首页> 外文会议>European Conference on Underwater Acoustics >MODE PARABOLIC EQUATIONS FOR THE MODELING OF THREE-DIMENSIONAL SOUND PROPAGATION EFFECTS IN SHALLOW WATER
【24h】

MODE PARABOLIC EQUATIONS FOR THE MODELING OF THREE-DIMENSIONAL SOUND PROPAGATION EFFECTS IN SHALLOW WATER

机译:用于浅水中三维声音传播效应建模的模式抛物线方程

获取原文

摘要

The modeling of acoustic wave propagation in three-dimensional environments is a subject of extensive study in last two decades. There exist large variety of substantially different approaches to these problems [1]. However, it is widely accepted that most direct methods of solution of acoustic wave equation in 3D-waveguides (e.g. using finite differences) are impractical due to computational complexity. This is the main reason why these equations are usually transformed into frequency domain where various parabolic approximations are applied [1-3] (since generally parabolic equations are more attractive in terms of efficiency). Most 3D parabolic equations however suffer from the inaccurate treatment of the inhomogeneities of interfaces - i.e. surfaces in the waveguides where media parameters have discontinuities. This problem emerges from the conventional derivation of the parabolic equations [1,3] (which is based on operator square root approximation) where in fact the absence of any horizontal inhomogeneities is assumed. Hence, such derivations do not lead explicitly to any consistent boundary conditions. This problem seems to severely affect even 2D parabolic equations theory where it is still subject of intensive study. Another common approach to 3D acoustic simulation is based on the normal modes theory [1]. Within this method at any horizontal point (x, y) acoustic field is represented as a linear combination of (local) acoustic spectral problem solutions - normal modes. The coefficients of this decomposition may depend on horizontal variables and must be calculated to obtain the acoustic field. This may be accomplished either by derivation of some equations governing the horizontal variations of these coefficients or by direct matching of sound field values along the boundaries of several regions in horizontal plane (x,y) in which the medium is assumed to be horizontally homogeneous [1,4]. The latter method (besides the limitations and inconveniences produced by such domain partitioning) leads to a computationally intensive procedure of the solution of very large system of linear equations. This is why in our opinion possible applications of such method are limited to construction of benchmark problems solutions (like in [4]). Approximate equations for the coefficients of the modal expansion of acoustic field seem to be much more fruitful in terms of efficiency and range of problems they can treat. Most of existing achievements within this approach were limited so far to the simulations of sound propagation in the environments with 3D-inhomogeneous sound speed profiles [5]. However the most interesting practical problems of shallow water acoustics usually involve 3D bottom relief (i.e. interface) inhomogeneities. A computationally efficient method for the solution of such problems is developed in [6] where the parabolic equations for modal coefficients accounting bottom relief and density inhomogenities were introduced. In this work we recall the derivation of these mode parabolic equations (MPEs) following [6] and address some issues arising in the applications of this method to the practical problems of shallow water acoustics. These equations are derived using the generalized multiple scale method [7]. The distinctive feature of this derivation is accurate and "uniform" treatment of both sound-speed and interface inhomogeneities.
机译:三维环境中声波传播的建模是过去二十年的广泛研究的主题。这些问题存在大量基本不同的方法[1]。然而,众所周知,由于计算复杂性,3D波导中的声波方程(例如,使用有限差异)的最大直接方法是不切实际的。这是这些等式通常转换成频域的主要原因,其中施加各种抛物面近似[1-3](因为通常抛物线方程在效率方面更具吸引力)。然而,大多数3D抛物线方程受到界面的不均匀性的不准确处理 - 即媒体参数具有不连续性的波导中的表面。该问题从抛物线方程的传统推导出来了[1,3](基于操作员平方根近似),实际上假设不存在任何水平不均匀性。因此,这种衍生不会明确地延伸到任何一致的边界条件。这个问题似乎严重影响了2D抛物线方程理论,它仍然是密集研究的主题。用于3D声学仿真的另一种常见方法是基于正常模式理论[1]。在任何水平点(x,y)处的该方法中,声场表示为(本地)声光谱问题解决方案的线性组合 - 正常模式。该分解的系数可以取决于水平变量,并且必须计算以获得声场。这可以通过导出控制这些系数的水平变化的一些方程或通过沿着水平面(x,y)中的若干区域的边界直接匹配的一些方程来实现,其中假设介质被培养为水平均匀[ 1,4]。后一种方法(除了通过这种域分区产生的局限性和不便)导致基于线性方程的非常大的系统解决方案的计算密集型过程。这就是我们认为这种方法可能应用的原因仅限于建造基准问题解决方案(如[4])。在声场的模态扩展系数的近似方程似乎在他们可以治疗的效率和问题范围内更加富有成效。到目前为止,这种方法内的大多数成就都是有限的,以便在具有3D非均匀声速分布的环境中进行声音传播的模拟[5]。然而,浅水声学的最有趣的实际问题通常涉及3D底部浮雕(即界面)不均匀性。在[6]中开发了用于解决这些问题的计算上有效的方法,其中引入了用于模数系计数底部浮雕和密度偏离的抛物线方程。在这项工作中,我们记得这些模式抛物线方程(MPE)的推导,并解决了这种方法应用于浅水声学的实际问题中出现的一些问题。使用广义多种刻度方法[7]导出这些方程。这种衍生的独特特征是准确的,并“均匀”处理声速和界面不均匀性。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号