Validation and evaluation of damage identification using probability-based diagnostic imaging on a stiffened composite panel. (November 2015)

• Record Type: Journal Article
• Authors: Wu, Zhanjun; Liu, Kehai; Wang, Yishou; Zheng, Yuebin
• Is Part Of: Journal of intelligent material systems and structures. Volume 26:Number 16(2015:Nov.)
• Page Start: 2181
• Page End: 2195
• Publication Date: 2015-11
• Subjects: Structure health monitor -- damage identification -- probability-based diagnostic image; Smart materials -- Periodicals; Intelligent control systems -- Periodicals; Artificial intelligence -- Periodicals; Matériaux intelligents -- Périodiques; Commande intelligente -- Périodiques; Intelligence artificielle -- Périodiques; 620.11
• DOI: 10.1177/1045389X14549873
• Languages: English
• ISSNs: 1045-389X

Abstract

Probability-based diagnostic imaging, as one of the damage identification methods using ultrasonic guided waves, has been attracting increasing attention by researchers in the community of structural health monitoring. However, the probability-based diagnostic imaging algorithm's influencing parameters, including the selection of certain damage index and frequency, the network of sensing paths, and the size of the effective elliptical distribution area, are empirically determined. This experience dependency limits the application of the method to identify damages in real-world practices. Therefore, it is important to clarify the influences of the above-mentioned various factors on the damage identification. However, the complexity of these factors makes it almost impossible to interpret the influencing mechanisms directly. Thus, a fusion image approach of multiple frequencies is employed to eliminate the influence of different frequencies, while a histogram-based method is proposed to evaluate the reliability of the fusion result. Meanwhile, a unit weight distribution function, considering both the network of sensing paths and the size of the effective elliptical distribution area, is presented in the analysis. Then, the influencing mechanisms are studied and discussed in detail, and a methodology is proposed to optimize the network and the scaling parameter which controls the affected zone.