透過您的圖書館登入
IP:3.140.195.167
  • 期刊

非破壞檢測可靠度模型之適合度探討

Study of Model Adaptability of NDE Reliability

摘要


在航空、核能、鋼鐵、石化等產業中,為了維護結構的安全性或產品的品質,人們常使用非破壞檢測來尋找結構中可能存在的裂縫。從檢測結構後得到的訊號,我們可以判斷結構是否存有裂縫,但在許多因素影響之下,檢測裂縫所得到的訊號難免存在不確定性,對此,我們可以線性迴歸模型加以闡述,假設裂縫長度之對數值與所得訊號之對數值成正比、而各大小不同裂縫量測所得訊號之對數值則是變異數相同的常態分佈,藉以建立所謂的檢測機率曲線,來探討非破壞檢測儀器對不同長度裂縫的檢出機率。然而實際上由於樣本數有限,量測不同裂縫所得雜訊對數值分佈之變異程度未必相同、也未必是常態分佈,因此我們需以統計學上的假設檢定來驗證。本文中即舉兩個例子作為檢驗的對象,檢驗結果發現兩例中雜訊的對數值皆可視為常態分佈,但量測不同大小裂縫所得雜訊對數值之變異程度則不盡然相同。

並列摘要


Non-Destructive Testing (NDT) and Non-Destructive Evaluation (NDE) play important roles in the safety assessment of structures. Based on the detected signal of an NDT instrument, one can judge whether cracks exist in a structure. However, the NDT signal is influenced by many factors including instrumental noises. It results in uncertainties when doing the evaluation. To resolve the uncertainty, the NDT data can be analyzed by a linear regression model. Through the model, the probability of detection (POD) of the instrument or its user can be evaluated based on the detected signal. In the present paper, a detailed statistical analysis of the NDT signals is introduced and performed. Statistical hypothesis testing is carried out as well. Two sets of NDT data are illustrated as numerical examples. It is found that NDT instrumental noises do follow the assumed lognormal distributions in both examples. However, depending on the crack sizes measured, their degrees of variation may not be the same.

被引用紀錄


邱民杰(2017)。考慮氣壓及水溫變化的潮汐調和分析〔碩士論文,國立交通大學〕。華藝線上圖書館。https://www.airitilibrary.com/Article/Detail?DocID=U0030-2212201712283423

延伸閱讀