统计最优近场声全息重建精度和计算速度优化方法

The method for improving the reconstruction accuracy and computational speed of the statistically optimized nearfield acoustic holography

分析了统计最优近场声全息(Statistically optimized nearfield acousticholography,简称SONAH)的重建过程发现采用SONAH重建噪声源表面法向振速时误差较大的原因在于正则化参数选取较小。在此基础上提出一种单元平面波优化选择方法,该方法保留了单元平面波中的全部传播波和部分倏逝波,去除了一些较高波数的倏逝波成分,保证了重建过程中正则化参数的准确选取。另外,采用单元平面波优化选择方法还可以降低SONAH中传递矩阵的阶数,从而解决SONAH的计算速度随着测量点数目的增加急剧变慢的问题。通过数值仿真和实验对所提出的单元平面波优化选择方法的有效性进行了验证,结果表明采用该方法后SONAH的计算速度和法向振速的重建精度都得到了较大提高。

By analyzing the reconstruction process of the statistically optimized nearfield acoustic holography （SONAH）, it is found that selecting regularization parameter too small is the essential reason for the low accuracy of SONAH when normal velocity on the surface of noise sources is reconstructed. In order to improve the reconstruction accuracy of SONAH, a method for optimized selection of elementary planar waves is proposed to guarantee the correct selection of the regularization parameter, in which all the propagating waves and parts of evanescent waves are retained and parts of the evanescent waves related to high wave number are eliminated. Moreover the order of the transfer matrix of SONAH decreases after using the proposed method, consequently the problem that the computational speed will become slow rapidly as soon as the number of measurement points increase can be solved simultaneously. The numerical simulation and the experiment are conducted to demonstrate the validity that both the reconstruction accuracy and the computational method of the proposed method, and the results of which show speed of SONAH are improved a lot by the proposed