A general substructure-based framework for input-state estimation using limited output measurements

This paper presents a general framework for estimating the state and unknown inputs at level of system subdomain using limited number output measurements, enabling thus component-based vibration monitoring or control providing novel approach to model updating hybrid testing applications. Under premise that dynamics are driven by (i) externally applied (ii) interface forces, with latter representing unmodeled components, problem output-only response prediction substructure can be tailored Bayesian input-state estimation context. As such, solution is recursively obtained fusing Reduced Order Model (ROM) structural interest available measurements via filter. The proposed without loss generality established on basis fixed- free-interface domain decomposition methods verified means three simulated Wind Turbine (WT) structure applications increasing complexity. performance assessed in terms achieved accuracy estimated quantities.