水轮机空化声发射信号降噪与混沌图像特征提取  

Denoising and chaotic feature extraction of acoustic emission signals of hydraulic turbine cavitation

作　　者：刘忠[1] 李显伟 邹淑云[1] 王文豪 周泽华 LIU Zhong;LI Xianwei;ZOU Shuyun;WANG Wenhao;ZHOU Zehua(School of Energy and Power Engineering,Changsha University of Science and Technology,Changsha 410114,China)

机构地区：[1]长沙理工大学能源与动力工程学院,湖南长沙410114

出　　处：《哈尔滨工程大学学报》2023年第8期1361-1367,共7页Journal of Harbin Engineering University

基　　金：国家自然科学基金项目(52079011);湖南省研究生科研创新项目(CX20220927);湖南省自然科学基金项目(2023JJ30032).

摘　　要：针对水轮机空化声发射信号存在噪声影响信号特征有效提取的问题,本文建立了基于傅里叶分解与多分辨奇异值分解的降噪和混沌特征提取的水轮机空化声发射信号处理方法。采用傅里叶分解算法将水轮机空化声发射信号分解为若干个固有频带函数,计算其相关系数。利用多分辨奇异值分解算法对相关系数较小的固有频带函数进行降噪,再将降噪后的固有频带函数与相关系数较大的固有频带函数进行重构,完成信号降噪。结果表明:将相空间重构得到相轨迹图和Poincaré截面图作为信号特征,本文降噪方法可以更好实现水轮机空化声发射信号降噪;混沌特征图像可以反映空化状态变化规律。This study aims to address the problem concerning the influence of noise on the effective extraction of hydraulic turbine cavitation acoustic emission signal characteristics.Therefore,a processing method for the turbine cavitation acoustic emission signal is established based on the Fourier decomposition method and multiresolution singular value decomposition FDM-MRSVD denoising and chaos feature extraction.First,the cavitation acoustic emission signal is decomposed into several Fourier intrinsic band functions(FIBFs)of instantaneous frequencies based on FDM,and the correlation coefficients are calculated.The FIBFs with a small correlation coefficient are denoised by using MRSVD,and the denoised FIBFs are reconstructed with a large correlation coefficient FIBFs to complete signal denoising.The phase space is reconstructed,obtaining the phase locus and Poincar section as the signal characteristics.The experimental results show that the denoising method of FDM-MRSVD can achieve superior noise reduction by cavitation acoustic emission of hydraulic turbines.The chaotic characteristic images can represent the change rule of the cavitation state.

关 键 词：水轮机 空化 声发射 傅里叶分解 多分辨奇异值分解 相空间重构 混沌特征 

分 类 号：TK73[交通运输工程—轮机工程]