請用此 Handle URI 來引用此文件:
http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/69154
標題: | 敲擊回音視波速現象之探討 A study of apparent velocity of the impact-echo test |
作者: | Chen-Yeh Yen 顏辰燁 |
指導教授: | 劉佩玲(Pei-Ling Liu) |
關鍵字: | 敲擊回音法,視波速因子,振動模態,柏松比,訊號取樣總長,寬高比, Impact-echo method,Empirical correction factor,Vibration modal,Poisson’s ratio,Sampling time,Aspect ratio, |
出版年 : | 2018 |
學位: | 碩士 |
摘要: | 敲擊回音法為應用於混凝土結構檢測中最為廣泛的非破壞檢測技術,其原理為於待測物表面敲擊激發應力波,以位移感測器量取敲擊點附近之縱向位移訊號,經由傅立葉轉換將位移時間域訊號轉換為頻率域訊號,即可由頻譜尖峰頻率判斷試體內部特徵。然而,由敲擊回音法所量測試體厚度與理論值卻存有一定比例誤差,稱之為視波速現象,使檢測人員產生量測上的不準確因素。
本研究目的旨在探討敲擊回音法理論頻率與實際量測所得頻率尖峰之差異,即視波速因子beta。分析對象分別為圓形,方形與無限大平板三種試體。利用Rayleigh-Lamb波動理論分析無限大扁平板之振動模態;以Aggarwal所提圓盤振動理論分析圓形試體振動模態;以及藉由有限元素軟體分析圓形及方形試體之振動模態。 本文考慮三種參數對敲擊回音尖峰頻率之影響:訊號取樣總長、材料常數與寬高比。首先,分析試體材料常數之影響,將Rayleigh-Lamb波動方程整理並化簡,將無限大平板之beta表示為柏松比函數,即此beta值只與試體柏松比相關,並以有限元素軟體模擬驗證理論值。 接著,分析訊號擷取時長影響,由數值模擬結果指出,當訊號擷取時長低於一門檻時,不論圓形或方形試體,因應力波於試體側向傳遞次數甚低,反應類似於一無限大平板中傳遞情形,因而所得結果與以Rayleigh-Lamb波動理論所得到beta值接近。當訊號擷取時長大於上述門檻後,此時應力波於試體側向傳遞次數增高,以致激發試體各式模態並反應於頻譜上,此時頻率尖峰對應試體模態振動。 前述門檻主要受試體幾何形狀與寬高比影響。本研究嘗試圓形及方形試體2 ~ 10之寬高比變化,數值模擬結果指出,寬高比4以上之圓形試體,模擬時長為底部回波週期40倍時約為激發試體模態振動之時間門檻。低於此寬高比之試體於相同訊號取樣長度時則易激發試體模態振動,進而影響所量測頻率,造成估算厚度之誤差。方形試體對於訊號取樣總長改變則無明顯反應,同一寬高比試體,由短時至長時進行量測,所取得頻率尖峰對應beta值皆十分相近。 寬高比亦對試體振動模態影響甚劇。於長時量測時,利用圓盤振動理論計算出各寬高比圓試體所對應模態頻率理論解與繪製其振態,並觀察到各寬高比圓試體頂部位移第一圈跨零點半徑均落於0.60 ~ 0.66倍試體厚度範圍內之現象。 綜上所述,建議進行敲擊回音試驗時可將量測訊號交由電腦分析其短時至長時反應,觀察頻率尖峰改變情形,並挑選短時分析結果,使用Rayleigh-Lamb理論所提供beta值修正試體厚度。 Impact-echo method is the most widely used nondestructive testing technique in the testing of concrete structure. Hitting the surface of the target and creating a stress wave, we can measure the vertical displacement signal near the impact point in time domain. We can determine the characteristics inside the target by studying the peaks’ frequency on the spectrum, transformed from time domain by Fourier transform. However, it exists a distinctive error between the results of the target’s thickness measured with impact-echo method and the theoretical value, which is called apparent velocity, suggesting an uncertain factor in the measuring process. The main purpose of this research is to investigate the difference between the peak of the theoretical frequency of impact-echo method and the frequency obtained by measurement, which is called apparent velocity factor beta, by analyzing circular, rectangular and infinitely large plates. Using Rayleigh-Lamb wave theory to analyze the vibration modal of infinitely large plates, Aggarwal’s theory of vibrating disks to analyze the vibration modal of circular plates and a finite element software to analyze the vibration modal of circular and rectangular plates. This research took into consideration the influence of three parameters toward impact-echo peak frequency: sampling time, material properties and aspect ratio. First analyzing the influence of the material properties, by simplifying Rayleigh-Lamb wave equation, expressing the beta of the infinitely large plate with Poisson’s ratio, making this beta only be related to Poisson’s ratio, and verifying this theoretical value with a finite element software. Then, by analyzing the influence of sampling time, we can see from the result of numerical simulations that when the sampling time is shorter than a threshold, since the number of stress wave’s lateral transfer is low, it reacted like if it was transferring on an infinitely large plate, whether the specimen is circular or rectangular, thus the result is similar to the beta value obtained by Rayleigh-Lamb wave theory. When the sampling time is above the threshold, the number of lateral transferring of stress wave in the specimen is increasing, leading to the exciting of all kinds of modals of the specimens and reflecting them on the frequency spectrum, and in this circumstance, the frequency peak corresponds to the specimen’s vibration modal . The above mentioned threshold is mainly decided by the shape and the aspect ratio of the specimen. This research tries to vary the aspect ratio of circular and rectangular specimens between 2~10, according to the results of numerical simulation, circular specimens with an aspect ratio above 4 have the threshold time of exciting the specimens modal vibration when the termination time is approximately 40 times the length of cycle of the specimens’ bottom reflected wave. Specimens with an aspect ratio below this threshold with the same sampling time tend to excite its vibration modal, influencing the measured frequency, leading to error in the estimation of the thickness. On the other hand, rectangular specimens do not show obvious difference faced with different sampling time. Measuring specimens with the same aspect ratio from short to long sampling time, we obtain frequency peaks corresponding very well to beta. Aspect ratio also greatly influences the vibration modal of specimens. When measuring over a long period, using circular plate vibration theory, we can calculate the theoretical value of modal frequency corresponding to various aspect ratio of circular plates and plot its vibration mode. We can observe that the zero-crossing point’s radius of the displacement’s first loop of the circular specimens’ top all lies within 0.60~0.66 times the thickness of the specimens. To sum up, it is suggested to use the computer to analyze the measured signals’ reactions from a short period of time to a long period of time when applying impact echo tests and observe the difference of frequency peaks and pick the results with short analyzing time to correct the thickness of the specimens with Rayleigh-Lamb theory. |
URI: | http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/69154 |
DOI: | 10.6342/NTU201801531 |
全文授權: | 有償授權 |
顯示於系所單位: | 應用力學研究所 |
文件中的檔案:
檔案 | 大小 | 格式 | |
---|---|---|---|
ntu-107-1.pdf 目前未授權公開取用 | 3.94 MB | Adobe PDF |
系統中的文件,除了特別指名其著作權條款之外,均受到著作權保護,並且保留所有的權利。