MODAL PARAMETER-IDENTIFICATION USING SIMULATED EVOLUTION

A global optimization algorithm based on the Darwinian evolutionary th eory is implemented for the identification of modal parameters of line ar vibrating structures. The capability of the evolutionary search met hod in locating the global minimum among numerous local minima is demo nstrated. The modal parameter identification procedure is based on the concept of modal sweep, according to which modes of a structure are i dentified individually and iteratively. A fast recursive algorithm is also developed for forward response calculation of a linear oscillator to lessen the computational burden of using simulated evolution in th e iterative identification process. The proposed modal parameter ident ification approach does not require any initial estimate of parameter values, Simulation results show that the approach is effective and rel iable in identifying dominant modes under noisy conditions.