is correct. Finally, we successfully developed a stand-alone FDI system applied to the rotating machinery.