Finite element modelling and simulation of guided wave propagation in steel structural members

In this thesis, a guided wave-based active structural health monitoring (SHM) system using PZT actuator/sensor network is developed for non-destructive defect identification for steel structural members, including steel pipes and plates. The development is conducted through analytical characterisation of guided wave propagation, wave mode selection, numerical modelling and simulations of guided wave propagation in plates and pipes containing defects in the form of notch, hole, and crack. Different finite element models for the elastic wave propagation in steel structures are developed to determine their characteristics and used to optimise wave mode and frequency for defect identification. In order to capture the electromechanical behaviours of the piezoelectric actuators of interests, exact models and effective models with/without adhesive layer are developed. Considering the shape and size effects of PZT actuators the finite element models are devised to determine the discrepancy when different PZT actuators used. Through employing different dynamic analysis techniques, three finite element modelling methods are developed and applied into the finite element modelling and simulations and they are Explicit Dynamic Analysis (EDA), Implicit Dynamic Analysis (IDA) and Combined Implicit-Explicit Dynamic Analysis (CIEDA). The numerical results show EDA and CIEDA both perform well in simulating the wave propagation in structures. The developed PZT models, finite element modelling and dynamic analysis techniques are further employed in two typical structural members – plates and pipes with/without defects and as a result, the characteristics of guided wave propagation in the structural members with/without defects are determined numerically which pave the way to design such a guided wave-based active structural health monitoring (SHM) system for steel plates and pipes