本文選題：損傷檢測 + 希爾伯特黃變換　； 參考：《中國海洋大學》2014年碩士論文

【摘要】：對結(jié)構(gòu)完整性進行監(jiān)測已經(jīng)成為許多工程研究領(lǐng)域的焦點。結(jié)構(gòu)的健康監(jiān)測離不開傳感器技術(shù)以及相應(yīng)的信號處理技術(shù)，信號處理技術(shù)是傳感器信號和結(jié)構(gòu)完整性檢測之間的紐帶。如何快速有效地從傳感器采集到的結(jié)構(gòu)振動響應(yīng)信號中提取出重要的特征信息，并判斷結(jié)構(gòu)的健康狀況，具有很大的應(yīng)用價值。基于希爾伯特黃變換的損傷檢測方法是一種全局性的檢測方法，與傳統(tǒng)的損傷檢測方法相比，其具有自適應(yīng)性、高時頻分辨率等優(yōu)點，具有非常廣闊的應(yīng)用前景。 本文系統(tǒng)的總結(jié)了目前主要的時頻分析方法，對其研究現(xiàn)狀、研究成果及存在的問題做了較全面的論述。闡述了希爾伯特黃變換的基本原理、發(fā)展歷程及其應(yīng)用于時頻分析的優(yōu)越性。結(jié)合之前有關(guān)希爾伯特黃變換的研究成果，對希爾伯特黃變換應(yīng)用于損傷檢測領(lǐng)域進行了進一步的研究。 考慮到經(jīng)驗?zāi)Ｊ椒纸獾木热菀资茉肼暩蓴_的缺點，利用小波變換的消噪功能結(jié)合希爾伯特黃變換進行損傷檢測，并依據(jù)HHT方法識別出的低階瞬時頻率變化定義了一個損傷指標。對一個三層剪切模型進行了數(shù)值模擬研究，并利用此指標研究了模型在不同損傷程度、不同噪聲背景下的損傷檢測情況，證明了該方法的有效性，為結(jié)構(gòu)狀況的在線監(jiān)測提供了一定的參考依據(jù)。 結(jié)構(gòu)的動力特性參數(shù)會受到環(huán)境因素和操作條件的影響，，而且環(huán)境因素和操作條件的改變所引起的結(jié)構(gòu)動力特性參數(shù)的變化，可能會掩蓋結(jié)構(gòu)由于損傷所引起的比較微小的變化。針對這一現(xiàn)狀，本文提出了希爾伯特黃變換結(jié)合主成分分析的損傷檢測方法。采用一個懸臂梁模型進行了數(shù)值模擬，用主成分分析排除溫度對由希爾伯特黃變換方法得到的模態(tài)頻率的影響，然后應(yīng)用異常檢測技術(shù)對主成分分析得到的預(yù)測誤差矩陣進行處理判斷損傷是否發(fā)生。數(shù)值研究結(jié)果表明這種方法可以有效排除溫度對模態(tài)頻率的影響，并可以準確地檢測到損傷的發(fā)生。 導管架平臺是淺海中比較常用的平臺形式，因此研究這類平臺的損傷檢測技術(shù)具有很重要的現(xiàn)實意義。本文分別建立了單立柱式和導管架式井口平臺的有限元模型，利用HHT方法分析研究了模型在不同損傷程度、不同外界激勵下的損傷檢測，并應(yīng)用添加噪聲的方式模擬外界環(huán)境的影響，得到了一些有用的結(jié)論。
[Abstract]:Structural integrity monitoring has become the focus of many engineering research fields. Structural health monitoring can not be separated from the sensor technology and the corresponding signal processing technology, signal processing technology is the link between sensor signal and structural integrity detection. How to extract the important characteristic information from the structural vibration response signal collected by the sensor quickly and effectively, and to judge the health condition of the structure, has great application value. The damage detection method based on Hilbert-Huang transform is a kind of global detection method. Compared with the traditional damage detection method, it has the advantages of self-adaptability, high time-frequency resolution and so on, so it has a very broad application prospect. In this paper, the main methods of time-frequency analysis are systematically summarized, and the research status, research results and existing problems are discussed comprehensively. The basic principle and development of Hilbert-Huang transform and its advantages in time-frequency analysis are described. The application of Hilbert-Huang transform to damage detection is further studied. Considering the disadvantage that the precision of empirical mode decomposition is easily disturbed by noise, the wavelet transform is used to remove noise and Hilbert-Huang transform is used to detect the damage. A damage index is defined according to the low order instantaneous frequency change identified by HHT method. The numerical simulation of a three-layer shear model is carried out, and the effectiveness of the method is proved by using this index to study the damage detection of the model in different damage degree and different noise background. It provides a certain reference for the on-line monitoring of the structure condition. The dynamic characteristic parameters of the structure will be affected by the environmental factors and the operating conditions, and the change of the structural dynamic characteristic parameters caused by the change of the environmental factors and the operating conditions. It may cover up minor structural changes caused by damage. In this paper, a damage detection method based on Hilbert-Huang transform and principal component analysis (PCA) is proposed. A cantilever beam model is used for numerical simulation. The influence of temperature on the modal frequency obtained by Hilbert-Huang transform is excluded by principal component analysis. Then the prediction error matrix obtained by principal component analysis is processed by anomaly detection technique to judge whether damage occurs or not. The numerical results show that this method can effectively eliminate the influence of temperature on modal frequency and can accurately detect the damage. Jacket platform is a commonly used platform in shallow water, so it is of great practical significance to study the damage detection technology of this kind of platform. In this paper, the finite element models of single column and conduit type wellhead platform are established, and the damage detection of the model under different damage degree and different external excitation is analyzed by using HHT method. Some useful conclusions are obtained by using the method of adding noise to simulate the influence of external environment.
【學位授予單位】：中國海洋大學
【學位級別】：碩士
【學位授予年份】：2014
【分類號】：P751

【參考文獻】

相關(guān)期刊論文 前8條

1 張強;嵇春艷;范佰明;;基于小波變換的固定式海洋平臺結(jié)構(gòu)損傷識別方法研究[J];海洋技術(shù);2008年02期

2 王龍;沈艷霞;季凌燕;;基于小波降噪和EMD方法的風力發(fā)電系統(tǒng)齒輪箱故障診斷[J];江南大學學報(自然科學版);2012年02期

3 續(xù)秀忠,李中付,華宏星,陳兆能;非平穩(wěn)環(huán)境激勵下線性結(jié)構(gòu)在線模態(tài)參數(shù)辨識[J];上海交通大學學報;2003年01期

4 張兆德 ,王德禹;基于動力特性與小波變換的損傷檢測方法在海洋平臺中的應(yīng)用[J];上海造船;2005年01期

5 宗周紅,任偉新,阮毅;土木工程結(jié)構(gòu)損傷診斷研究進展[J];土木工程學報;2003年05期

6 陳雋,徐幼麟;HHT方法在結(jié)構(gòu)模態(tài)參數(shù)識別中的應(yīng)用[J];振動工程學報;2003年03期

7 羅忠輝,薛曉寧,王筱珍,吳百海,何真;小波變換及經(jīng)驗?zāi)Ｊ椒纸夥椒ㄔ陔姍C軸承早期故障診斷中的應(yīng)用[J];中國電機工程學報;2005年14期

8 徐仁林;安偉;;小波降噪在信號基于EMD的Hilbert變換中的應(yīng)用[J];噪聲與振動控制;2008年03期

本文編號：1891230