Smart monitoring of aeronautical composites plates based on electromechanical impedance measurements and artificial neural networks

Abstract

This paper presents a structural health monitoring (SHM) method for in situ damage detection and localization in carbon fiber reinforced plates (CFRPs). The detection is achieved using the electromechanical impedance (EMI) technique employing piezoelectric transducers as high-frequency modal sensors. Numerical simulations based on the finite element method are carried out so as to simulate more than a hundred damage scenarios. Damage metrics are then used to quantify and detect changes between the electromechanical impedance spectrum of a pristine and damaged structure. The localization process relies on artificial neural networks (ANNs) whose inputs are derived from a principal component analysis of the damage metrics. It is shown that the resulting ANN can be used as a tool to predict the in-plane position of a single damage in a laminated composite plate.

Kappa-PSO-FAN based method for damage identification on composite structural health monitoring

“…PZTs PNN SFAN FAN Kappa-PSO-FAN P1 PZT #1 96% 98% 98% 100% PZT #4 62% 82% 82% 84% P2 PZT #1 72% 82% 82% 86% PZT #2 68% 76% 76% 84% P3 PZT #3 80% 100% 100% 100% PZT #4 60% 84% 84% 84% P4 PZT #3 64% 40% 40% 60% PZT #4 78% 68% 68% 88% P5 PZT #3 48% 68% 68% 68% PZT #4 72% 70% 74% 75% Investigating the results presented in Table 4 , it can be observed that the PNN based method presented the worst performance by taking into account the success rate for identifying structural damage. It is important to mention that PNN based methods have been widely recurrent in the SHM literature ( De Oliveira & Inman, 2017;Na & Lee, 2013;Palomino et al, 2014;Selva et al, 2013 ). As observed, results for SFAN and FAN have presented practically the same performance for the overall success rate when working in the same conditions.…”

“…One reason for the increasing effort s in the related research is that those methods can be applied to different types of structures and several different damage scenarios. For example, methods based on PNN, applied to damage identification in SHM, are addressed in Na and Lee (2013), Palomino, Steffen, and Finzi Neto (2014) and Selva, Cherrier, Bundinger, Lachaud, and Morlierb (2013) . According to Na and Lee (2013) , methods based on PNN present a faster training procedure and they are easier to implement than the traditional Back-Propagation based methods.…”

Improved stochastic methods for modelling imperfections for buckling analysis of composite cylindrical shells

“…Unstiffened composite shells are used in a wide range of critical engineering structures [1][2][3]. These shells are prone to buckling and are highly sensitive to imperfections which arise during the manufacturing process [4][5][6][7][8].…”

Research on FBG-Based CFRP Structural Damage Identification Using BP Neural Network

“…The neural network [10,11] is a potentially powerful algorithm because of the ability to learn from past experiences and memorize the patterns in the form of an associative memory, and the special attention has been drawn to damage recognition and classification. Li et al [12] presented a thorough investigation into a vibration-based damage identification method utilizing dimensionally reduced residual frequency response function data in combination with neural networks to identify locations and severities of damage in numerical and experimental beam structures.…”