【摘要】：1979年W.Munk等人提出利用聲學(xué)方法對(duì)中尺度海洋環(huán)境進(jìn)行監(jiān)測(cè)的海洋聲層析方法,解決逆介質(zhì)問(wèn)題。該方法利用傳播時(shí)延等聲傳播觀測(cè)量反演經(jīng)過(guò)海域的聲速與流速等海洋參數(shù)。聲層析是結(jié)合了海洋測(cè)量、聲傳播建模和聲學(xué)逆問(wèn)題的多維問(wèn)題,傳統(tǒng)的聲層析方法是基于射線模型的微擾法。微擾法利用從聲源到接收陣元的多徑信號(hào)走時(shí)來(lái)反演垂直分層的聲速剖面。聲線路徑與路徑上慢度(聲速的倒數(shù))的線性積分關(guān)系構(gòu)成了聲學(xué)逆介質(zhì)問(wèn)題的基本方程,通過(guò)最小二乘等逆問(wèn)題求解方法可以估計(jì)出實(shí)際的聲速剖面。但受發(fā)射信號(hào)傳播時(shí)延測(cè)量的分辨力與射線模型的高頻近似制約,在低頻深海環(huán)境下微擾法的分辨力和精度都有所不足。本文提出的傳播時(shí)延敏感核聲層析方法從波動(dòng)理論與波恩近似推導(dǎo)出聲速擾動(dòng)與傳播時(shí)延的三維積分關(guān)系,可以利用兩節(jié)點(diǎn)間的低頻聲信號(hào)反演海域中任意位置的聲速擾動(dòng)。傳播時(shí)延敏感核方法利用信號(hào)的波峰時(shí)延替代多路徑波包,增加了信號(hào)的觀測(cè)量；同時(shí),三維的傳播時(shí)延敏感核函數(shù)將聲線路徑積分提升到三維的空間積分,提高了觀測(cè)的空間分辨力。針對(duì)傳播時(shí)延敏感核聲層析方法,本文研究了核函數(shù)的理論推導(dǎo)、波峰模板匹配以及聲層析逆問(wèn)題的求解方法。首先在寬帶模型下,用簡(jiǎn)正波模型推導(dǎo)出三維核函數(shù)表達(dá)；然后提出了基于相關(guān)的波形模板匹配方法,能有效分辨出匹配的波峰；最后,針對(duì)不同的海洋聲擾動(dòng)環(huán)境,提出了帶有平滑約束的最小二乘方法、截?cái)郤VD方法以及凸優(yōu)化方法,來(lái)求解聲層析逆問(wèn)題。本文對(duì)傳統(tǒng)微擾法聲層析與傳播時(shí)延敏感核方法聲層析分別進(jìn)行了實(shí)驗(yàn)仿真,并比較了兩者的聲速反演結(jié)果,驗(yàn)證了傳播時(shí)延敏感核方法能有效提高精度與分辨力。并在以上研究結(jié)果的基礎(chǔ)上,開(kāi)發(fā)出一套海洋聲層析軟件。
[Abstract]:In 1979 W.Munk et al proposed an acoustic tomography method to monitor the mesoscale marine environment using acoustic method to solve the problem of inverse medium. In this method, the ocean parameters such as velocity and velocity of sound passing through the sea area are inversed by equal-acoustic propagation observations with propagation delay. Acoustic tomography is a multidimensional problem which combines ocean measurement, acoustic propagation modeling and acoustic inverse problem. The traditional acoustic tomography method is based on the perturbation method of ray model. The multipath signal travel time from the sound source to the receiving array element is used to retrieve the vertical stratified sound velocity profile by the perturbation method. The linear integral relation between sound path and slowness (reciprocal of sound velocity) constitutes the basic equation of acoustic inverse medium problem. The actual sound velocity profile can be estimated by solving inverse problem such as least square method. However, due to the resolution of transmission delay measurement and the high frequency approximation of the ray model, the resolution and accuracy of the perturbation method in the low-frequency deep-sea environment are insufficient. The propagation delay-sensitive nuclear acoustic tomography proposed in this paper deduces the three-dimensional integral relationship between the acoustic velocity disturbance and the propagation delay from the wave theory and the Bonn approximation. The low-frequency acoustic signal between the two nodes can be used to retrieve the acoustic disturbance at any position in the sea area. The propagating delay sensitive kernel method uses the wave peak delay of the signal to replace the multipath wave packet, which increases the observation quantity of the signal. At the same time, the three-dimension propagation delay sensitive kernel function promotes the acoustic path integral to the three-dimensional spatial integral, which improves the spatial resolution of the observation. In this paper, the theoretical derivation of kernel function, the matching of wave peak template and the solution of inverse problem of acoustic tomography are studied for propagation delay sensitive nuclear acoustic tomography. Firstly, the expression of 3D kernel function is deduced by using the normal wave model under the wideband model, and then the correlation-based waveform template matching method is proposed, which can effectively distinguish the matching peak. Finally, for different ocean acoustic disturbance environments, a least square method with smooth constraints, a truncated SVD method and a convex optimization method are proposed to solve the inverse problem of acoustic tomography. In this paper, the acoustic tomography of the traditional perturbation method and the propagating delay sensitive kernel method are simulated, and the results of the acoustic velocity inversion are compared. The results show that the propagating delay sensitive kernel method can effectively improve the accuracy and resolution. On the basis of the above research results, a set of ocean acoustic tomography software is developed.
【學(xué)位授予單位】：浙江大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2016
【分類(lèi)號(hào)】：P733.2

【相似文獻(xiàn)】

相關(guān)期刊論文 前8條

1 吳海濤;宋金安;;預(yù)報(bào)衛(wèi)星信號(hào)傳播時(shí)延的統(tǒng)計(jì)方法[J];陜西天文臺(tái)臺(tái)刊;1992年01期

2 梁仲寰;預(yù)測(cè)地波定時(shí)信號(hào)傳播時(shí)延的經(jīng)驗(yàn)方法[J];陜西天文臺(tái)臺(tái)刊;1998年01期

3 梁仲寰;;復(fù)雜地面上低頻時(shí)間信號(hào)傳播時(shí)延預(yù)測(cè)精度問(wèn)題[J];天文學(xué)報(bào);1988年04期

4 ;突然低頻傳播擾動(dòng)對(duì)低頻定時(shí)校頻的影響及其修正[J];時(shí)間頻率公報(bào);1992年03期

5 彭金鳳;;廣州人衛(wèi)站對(duì)BPL地波信號(hào)的監(jiān)測(cè)結(jié)果及其誤差討論[J];陜西天文臺(tái)臺(tái)刊;1985年01期

6 ;一跳天波場(chǎng)強(qiáng)和傳播時(shí)延的計(jì)算[J];陜西天文臺(tái)臺(tái)刊;1986年02期

7 梁仲寰;陳洪卿;李達(dá)銘;王發(fā)潮;;BP L在幾條路徑上的傳播時(shí)延和大地等效電導(dǎo)率[J];陜西天文臺(tái)臺(tái)刊;1984年02期

8 ;[J];;年期

相關(guān)會(huì)議論文 前3條

1 趙景斐;趙峗;孫海燕;史俊文;;地球同步衛(wèi)星導(dǎo)航系統(tǒng)授時(shí)中的大氣傳播時(shí)延誤差[A];第二屆中國(guó)衛(wèi)星導(dǎo)航學(xué)術(shù)年會(huì)電子文集[C];2011年

2 高大治;趙振東;劉進(jìn)忠;王好忠;王寧;;AEYFI-05簡(jiǎn)正波傳播時(shí)延“高頻”起伏的原因?[A];2008年全國(guó)聲學(xué)學(xué)術(shù)會(huì)議論文集[C];2008年

3 周慶勇;杜蘭;;一種脈沖星計(jì)時(shí)模型與其精度評(píng)定[A];第一屆中國(guó)衛(wèi)星導(dǎo)航學(xué)術(shù)年會(huì)論文集（中）[C];2010年

相關(guān)重要報(bào)紙文章 前1條

1 ;WAN仿真器簡(jiǎn)化應(yīng)用測(cè)試[N];網(wǎng)絡(luò)世界;2006年

相關(guān)碩士學(xué)位論文 前1條

1 汪非易;深海聲層析傳播時(shí)延敏感核方法[D];浙江大學(xué);2016年

，

本文編號(hào)：2306336