This paper presents a two-level neural network scheme for finite element (FE) model updating in which both the structural parameters and the damping ratios are updated.
As predicted, at the critical panel, we observed an ERP index of mental model updating in a late (400 900 ms) posterior positivity (Brouwer et al., 2012; Donchin & Coles, 1988).
To demonstrate the effectiveness of the proposed methods, FE models of truss structure and cantilever beam are used for numerical simulations of model updating in presence of random and systematic errors.
Although either static or dynamic measurements have been used for model updating in a damage identification procedure, when a generally valid and accurate model is sought, different types of measurements should be combined.
The approach does not use identification, operates without constraints between the number of sensors and the number of model degrees of freedom and differs from model updating in that only the damage distribution (and not the severity) enters the formulation.
It is the authors' hope that this work will prove to be of value, especially to those who are getting acquainted with the research base and aim to participate in the application of model updating in industry, where a pressing need exists.
