摘要为有效解决近场声全息(Near field acoustic holograph NAH)技术在水下振动声源识别方法中面临不适定性问题,将平面NAH技术的声源识别过程转化为线性系统的求解过程,探明声源识别中不适定性问题产生的根源,考虑全息面测量声压和传递矩阵均存在误差,提出牛顿迭代的总体最小二乘(NTLS)正则化方法稳定NAH重建过程。新方法以TLS正则化算法为基础建立目标函数,将目标函数改化为具有凸函数的性质;然后采用具有二阶收敛速度牛顿迭代法求解;基于L曲线法特性,提出自适应迭代方法确定正则化参数。开展了NAH仿真和试验研究,提出构建良态传递矩阵的策略,最终验证了NTLS正则化的NAH技术在水下振动声源识别和定位中具有较好的精度。
Near field acoustic holography (NAH) is a very powerful tool for identifying sound source. The identification process based on NAH was firstly transformed into linear equation in this paper, which is a linear, ill-posed inverse problem. And then, an ill-conditioned property of the ill-posed problem was ascertained. To find meaningful solutions of such systems, a novel Newton iteration total least squares (NTLS) method when both the transfer matrix and the observation term were contaminated by some errors was proposed. The NTLS method included three creative schemes: we firstly converted the Tikhonov regularization TLS problem to an unconstrained optimization problem with the properties of a convex function; the optimization problem would be solved with the Newton iteration method, and moreover, the adaptive strategy for selecting regularization parameter was also established. Finally, the robustness and effectiveness of the proposed methods are demonstrated by simulations and test of NAH. The results prove that the NTLS method provide a significant improvement in identifying and locating sound source.
收稿日期: 2015-07-14
出版日期: 2016-10-25
引用本文:
张磊 曹跃云 杨自春. 迭代总体最小二乘正则化的近场声全息方法研究[J]. 振动与冲击, 2016, 35(21): 96-101.
ZHANG Lei, CAO Yun-yue, YANG Zi-chun. Iteration total least square regularization for near field acoustic holography. JOURNAL OF VIBRATION AND SHOCK, 2016, 35(21): 96-101.
[1] 曹跃云,张磊,杨自春.舰船振动噪声传递路径分析及试验验证[J].振动与冲击,2013,32(22):158-162.
ZHANG Lei, CAO yue-yun, YANG Zi-chun. Transfer path analysis of ship vibration sources and experimental validation[J]. Journal of Vibration and Shock,2013,32(22):158-162.
[2] 郭小霞.水下结构辐射噪声源快速诊断识别研究[D].哈尔滨:哈尔滨工程大学,2012.
[3] Williams E G,Houston B H.Fast fourier transform and singular value decomposition formulations for patch nearfield acoustical holography[J].J. Acoust. Soc. Am., 2003,114(3): 1322-1333.
[4] Delillo T,Hrycak T,Isakov V.Theory and boundary element methods for near-field acoustic holography[J].Journal of Computational Acoustics,2005,13(1):163-185.
[5] 毕传兴,陈心昭,周蓉,等.Landweber迭代近场声全息[J].科学通报,2006,51(9):1101-1111.
BI Chuan-xing, CHEN Xin-zhao, ZHOU Rong, et al. The near-field acoustic holography based on iteration landweber [J]. Chinese Science Bulletin, 2006,51(9):1101-1111.
[6] Jesper G..A study on regularizration parameter choice in near-field acoustical holography[J].J. Acoust. Soc. Am.,2008,123(5):3385.
[7] 李凌志,李骏,卢炳武, 等.平面近场声全息中的正则化参数的确定[J].声学学报, 2012,35(2): 170-178.
LI Ling-zhi, LI Jun, LU Bing-wu, et al. The determination of regularization parameters in planar nearfield acoustic holography[J]. ACTA Acoustic, 2012,35(2): 170-178.
[8] 贺春东,毕传兴,徐亮,等.用于近场声全息正则化的共轭梯度法[J].振动工程学报,2011,24(1):67-72.
HE Chun-dong, BI Chuan-xin, XU Liang, et al. Regularization of the distributed source boundary point method based nearfield acoustic holography by using conjugate gradient algorithm[J]. Journal of Vibration Engineering, 2011,24(1):67-72.
[9] Maziar S,Hossein Z.Computational experiments on the tikhonov regularization of the total least squares problem[J].Computer Science Journal of Moldova,2009,17(1):14-25.
[10] Beck A,Ben-Tal A,Kanzow C.A fast method finding the global solution of the regularized structured total least squares problem for image deblurring[J].SIAM Matrix Anal. Appl.,2008,30(1):419-443.
[11] Oraintara S,Karl W C,Castanon D A,et al.A method for choosing the regularization parameter in generalized Tikhonov regularization linear inverse problems.Image Processing[J].Proceedings,2000,1:93-96.
[12] 辛雨.基于空间声场变换的近场声全息参数选取及重建误差研究[D].合肥:合肥工业大学,2009.