No abstract

No abstract

No abstract

System identification is primarily studied for unidirectional excitation using the Bouc-Wen model, neglecting the torsional coupling, even though real structure experiences multidirectional seismic excitation. Moreover, the high damping rubber bearings exhibit bidirectional effects, thereby requiring coupled biaxial Bouc-Wen (BBW) model and demand the estimation of model parameters for structural health monitoring. The current work presents three numerical case studies followed by experimental validation to demonstrate the applicability and efficacy of Bayesian filters named constraint unscented Kalman filter (CUKF) in identifying model parameters for the nondeteriorating system as well as deteriorating systems. With limited measurements and increased states, a two-stage framework of the CUKF is used to enhance the performance in identifying the hysteresis parameters and system dynamics of the nondeteriorating systems. For the deteriorating system, the Paris-Erdogan law is coupled with the stiffness in the BBW model to introduce degradation as per the acceleration fatigue crack growth. The degradation parameters and deteriorating stiffness is captured through CUKF accurately.The application of CUKF to the experimental responses proves the robustness of the algorithm for coupled biaxial hysteresis system. Additionally, a unified structural health monitoring (SHM) framework is proposed for condition monitoring during extreme events and long-term periodic maintenance through ambient vibrations. Overall, the result concludes that CUKF is a reliable Bayesian estimator for coupled biaxial hysteresis systems and demonstrates promising potential in identifying fatigue-induced deterioration.

The implementation of piezoelectric sensors is degraded due to surface defects, delamination, and extreme weathering conditions, to mention a few. The sensor needs to be diagnosed before the efficacious implementation in the structural health monitoring (SHM) framework. A novel experimental method based on Coulomb coupling is utilised to visualise the evolution of elastic waves and interaction with the surface anomaly in the Lead Zirconate Titanate (PZT) substrate. Machine learning is expeditiously becoming an essential technology for scientific computing, with several possibilities to advance the field of structural health monitoring (SHM). This study employs a deep learning-based autoencoder neural network in conjunction with image registration and PSNR to diagnose the surface anomaly in the PZT substrate. The autoencoder extracts the significant damage-sensitive features from the complex waveform big data. The autoencoder provides a non-linear input-output model that is well suited for the non-linear interaction of the wave with the surface anomaly and boundary of the substrate. The mean absolute error (MAE) in the reconstruction of the sequential signal from the autoencoder network is evaluated. The MAEs are sensitive to the anomaly that lies in the PZT substrate. Image registration is leveraged to overcome the challenge due to the offset in the data. Further, the localisation and quantification of the anomaly are performed by computing PSNR values. This work proposes an advanced, efficient damage detection algorithm in the scenario of big data that is ubiquitous in SHM.