INSTITUTE OF PHYSICS PUBLISHING SMART MATERIALS AND STRUCTURES Smart Mater. Struct. 12 (2003) 776–784 PII: S0964-1726(03)67504-X Structural health monitoring and damage detection using a sub-domain inverse method F J Carrion 1 , J F Doyle 2 and A Lozano 3 1 Instituto Mexicano del Transporte, Km 12000 Carretera Quer´ etaro-Galindo, Sanfandila Municipio de Pedro Escobedo, 76700 Quer´ etaro, Mexico 2 Aeronautics and Aerospace Department, Purdue University, West Lafayette, IN, USA 3 Consejo de Ciencia y Tecnolog´ ıa del Estado de Quer´ etaro, 76000 Quer´ etaro, Mexico Received 26 April 2002, in final form 19 June 2003 Published 25 September 2003 Online at stacks.iop.org/SMS/12/776 Abstract Health monitoring of modern complex structures poses new challenges as regards increased safety and operational reliability, while recent sensor technology opens up many possibilities for previously undreamt of methods of analysis. The need for different schemes or algorithms for damage detection, remote monitoring and continuous real time evaluation is increasingly pressing. In this work, a sub-domain inverse method based on the analysis of the wave propagation in structures is used for the identification of the damage response vectors, with an adapted minimum rank perturbation formulation for the calculation of the perturbation stiffness matrix of a damaged structure. Results show that simple experimental procedures can be utilized and several strategies can be adopted for fast monitoring or detailed analysis. Analyses in sub-regions or complete structures are possible and testing can be done under random excitations and with structural non-linear behaviour or unknown structural parameters. The proposed method of using smart structures with many embedded sensors is potentially useful for structural health monitoring and damage detection with almost no limitation on the type or distribution of the sensors. (Some figures in this article are in colour only in the electronic version) 1. Introduction Modern civil and aeronautical structures are getting more complex in size and design, new concepts and materials are increasingly being used and the environmental and operational conditions are ever more demanding. Moreover, structures are becoming important components of integrated systems (e.g. highway bridges in integrated transport systems); for these, continuous evaluation of the structural condition is necessary and, in some cases, it must be in real time [1]. At the same time, sensor technology has grown and has expanded very many possibilities for measuring, evaluation and control of engineering systems; multiple sensors of different types are now affordable and we can envisage the use of thousands of embedded sensors in smart structures [2–4]. Under these operational conditions, health monitoring of new structures and sensor technology developments pose various challenges and alternatives for innovative theory and experimental schemes [5]. Structural evaluation to determine damage, deterioration and/or operational anomalies in modern complex civil and aeronautic structures is essential to determine safety, operational reliability and residual life [1]. Traditionally, most of the damage detection programs are based on visual inspections, which are costly and difficult to carry out due to the inaccessibility of most of the parts of the structure. Furthermore, internal damage cannot be detected and no quantitative value of the damage and the residual strength of the structure is produced by the visual inspections [6]. Recent health monitoring systems have included different non- 0964-1726/03/050776+09$30.00 © 2003 IOP Publishing Ltd Printed in the UK 776