Open Access

ARTICLE

A Hybrid of Interval Wavelets and Wavelet Finite Element Model for Damage Detection in Structures

Jiawei Xiang¹, Toshiro Matsumoto², Yanxue Wang³, Zhansi Jiang⁴

1 Department of Mechanical Science and Engineering, Nagoya University, Furo-cho, Chikusa-ku, Nagoya, 464-8603, Japan; School of Mechanical and Electrical Engineering, Guilin University of Electronic Technology, Guilin, 541004, China. Corresponding Author: Jiawei Xiang, E-mail: wxw8627@163.com
2 Department of Mechanical Science and Engineering, Nagoya University, Furo-cho, Chikusa-ku, Nagoya, 464-8603, Japan
3 School of Mechanical and Electrical Engineering, Guilin University of Electronic Technology, Guilin, 541004, China
4 School of Mechanical and Electrical Engineering, Guilin University of Electronic Technology, Guilin, 541004, China

Computer Modeling in Engineering & Sciences 2011, 81(3&4), 269-294. https://doi.org/10.3970/cmes.2011.081.269

Abstract

Damages occurred in a structure will lead to changes in modal parameters (natural frequencies and modal shapes). The relationship between modal parameters and damage parameters (locations and depths) is very complicated. Single detection method using natural frequencies or modal shapes can not obtain robust damage detection results from the inevitably noise-contaminated modal parameters. To eliminate the complexity, a hybrid approach using both of wavelets on the interval (interval wavelets) method and wavelet finite element model-based method is proposed to detect damage locations and depths. To avoid the boundary distortion phenomenon, Interval wavelets are employed to analyze the finite-length modal shape to decompose into approximation and detailed signals. Damage locations will be detected by showing some peaks on the figures of detailed signal. To detect damage depths, the relationship between natural frequencies and damage depths (the damage depth detection database) is constructed using wavelet finite element method. Several natural frequencies obtained by experimental modal analysis are employed as inputs to the constructed database using particle swarm optimization (PSO) to search for damage depths. Numerical examples of beam and plate structures show that the new approach is robust to boundary distortion phenomenon and environment noise.

---

Keywords

---