本文選題：探地雷達(dá) + 缺陷巖體��； 參考：《中南林業(yè)科技大學(xué)》2014年碩士論文

【摘要】：探地雷達(dá)是一種具有高效、快速、便捷、連續(xù)的無損檢測物理探測方法,是一種目前先進(jìn)的高頻電磁波勘探技術(shù)。由于它對淺層目標(biāo)體具有良好的探測效果,因而廣泛地應(yīng)用于工程結(jié)構(gòu)檢測、工程地質(zhì)及水文地質(zhì)勘查、隧道工程超前預(yù)報(bào)與檢測、考古調(diào)查以及生態(tài)環(huán)境檢測等方面。由于電磁波傳播的高頻特性,傳播環(huán)境條件干擾大,以及探測環(huán)境復(fù)雜性等,使得探地雷達(dá)在進(jìn)行缺陷及缺陷注漿的探測中存在著不足,例如：數(shù)據(jù)分辨率不高、雷達(dá)圖像的識(shí)別主要依靠人工識(shí)別等。因此,開展巖體缺陷探測及缺陷注漿探測的物理正演模擬工作,研究電磁波在缺陷巖體中的傳播規(guī)律,有利于對雷達(dá)采集數(shù)據(jù)的分析、雷達(dá)圖像識(shí)別以及對地下巖體結(jié)構(gòu)形態(tài)判別的經(jīng)驗(yàn)積累；通過快速有效地的物理模型試驗(yàn)研究,可進(jìn)一步提高探地雷達(dá)回波信號(hào)識(shí)別的準(zhǔn)確性,對解釋和識(shí)別地下介質(zhì)或目標(biāo)體具有重要意義。 本論文通過物理模型試驗(yàn)、數(shù)值正演模擬,采用希爾伯特變換和小波變換消噪技術(shù)等方面對巖體缺陷及注漿效果探測技術(shù)進(jìn)行了較為系統(tǒng)深入的研究。利用LTD-2100探地雷達(dá)系統(tǒng)進(jìn)行實(shí)測以及GprMax2D軟件進(jìn)行仿真實(shí)驗(yàn),利用雷達(dá)后處理軟件IDSP6.0和數(shù)值分析軟件MALAB分別選取希爾伯特變換和小波變換消噪的方法進(jìn)行了試驗(yàn)數(shù)據(jù)的處理與分析,主要研究內(nèi)容和成果如下： 一、研究內(nèi)容 (1)以MATLAB數(shù)值分析平臺(tái),編制程序?qū)崿F(xiàn)了LTD-2100探地雷達(dá)系統(tǒng)數(shù)據(jù)導(dǎo)入,即原始雷達(dá)數(shù)據(jù)的文件存儲(chǔ)中信息以及數(shù)據(jù)文件*.LTE格式文件的讀取,從而可以方便地使得探地雷達(dá)數(shù)據(jù)進(jìn)行二次開發(fā)、分析、處理和成像。 (2)根據(jù)電磁波的傳播理論,巖體介質(zhì)中傳播的高頻電磁波服從麥克斯韋方程組,分析了二維時(shí)域有限差分方程,以GprMax2D軟件為數(shù)值模擬平臺(tái),選擇與實(shí)際探測天線接近的雷達(dá)激勵(lì)源——雷克子波,建立了水、混凝土、濕砂等不同充填介質(zhì)在巖體缺陷結(jié)構(gòu)中的探地雷達(dá)電磁波傳播模型。模擬了不同尺度、不同形狀、不同填充狀態(tài)下二維的探地雷達(dá)波場,在此基礎(chǔ)上研究了不同巖體缺陷及注漿效果對雷達(dá)反射(回波)信號(hào)的影響。 (3)進(jìn)行了不同巖體缺陷結(jié)構(gòu)(形狀、規(guī)模、充填物)傳播的物理模型試驗(yàn),在利用雷達(dá)后處理軟IDSP6.0進(jìn)行常規(guī)的數(shù)據(jù)處理,包括增益調(diào)節(jié)、水平濾波和垂直濾波等的基礎(chǔ)上,對雷達(dá)數(shù)據(jù)進(jìn)行了希爾伯特變換分析,提取了瞬時(shí)振幅、瞬時(shí)相位和瞬時(shí)頻率的“三瞬”屬性特征,利用屬性參數(shù)進(jìn)行了雷達(dá)數(shù)據(jù)分析和解釋。 (4)基于探地雷達(dá)數(shù)據(jù)采集信號(hào)的非平穩(wěn)性,以及高頻脈沖信號(hào)的非線性衰減等特點(diǎn),對探地雷達(dá)回波信號(hào)的小波消噪處理技術(shù)進(jìn)行了探索。利用MATLAB軟件,研究分析了二維小波變換在探地雷達(dá)回波信號(hào)處理中的分解、重構(gòu)和消噪的算法,選取混凝土注漿探測數(shù)據(jù)進(jìn)行小波消噪,分析結(jié)果與希爾伯特變換對比,研究兩方法的實(shí)用性。 二、主要成果 (1)同一探測環(huán)境下,以水、混凝土、干砂、濕砂、淤泥和粘土為充填介質(zhì),進(jìn)行缺陷巖體填充程度注漿模擬。研究表明：滿充填程度時(shí),水、濕砂、淤泥和粘土介質(zhì)充填時(shí)隱約見到模糊的頂角波形,濕砂和淤泥的深層信號(hào)形成較強(qiáng)的二次繞射波,混凝土和干砂較難分辨；其原因是充填介質(zhì)與探測環(huán)境的相對介電常數(shù)相差越小,波形圖越難識(shí)別。半充填條件下,波形剖面都能很容易的辨別缺陷頂角的小波形,對比深部波形信息,發(fā)現(xiàn)水、濕砂、淤泥和粘土充填比混凝土、干砂子和空氣充填含有更強(qiáng)多次的繞射波。 (2)不同探測環(huán)境下的數(shù)值模擬,研究發(fā)現(xiàn)：滿填條件下,空氣、水、混凝土、濕砂、淤泥和粘土含有的繞射波信息由多到少的探測環(huán)境是干燥環(huán)境、分層環(huán)境和潮濕環(huán)境,干砂則相反；半充填條件下,分辨程度由易到難的探測環(huán)境是潮濕、分層和干燥,各種環(huán)境條件下水、濕砂、淤泥和粘土較易分辨。 (3)利用不同形狀、充填程度的混凝土數(shù)值模擬,進(jìn)行了理論探測和現(xiàn)場試驗(yàn)探測對比模型研究,分析表明現(xiàn)場試驗(yàn)?zāi)M理論滿注漿和半注漿的半圓形、三角形存在缺陷時(shí),雷達(dá)波形有較明顯差異。 (4)試驗(yàn)數(shù)據(jù)處理分析表明：缺陷巖體以水、濕泥、濕砂、干砂子填注�？梢苑治銎鋵傩詤�(shù)：瞬時(shí)振幅圖形,可利用強(qiáng)白區(qū)和分層突變線進(jìn)行判別半注漿和滿注漿。若有強(qiáng)白區(qū)或分層突變線,表示為缺陷巖體半注漿或未注滿；若沒有分層突變線存在,表明注漿探測結(jié)果為滿注。 (5)研究表明：半注漿或未注滿時(shí)能形成分層突變線但沒有強(qiáng)白區(qū),這是由于充填混凝土未振搗壓實(shí),注漿表層面沒有形成薄水膜造成；滿注漿時(shí)明顯發(fā)現(xiàn)有矩形白區(qū)域,由于充填混凝土振搗壓實(shí)、滿注流漿,注漿表層形成了薄水膜造成；建議：注漿完成后的檢測應(yīng)盡早進(jìn)行。根據(jù)這一特性結(jié)合探測現(xiàn)場環(huán)境和地質(zhì)資料可以進(jìn)行施工現(xiàn)場注漿效果探測和評價(jià)。這為運(yùn)用探地雷達(dá)進(jìn)行注漿效果現(xiàn)場監(jiān)測提供了技術(shù)。 (6)研究表明瞬時(shí)振幅強(qiáng)弱與衰減的性質(zhì)較容易識(shí)別介質(zhì)層位；瞬時(shí)相位剖面上的同相軸連續(xù)性較易判別缺陷巖體頂面位置,利用其深層信號(hào)特征較易判別巖體注漿程度；利用瞬時(shí)頻率特征分析介質(zhì)變化性質(zhì)難以判別缺陷形狀,但結(jié)合瞬時(shí)振幅特征可以共同判定缺陷巖體注漿效果。綜合“三瞬”屬性特性分析,可知：雷達(dá)探測注漿效果的敏感性因子是水,包含充填介質(zhì)的含水率,含水量高則探測結(jié)果敏感。 (7)利用小波二維窗口,選取dbl小波函數(shù),以Ievel2為層數(shù),分解并進(jìn)行消噪處理,選用軟閾值和尺度白噪聲,分別從水平方向、斜線方向和垂直方向進(jìn)行量化系數(shù)調(diào)試,消噪、重構(gòu),可以從Residuals中獲取了清晰可分辨的巖體缺陷形狀及缺陷注漿效果圖。 (8)小波消噪和希爾伯特變換對比分析研究表明：利用二維小波變換的雷達(dá)回波信號(hào)消噪方面有較好的效果,較易識(shí)別缺陷巖體形狀及缺陷頂面、底面位置,比希爾伯特變換更為精細(xì)、直觀地反映缺陷巖體及其缺陷混凝土注漿效果,同時(shí)將倒三角形和半圓形容器注漿時(shí)形成的兩個(gè)小缺陷孔洞波形反映出來,但在有缺陷孔洞探測條件下兩形狀容器未滿填形狀不易辨別。
[Abstract]:Ground penetrating radar is a kind of advanced high frequency electromagnetic wave exploration technology , it is a kind of advanced high frequency electromagnetic wave exploration technology . Because it has good detection effect on shallow target body , it is widely used in engineering structure detection , engineering geology and hydrogeologic exploration , advanced prediction and detection of tunnel engineering , archaeological investigation and ecological environment detection .
Through the rapid and effective physical model test research , it can further improve the accuracy of the detection of the radar echo signal , which is of great significance to the interpretation and identification of the underground medium or target body .

In this paper , through physical model test , numerical forward modeling , using Hilbert transform and wavelet transform de - noising technique , this paper makes a systematic and deep research on the technology of detecting the defects and grouting effect of rock mass . By using LTD - 2100 ground penetrating radar system and simulation experiment with GprMax2D software , this paper deals with the processing and analysis of the test data by using the radar post - processing software IDSP6.0 and the numerical analysis software MALAB . The main research contents and results are as follows :

I . Contents of the study

( 1 ) Based on the MATLAB numerical analysis platform , the data import of the LTD - 2100 ground penetrating radar system is realized by the programming procedure , that is , the information in the file storage of the original radar data and the data file * . The reading of the LTE format file can facilitate the secondary development , analysis , processing and imaging of the ground penetrating radar data .

( 2 ) According to the propagation theory of electromagnetic wave , the high - frequency electromagnetic waves propagating in the rock mass are subject to Maxwell ' s equations , the finite difference equation of the two - dimensional time domain is analyzed , and the electromagnetic wave propagation model of ground penetrating radar in the rock mass defect structure is selected by using the GprMax2D software as the numerical simulation platform .

( 3 ) The physical model test of different rock mass defect structure ( shape , scale , filling ) propagation is carried out . On the basis of using radar post - processing soft IDSP6.0 to carry out routine data processing , including gain adjustment , horizontal filtering and vertical filtering , Hilbert transform analysis is carried out on radar data , instantaneous amplitude , instantaneous phase and instantaneous frequency ' s ' three transient ' attribute characteristics are extracted , and the attribute parameters are used for radar data analysis and interpretation .

( 4 ) Based on the non - stationarity of ground penetrating radar data acquisition signal and the nonlinear attenuation of high frequency pulse signal , the wavelet de - noising processing technique of ground penetrating radar echo signal is explored .

II . Main results

( 1 ) In the same detection environment , water , concrete , dry sand , wet sand , sludge and clay are used as filling medium to simulate the filling degree of defective rock mass .
The reason is that the smaller the relative dielectric constant of the filling medium and the detection environment , the more difficult it is to identify the waveform chart . Under the semi - filling condition , the waveform profile can easily distinguish the small waveform of the top angle of the defect and compare the deep waveform information . It is found that the water , the wet sand , the sludge and the clay are filled more than the concrete , the dry sand and the air are filled with more intense diffraction waves than the concrete , the dry sand and the air .

( 2 ) Numerical simulation under different detection environment , it is found that under the condition of full filling , air , water , concrete , wet sand , sludge and clay contain the diffraction wave information from many to less detection environment is dry environment , layered environment and moist environment , dry sand is the opposite ;
Under the condition of semi - filling , the detection environment is moist , layered and dry , and water , wet sand , sludge and clay are more easily distinguished in various environmental conditions .

( 3 ) Based on the numerical simulation of concrete with different shapes and filling degree , a comparative model of theoretical detection and field test is carried out . The results show that the simulation theory of field test is full of semi - circular and triangular shape of grouting and semi - slip casting , and the radar waveform has obvious difference .

( 4 ) The analysis of the data shows that the defect rock mass is filled with water , wet mud , wet sand and dry sand . The attribute parameters of the defect rock mass can be analyzed : the instantaneous amplitude graph , the strong white area and the layered mutant line can be used to judge the semi - grouting and full grouting .
If there is no stratified mutation line , it indicates that the grouting detection result is full .

( 5 ) The research shows that the layered abrupt line can be formed without strong white area when the semi - slip casting or the filling is not full , because the filling concrete is not vibrated and compacted , and the surface of the grouting surface layer does not form a thin water film ;
When full slip casting , the rectangular white area is obviously found , which is caused by the filling concrete vibrating compaction , the full injection of the slurry and the formation of thin water film on the surface layer of the slip casting ;
It is suggested that the detection of grouting effect should be carried out as soon as possible after grouting is completed . According to this characteristic , the site environment and geological data can be used for detecting and evaluating the grouting effect on the construction site . This provides a technique for field monitoring of grouting effect using ground penetrating radar .

( 6 ) The study shows that the nature of the instantaneous amplitude and attenuation is easier to identify the media horizon ;
The continuity of the same - phase axis on the instantaneous phase profile is easy to distinguish the position of the top surface of the defective rock mass , and it is easy to judge the grouting degree of the rock mass by using its deep signal characteristics ;
It is difficult to distinguish the defect shape with the characteristic of instantaneous frequency characteristic , but it can be used to determine the grouting effect of the defective rock mass in combination with the instantaneous amplitude characteristic .

( 7 ) using the wavelet two - dimensional window , selecting dbl wavelet function , taking Ievel2 as the number of layers , decomposing and carrying out denoising treatment , selecting soft threshold and scale white noise , performing quantization coefficient debugging , noise elimination and reconstruction from the horizontal direction , the oblique line direction and the vertical direction respectively , and obtaining clear and distinguished rock mass defect shapes and defect grouting effect graphs from Residences .

( 8 ) The comparison and analysis of wavelet de - noising and Hilbert transform shows that the two - dimensional wavelet transform radar echo signal denoising method has better effect , it is easier to identify the shape of the defective rock body and the top surface and the bottom surface of the defect , and can directly reflect the grouting effect of the defective rock mass and the defect concrete , and simultaneously reflect the two small defect hole waveforms formed when the inverted triangle and the semi - circular container are injected , but the shape of the two - shape container under the detection condition of the defective hole is not easy to distinguish .
【學(xué)位授予單位】：中南林業(yè)科技大學(xué)
【學(xué)位級別】：碩士
【學(xué)位授予年份】：2014
【分類號(hào)】：TU195.1;TU45

【參考文獻(xiàn)】

相關(guān)期刊論文 前10條

1 周輝林;田茂;熊俊志;;探地雷達(dá)回波信號(hào)的特征提取和分類[J];電波科學(xué)學(xué)報(bào);2006年04期

2 楊光亮;朱元清;于海英;;基于HHT的地震信號(hào)自動(dòng)去噪算法[J];大地測量與地球動(dòng)力學(xué);2010年03期

3 梁北援,郭鐵拴,申旭輝;地質(zhì)雷達(dá)雙域數(shù)據(jù)處理軟件及其應(yīng)用[J];地球物理學(xué)進(jìn)展;2005年02期

4 李華;魯光銀;何現(xiàn)啟;鄧珂;;探地雷達(dá)的發(fā)展歷程及其前景探討[J];地球物理學(xué)進(jìn)展;2010年04期

5 趙文軻;陳國順;田鋼;王幫兵;;探地雷達(dá)屬性技術(shù)進(jìn)展[J];地球物理學(xué)進(jìn)展;2012年03期

6 劉生貴;;采空區(qū)注漿加固效果檢測方法的現(xiàn)狀研究[J];工程勘察;2007年10期

7 李秀坤;謝磊;秦宇;;應(yīng)用希爾伯特黃變換的水下目標(biāo)特征提取[J];哈爾濱工程大學(xué)學(xué)報(bào);2009年05期

8 李巧梅;王獻(xiàn)杰;朱麗;;地震屬性提取技術(shù)研究[J];內(nèi)蒙古石油化工;2012年07期

9 舒志樂;劉�？h;張玲;;隧道前方不良地質(zhì)體探地雷達(dá)正演模擬研究[J];西華大學(xué)學(xué)報(bào)(自然科學(xué)版);2012年04期

10 王世瑞;王樹平;狄?guī)妥?任敦占;王振國;陸道林;;基于地震屬性特征的河道砂體預(yù)測方法[J];石油地球物理勘探;2009年03期

相關(guān)博士學(xué)位論文 前4條

1 王婷;EMD算法研究及其在信號(hào)去噪中的應(yīng)用[D];哈爾濱工程大學(xué);2010年

2 史凌峰;探地雷達(dá)檢測中的關(guān)鍵技術(shù)研究[D];西安電子科技大學(xué);2008年

3 馮宏;調(diào)頻連續(xù)波探地雷達(dá)關(guān)鍵技術(shù)研究[D];長安大學(xué);2012年

4 趙文軻;探地雷達(dá)屬性技術(shù)及其在考古調(diào)查中的應(yīng)用研究[D];浙江大學(xué);2013年

，

本文編號(hào)：2037582