An innovative data-driven probabilistic approach for damage detection in Electromechanical Impedance Technique. (1st September 2022)

• Record Type: Journal Article
• Authors: Singh, Shishir K.; Malinowski, Pawel H.
• Is Part Of: Composite structures. Volume 295(2022)
• Publication Date: 2022-09-01
• Subjects: Electromechanical impedance -- Piezo-actuator -- Data fusion -- Probability density function -- PDF -- Guassian function; Composite construction -- Periodicals; Composites -- Périodiques; 624.18
• DOI: 10.1016/j.compstruct.2022.115808
• Languages: English
• ISSNs: 0263-8223

Abstract

Abstract: The electromechanical impedance method (EMI) has shown its damage detection capacity under hazardous environments. Moreover, it is cost-effective, because it involves a bonded piezoelectric transducer (PZT). But the daunting task in the EMI method is the selection of robust and effective damage index, frequency ranges, and combining the information for damage quantification. In this paper, an innovative standard deviation approach is used for the selection of effective frequency ranges. The novice nature of frequency range selection is based on the difference between healthy and damaged state data. Further, a data fusion technique is introduced for damage detection and classification using analytical and experimental data in this effective frequency range. This paper presents four samples of metal and composite with different damage scenarios to demonstrate the applicability of the method. The paper investigates damage in thin aluminium (Al) plate with holes using the EMI method. Further, simulated mass damage and impact damage study to the glass fiber reinforced polymer (GFRP) composite plate. A combined C-index statistical data-driven damage matrices are calculated and further compared with the RMSD index approach. The method looks suitable for identifying the damage location and damage severity simultaneously and is more effective for the less severe damage cases.