小波
小波包分解
激发
振动
能量(信号处理)
网络数据包
声学
结构健康监测
稳健性(进化)
小波变换
计算机科学
结构工程
物理
数学
工程类
人工智能
统计
计算机网络
生物化学
化学
量子力学
基因
标识
DOI:10.1142/s0219455422500912
摘要
Damage detection technology based on wavelet packet energy has been widely applied to structural health monitoring, especially for one-dimensional structures. With more complex vibration characteristics than one-dimensional structure, two-dimensional structure is common in ship structure. It is a challenge for using wavelet packet energy methods to detect damage in two-dimensional structures. In this study, based on the wavelet packet energy eigenvector, a novel concept of detection index ratio (DIR) is defined to characterize the damage location ability of the excitation force. Based on the above research, a multi-excitation method of damage location in plate-like structures is proposed. In this method, a series of half-sine pulse excitation forces with different frequencies are applied to the structure, and wavelet packet transform (WPT) is used to decompose the vibration response of the intact structure and the damaged structure. The energy ratio changes in frequency bands on all measuring points are taken as the preliminary damage detection index, and the final detection index is obtained by synthesizing the preliminary indexes under various excitation. Moreover, to verify the effect of the proposed method, several tested damage cases are presented. The results show that the proposed method can locate single and multiple damage more accurately and has better noise robustness than the traditional detection method under single excitation.
科研通智能强力驱动
Strongly Powered by AbleSci AI