摘要
为降低初始频率和信号噪声对变分非线性chirp模态分解(variational nonlinear chirp mode decomposition,VNCMD)的影响,提出了一种基于数据驱动自适应变分非线性chirp模态分解(data-driven adaptive variational nonlinear chirp mode decomposition,DDAVNCMD)的方法。通过模态能量占比确定响应信号的模态个数,同时采用导数归一化算法初步估算模态分量的初始频率,并添加迭代时变滤波器来降低噪声的影响,在此基础上再对响应信号进行VNCMD。通过单分量和多分量解析信号及拉索结构试验对所提方法进行验证。研究结果表明,基于DDAVNCMD的瞬时频率识别方法具有较好的准确性和抗噪性。
To reduce the impact of initial instantaneous frequency and signal noise on the variational nonlinear chirp mode decomposition(VNCMD),a data-driven adaptive variational nonlinear chirp mode decomposition(DDAVNCMD)was proposed in this paper.The modal number of the response signal was obtained by the proportion of modal energy,and the derivative normalization algorithm was used to preliminarily estimate the initial instantaneous frequencies of the modal components.An iterative time-varying filter was also added to reduce the noise effect.Based on this,the response signal was then subjected to VNCMD.The proposed method was validated through single-component and multi-component analytic signals,as well as a cable structure experiment.The research results indicate that the instantaneous frequency identification method based on DDAVNCMD has good accuracy and anti-noise performance.
作者
袁平平
满镇
赵周杰
任伟新
YUAN Pingping;MAN Zhen;ZHAO Zhoujie;REN Weixin(School of Civil Engineering and Architecture,Jiangsu University of Science and Technology,Zhenjiang 212100,China;China Construction Second Engineering Bureau Ltd.,Beijing 100160,China;College of Civil and Transportation Engineering,Shenzhen University,Shenzhen 518061,China)
出处
《振动与冲击》
EI
CSCD
北大核心
2024年第20期18-25,共8页
Journal of Vibration and Shock
基金
国家自然科学基金面上项目(51979130)。
