透過您的圖書館登入
IP:3.142.144.40
  • 會議論文

局部穩態隨機振動量測資料分析實例探討

摘要


實務上遭遇的大部分隨機資料,若以訊號之整體特性來看大多屬非穩態。欲處理此類非穩態資料,須蒐集相似條件下之重複樣本,以便計算資料之整體統計特性。但是有時要獲得重複樣本有實質上的困難,甚至不可能,因而必須針對單一樣本進行分析工作。在這種情況下,僅一些特殊型態之非穩態資料可獲致適切的分析結果。本文係針對飛機外掛物,於掛載飛行過程中實際量測所得之振動資料進行分析。文中首先以分段方式進行頻譜分析,然後利用統計方法(Cochran's test)檢定分段頻譜,以確認其是否爲等變異數(符合局部穩態模式),結果證實此分析方式可適用於此類飛機外掛物上量測所得之振動資料分析。

關鍵字

無資料

並列摘要


Much of the random data of interest in practice is nonstationary when viewed as a whole. To analyze such nonstationary data, it is necessary to gathering repeated samples under statistical similar conditions to calculate ensemble statistic. In a number of situations where to collect data described above is not feasible, and individual sample records must be analyzed. In such cases, only the data have a specific type of nonstationarity can be analyzed well. This paper is concerned with analysis of practical measurement data on external stores of fighter during captive flight. First, the APSD is computed from individual 1-sec segments of the 5-sec-long sample record. Next, a Cochran's test is used to check whether the APSD is variance homogeneity. The results validate that the application of locally stationary model to analyze this type of vibration data is feasible.

並列關鍵字

locally stationary random vibration

延伸閱讀