Introduction to the use of wavelet multiresolution analysis for intelligent structural health monitoring M.M. Reda Taha, A. Noureldin, A. Osman, and N. El-Sheimy Abstract: This paper suggests the use of wavelet multiresolution analysis (WMRA) as a reliable tool for digital signal processing in structural health monitoring (SHM) systems. A damage occurrence detection algorithm using WMRA augmented with artificial neural networks (ANN) is described. The suggested algorithm allows intelligent monitoring of structures in real time. The probability of damage occurrence is determined by evaluating the wavelet norm index (WNI) representing the energy of a signal describing the change in the system dynamics due to damage. An example application of the proposed algorithm is presented using a finite element simulated structural dynamics of a prestressed concrete bridge. The new algorithm showed very promising results. Key words: structural health monitoring, neural networks, wavelet analysis, signal processing, damage detection. Résumé : Cet article suggère l’utilisation de l’analyse de multi-résolution des ondelettes (WMRA) comme outil fiable de traitement des signaux numériques dans les systèmes de contrôle de l’état structural. Un algorithme de détection d’occurrence de dommages employant la WMRA à laquelle on a ajouté des réseaux neuronaux artificiels est décrit. L’algorithme suggéré permet la surveillance intelligente des structures en temps réel. La probabilité de l’occurrence de dommages est déterminée en évaluant l’indice normatif des ondelettes représentant l’énergie d’un signal décrivant le changement dans la dynamique du système en raison des dommages. Un exemple de l’application de l’algorithme pro- posé est présenté en utilisant une dynamique structurale simulée par éléments finis d’un pont en béton précontraint. Le nouvel algorithme montre des résultats très prometteurs. Mots clés : contrôle de l’état structural, réseaux neuronaux, analyse des ondelettes, traitement des signaux, détection des dommages. Reda Taha et al. 731 1. Introduction Fourier transform (FT) has been used for many years as a reliable tool in signal analysis and has proven incredibly ver- satile in applications ranging from structural dynamics anal- ysis (Humar 2002) to image processing (Gonzalez and Woods 2002). Recently, a new type of transform, the wavelet transform (WT), has been shown to be as powerful and ver- satile as FT, yet without some of the limitations of FT (e.g., Strang and Nguyen 1997). Most current structural health monitoring systems utilize the modal analysis, with its limi- tations represented by insensitivity to localized damage and difficulty to identify damage when significant noise exists. Here, we introduce WT as a new technique for damage iden- tification in structural health monitoring systems that can be integrated with artificial intelligence algorithms to separate the main components of the signals in the time domain and identify changes in these signals due to structural damage. A comparison between WT and FT as means for analyzing transient signals and their advantages and limitations will follow. 1.1. Fourier transform Fourier transform is used to decompose a function in terms of a set of basis functions. The set of complex expo- nentials {exp(iωn), −∞ < f < ∞} forms the set of these basis Can. J. Civ. Eng. 31: 719–731 (2004) doi: 10.1139/L04-022 © 2004 NRC Canada 719 Received 20 November 2003. Revision accepted 19 February 2004. Published on the NRC Research Press Web site at http://cjce.nrc.ca on 1 October 2004. M.M. Reda Taha. 1 Department of Civil Engineering, University of New Mexico, MSC 01 1070, Albuquerque, NM 87131-0001, USA. A. Noureldin. Department of Electrical and Computer Engineering, Royal Military College of Canada, Kingston, ON K7K 7B4, Canada. A. Osman. Department of Electrical and Computer Engineering, The University of Calgary, Calgary, AB T2N 1N4, Canada. N. El-Sheimy. Canada Research Chair in Mobile Multi-sensor Geomatics Systems, Department of Geomatics Engineering, The University of Calgary, Calgary, AB T2N 1N4, Canada. Written discussion of this article is welcomed and will be received by the Editor until 28 February 2004. 1 Corresponding author (e-mail: mrtaha@unm.edu).