Parameters Optimization of Wavelet Transform in Modal Parameter Identification with Closely Spaced Modes

In recent years, the identification of modal parameters with closely spaced modes based on wavelet transform has been widely applied in the research area of structural health monitoring. But the parameters of wavelet, i.e. center frequency and its bandwidth have great effect on the results of identification. If the parameters of wavelet are inappropriate, the closely spaced modes could not be identified. In order to optimize the parameters of wavelet, a method based on adaptive genetic algorithm is proposed in this paper. The center frequency and its bandwidth are optimized with the adaptive genetic algorithm, whose objective function is the standard deviation between wavelet-ridge and fitting line. The results of numerical simulation with three closely spaced modes show that, the method proposed in this paper could optimize the wavelet parameters and the closely spaced modes could be identified by the wavelet transform with adaptive genetic algorithm.