笛卡尔坐标系
声学
网格
传感器
点源
规则网格
超声波传感器
物理
点(几何)
光学
几何学
计算机科学
数学
作者
Eleanor Martin,Yan To Ling,Bradley E. Treeby
出处
期刊:IEEE Transactions on Ultrasonics Ferroelectrics and Frequency Control
[Institute of Electrical and Electronics Engineers]
日期:2016-10-01
卷期号:63 (10): 1535-1542
被引量:35
标识
DOI:10.1109/tuffc.2016.2600862
摘要
Accurately representing the behavior of acoustic sources is an important part of ultrasound simulation. This is particularly challenging in ultrasound therapy where multielement arrays are often used. Typically, sources are defined as a boundary condition over a 2-D plane within the computational model. However, this approach can become difficult to apply to arrays with multiple elements distributed over a nonplanar surface. In this paper, a grid-based discrete source model for single- and multielement bowl-shaped transducers is developed to model the source geometry explicitly within a regular Cartesian grid. For each element, the source model is defined as a symmetric, simply connected surface with a single grid point thickness. Simulations using the source model with the open-source k-Wave toolbox are validated using the Rayleigh integral, O'Neil solution, and experimental measurements of a focused bowl transducer under both quasi-continuous wave and pulsed excitations. Close agreement is shown between the discrete bowl model and the axial pressure predicted by the O'Neil solution for a uniform curved radiator, even at very low grid resolutions. Excellent agreement is also shown between the discrete bowl model and experimental measurements. To accurately reproduce the near-field pressure measured experimentally, it is necessary to derive the drive signal at each grid point of the bowl model directly using holography. However, good agreement is also obtained in the focal region using uniformly radiating monopole sources distributed over the bowl surface. This allows the response of multielement transducers to be modeled, even where measurement of an input plane is not possible.
科研通智能强力驱动
Strongly Powered by AbleSci AI