噪声源识别的混合波叠加法及其数值仿真研究  

Noise source identification based on combined wave superposition and numerical simulation

作　　者：薛玮飞[1] 陈进[1] 李加庆[1] 

机构地区：[1]上海交通大学振动,冲击,噪声国家重点实验室,上海200240

出　　处：《声学技术》2007年第3期455-459,共5页Technical Acoustics

基　　金：国家自然科学基金资助项目(50335030);日本Fujitec公司资助项目。

摘　　要：提出了以混合波叠加法作为声场的全息变换算法,建立了基于混合波叠加法在声源识别中的理论模型。此法用相对少量的测点数据就可重建任意形状源表面的声场,克服了基于边界元法的声全息中的奇异值积分和解的非唯一性难题,数值仿真结果表明:它具有计算速度快、重建精度高、测量成本低和容易实施等优点,可以精确地识别和定位机械噪声源,在工程实践中具有广泛的应用前景。A combined wave superposition method is proposed to transform near-field acoustic holography. A theoretical model to identify sound sources is established. The method allows reconstruction of the acoustic field from an arbitrary object with relatively few measurements, and overcomes nonuniqueness and ill-posedness difficulties. Numerical simulation shows that it can identify and localize sound sources. avoid reconstruction very time, and reduce cost, thus significantly enhancing the reconstruction efficiency. It can become a powerful noise diagnostic tool in engineering.

关 键 词：波叠加法 声源识别 声场重建 HELS 

分 类 号：TB532[理学—物理] TN911.7[理学—声学]