A Damage Classification technique for Impedance-Based Health Monitoring of Helicopter Blades Jose R. V. Moura, Jr* a , Valder Steffen, Jr a , Daniel J. Inman b a School of Mec. Eng., Fed. Univ. of Uberlandia, 2160 Joao Naves de Avila Av., Uberlandia, MG, Brazil 38400-902; b CIMSS, Virginia Tech, 310 New Eng. Building, MC0261, Blacksburg, VA USA 24061 ABSTRACT One of the most sensitive problems regarding the application of SHM (Structural Health Monitoring) is found in the aeronautical segment. This field presents the necessity of monitoring small structural changes representing damage, due both to economic aspects and safety. In this contribution two helicopter blade structures (pertaining to a civil and a military helicopter) are studied. In both cases, two types of damage are inserted, namely holes and cracks. Through the impedance-based structural health monitoring method, an identification procedure using cluster analysis techniques was performed aiming at distinguishing these two types of damage. Then, a meta-model based on a probabilistic neural network was built for fault position identification. Keywords: Impedance-based SHM, Structural health monitoring, Helicopter blades, Damage detection, Fuzzy C- Means. 1. INTRODUCTION This work is devoted to the development of a methodology for the identification of damage in a real world complex aeronautical structure, highlighting the potentialities of a health-monitoring technique that is based on impedance signals and meta-modeling. For that purpose, two aeronautical structures (helicopter blades) were chosen to test the methodology, namely a civil structure and a military structure. Both structures are made of composite materials and can be considered as complex structures. In the recent years, the significant number of studies involving impedance-based structural health monitoring demonstrates the interest on this topic. The literature shows different approaches to deal with the problem of structural health monitoring [1-4, 6-9, 11, 13-31, 34]. As this technique is able to identify small structural variations, it has been applied to a number of aeronautic and airspace structures in the context of structural reliability. Initially, a clustering analysis procedure for the identification of the type of failure was conducted by using Fuzzy C- Means and K-Means techniques, by evaluating each case of inserted damage in both structures. Finally, two approaches to meta-modeling the impedance-based health-monitoring process are presented. The first one uses probabilistic neural networks and the other uses statistics surface responses to build the meta-models. 2. IMPEDANCE-BASED STRUCTURAL HEALTH MONITORING The impedance-based health monitoring technique has been developed as a promising tool for detecting structural damage in real time and is considered to be a new method for non-destructive evaluation [29]. It has been shown that the electrical impedance from the piezoelectric ceramic (PZT) surface bonded to a structure can be directly associated with the mechanical impedance of the structure to which it is bonded. By using the same piezoelectric element as a sensor/actuator, a simple testing device, containing a smaller number of components and cables has been developed [29]. *mourajr@mourajr.com; phone 55 34 3239-4148; fax 55 34 3239-4206; www.mecanica.ufu.br Sensors and Smart Structures Technologies for Civil, Mechanical, and Aerospace Systems 2008, edited by Masayoshi Tomizuka, Proc. of SPIE Vol. 6932, 69323S, (2008) 0277-786X/08/$18 · doi: 10.1117/12.776751 Proc. of SPIE Vol. 6932 69323S-1 2008 SPIE Digital Library -- Subscriber Archive Copy