Numerical study of nonlinear interaction of the guided wave due to breathing type debonding in stiffened panel
The common tool for assessment of breathing-type debonding in metallic or composite structures is nonlinear guided wave-based technique. The past studies show that with debonding size, the strength of the nonlinearity does not exhibit strictly increasing or decreasing trends, or that the monocity is valid up to a certain size limit of debonding. This paper presents the study of non-linear interaction of guided waves in the debonding interface of a metallic stiffened panel. The study attempts to establish a relationship between the contact energy generated by the contact acoustic nonlinearity (CAN) at the debonding interface and the associated nonlinearity strength for various debonding sizes at various excitation frequencies. A numerical model of the stiffened panels is developed in three-dimensional finite element (FE) and validated with experiments for the study of interaction of nonlinear guided waves. The validated FE model is used to conduct studies on nonlinear interactions in debonding. The outcome of this study contributes to a better understanding of how guided waves can be used to effectively assess the debonding in metallic stiffened panels by considering non-linear interactions at the debonding interface. The study also provides insights into a more accurate and consistent quantification of the debonding using higher harmonic signals and contact energy produced by non-linear interactions.
Transforming Simulated Data into Experimental Data Using Deep Learning for Vibration-Based Structural Health Monitoring
While machine learning (ML) has been quite successful in the field of structural health monitoring (SHM), its practical implementation has been limited. This is because ML model training requires data containing a variety of distinct instances of damage captured from a real structure and the experimental generation of such data is challenging. One way to tackle this issue is by generating training data through numerical simulations. However, simulated data cannot capture the bias and variance of experimental uncertainty. To overcome this problem, this work proposes a deep-learning-based domain transformation method for transforming simulated data to the experimental domain. Use of this technique has been demonstrated for debonding location and size predictions of stiffened panels using a vibration-based method. The results are satisfactory for both debonding location and size prediction. This domain transformation method can be used in any field in which experimental data for training machine-learning models is scarce.