ABSTRACT
The Modal Assurance Criterion (MAC) is the most widely used indicator to correlate mode shapes. However, it is recognized that the single index nature of the MAC is insufficient to detect localized differences in mode shapes. Mode shapes of structures made from Glass fibre-reinforced polymer (GFRP) can exhibit localized features. Consequently, model updating process of GFRP structures may not be optimal with the MAC. This paper proposes a shape-descriptor–based model updating procedure using Tchebichef Moment Descriptors (TMDs) for improved model updating of GFRP structures. Shape descriptors are advantageous in describing mode shape features at a range of fidelities, and so are well suited for use in model updating procedures. The proposed model updating framework is implemented on a benchmark GFRP footbridge which exhibit localized differences in mode shapes. The method is validated using a simulated model updating of the footbridge. A combination of shape descriptors and natural frequencies for the updating objective yield the most improved numerical model. The updating results suggest that the proposed method is an improvement for model updating of GFRP structures.
