基于CYCBD和CPAD的螺旋桨空化噪声特征提取方法  

A feature extraction method of propeller cavitation noise based on CYCBD and CPAD

作　　者：李阳 崔立林[1] 陈景兵[1] 余文晶[1] LI Yang;CUI Lilin;CHEN Jingbing;YU Wenjing(Inst.of Noise and Vibration,Naval Univ.of Engineering,Wuhan 430033,China)

机构地区：[1]海军工程大学振动与噪声研究所,武汉430033

出　　处：《海军工程大学学报》2023年第5期74-79,共6页Journal of Naval University of Engineering

基　　金：海军工程大学自主立项基金资助项目(4142Z3391233)。

摘　　要：为实现对螺旋桨空化起始的有效检测,设计了一种基于最大二阶循环平稳盲解卷积(maximum second-order cyclic stationary blind deconvolution,CYCBD)和空化脉冲检测器(cavitation pulse activity detector,CPAD)的螺旋桨空化特征提取方法。首先,利用CYCBD方法对螺旋桨噪声信号进行预处理,提取信号中的周期性脉冲成分;然后,构建基于局部能量密度检测和Hilbert边际谱质心(Hilbert marginal spectral centroid,HMSC)判别串联结构的检测器,提取信号中的高频脉冲;最后,根据一定旋转周期内的高频脉冲数目,对空化状态进行区分。试验结果表明:所提出的方法可有效提取螺旋桨空化噪声特征,为螺旋桨空化状态检测提供依据。In order to effectively detect the inception of propeller cavitation,a method based on maximum second-order cyclic stationary blind deconvolution(CYCBD)and cavitation pulse activity detector(CPAD)was designed to extract propeller cavitation features.Firstly,the CYCBD method was used to preprocess the propeller noise signal and extract the periodic pulse component in the signal.Then,a CPAD based on local energy density detection and Hilbert marginal spectral centroid(HMSC)was constructed,and the cavitation pulse in the signal was extracted.Finally,the cavitation states were identified based on the number of cavitation pulses in a given period of rotation.The experimental results show that the proposed method can effectively extract the characteristics of propeller cavitation noise and provide a basis for the detection of propeller cavitation state.

关 键 词：螺旋桨 空化噪声 特征提取 CYCBD 脉冲检测 

分 类 号：O427.4[理学—声学]