A contribution to structural healthy monitoring systems electromechanical impedance-based under low signal-to-noise ratio applications

Author(s)

Bruno Albuquerque de Castro

• Document type: Doctoral Thesis
• Press: Bauru. 2019-11-04.
• Institution: Universidade Estadual Paulista (Unesp). Faculdade de Engenharia. Bauru
• Defense date: 2019-11-01
• Advisor: Fabricio Guimarães Baptista

Abstract

Structural Health Monitoring (SHM) systems aim to avoid and detect structural damages at the initial stage in order to promote human safety and reduction of maintenance costs. Among numerous SHM methods for damage detection, the electromechanical impedance (EMI) technique stands out to be a simple methodology that uses low-cost piezoelectric transducers. Basically, the structural damage diagnosis is achieved by extracting the electrical impedance of a transducer attached to a host structure and the failure quantification is accomplished by traditional statistical indices such as the root mean square deviation (RMSD) and the correlation coefficient deviation metric (CCDM). Although the EMI stands out to be an effective and promising SHM technique, research involving the electromechanical impedance is commonly performed in laboratories under ideal conditions. However, in practical applications, during the impedance measurements, it’s inevitable the signal contamination by noise, which can significantly influence and impair the diagnosis of structural condition, limiting the applicability of the EMI. Based on this issue, the objective of this work is to explore new signal processing techniques from impedance measurements of piezoelectric transducers, aiming to expand the applicability of this promising SHM approach for low signal-to-noise ratio inspections conditions. Experimental tests were carried out on a carbon fiber reinforced plate and an aluminum component, which are widely used in the aerospace and naval industry. A study of the noise influence on the traditional damage characterization was accomplished and three types of noise compensation techniques were proposed in this work that aims to expand the applicability of EMI technique in noisy inspections. These advanced signal processing techniques were based on the Hinkley statistical criterion, the cross-correlation function and the chromatic technique of data cluster. The results indicated that the newly proposed methods were effective in the diagnosis of structural failure, expanding the applicability of this promising structural monitoring technique for inspections where structural noise is not controlled, as in real applications (AU)

• FAPESP's process: 15/24903-5 - Structural Health Monitoring Systems Based on the Electromechanical Impedance in Applications with Low Signal-to-Noise Ratio
• Grantee: Bruno Albuquerque de Castro
• Support Opportunities: Scholarships in Brazil - Doctorate