摘要
提出了一种基于波叠加的噪声源识别方法,并通过仿真验证了在自由场条件下该方法的正确性。根据波叠加的原理,声源产生的空间声场可以用其内部的一系列虚源点来等效代替,而虚源源强可以通过匹配场点的声压来求得,进而由这些虚源重构任意场点的各种声学量。由于噪声源识别问题对测量误差非常敏感,采用了Tikhonov方法进行正则化滤波。通过两个典型的仿真算例,研究了其定位精度与分辨率精度,取得了良好的效果,表明该方法可以在自由场条件下对任意形状物体进行噪声源识别。
An efficient technique of noise source identification(NSI) based on the Wave Superposition Method(WSM) is proposed.According to the theory of WSM,the total acoustic field radiated by an arbitrary-shaped noise radiator can be substituted by a series of virtual source points.These point sources locate on a surface interior to the body of the radiator,and their strengths can be evaluated in terms of a pressure measurement on another surface.Once calculated,the strengths of these point sources allow the pressure,intensity and instantaneous point velocity etc distribution exterior to the radiator to be calculated.With the visualized maps and animations,main noise sources can be easily located and quantified,also the noise transmission path can be found.A regularization method(Tikhonov) is applied to the technique since there inevitably exits errors in a real measurement.Through two typical simulations,the validity and efficiency of this technique are tested.In the further research,all these techniques including measurement technique,WSM and visualization technique can be integrated into a novel NSI system.
出处
《振动与冲击》
EI
CSCD
北大核心
2006年第4期58-60,共3页
Journal of Vibration and Shock
基金
国家自然科学基金重点资助项目(50335030)
日本Fujitec公司资助项目
关键词
噪声源识别
波叠加
正则化
noise source indentification,wave superposition method,regularization
