Detecting nonlinearity in concrete beams using bi-coherence of vibration data

Abstract

Vibration characteristics of structures have gained recent popularity in detecting damage of structures. Most of these works have been based on the assumption that the vibrating system behaves linearly. The aim of the present work is to explore possibility of identifying change in the nonlinear vibration characteristics to detect damage in RC beams. To understand the nonlinearities in the vibration characteristics, a time series data based model is discussed which is based on the bicoherence of vibration data. The model is capable of identifying the extent of nonlinearity in any time series and it also proposes some indexes to measure the extent of nonlinearity in vibrating time series. Static loads were applied to RC beams to induce damage and vibrating time series data at various damage levels were recorded. Time series data were recorded for four different vibrating conditions, ambient, one impact, interval impact and continuous impact excitations. These time series data were used to calculate the indexes proposed by the time series based model. These indexes were used to develop some relations with the increasing nonlinearity due to damage loading in RC beams. After that frequency domain data of time series were further studied to know the extent of cracks in RC beams due to the loading. The beam tests demonstrated that there was a change in nonlinear vibration behavior with damage. The change in frequency domain data and corresponding area is more prominent for flexure control beams. In case of shear control beams we can identify the propagation of crack from the change of frequency domain data of time series.