Prediction of fatigue damage accumulation in metallic structures by the estimation of strains from operational vibrations

This work deals with the problem of estimating damage accumulation due to fatigue in the entire body of a metallic structure using operational vibration measurements collected from a limited number of sensors installed on a structure. A recently proposed joint input-state estimation filter is extended to estimate the strain response time histories in the entire body of the structure using the output-only vibration measurements. The estimates hold for any excitation with arbitrary temporal variation and spatial distribution over the structure. Such predictions are then integrated with damage accumulation models, S-N fatigue curves and rainflow stress cycle counting methods to estimate fatigue damage accumulation maps covering the entire body of the structure. The method is validated using simulated vibration measurements for a laboratory steel beam, generated from impulse and stochastic excitations