In this paper, the principle of parameter identification of structure nonlinear vibration model based on the complex Morlet wavelet transform was introduced firstly. The influence of complex Morlet wavelet transform on the identification accuracy of nonlinear model parameters was analyzed under the different cases of edge-effect of wavelet transform and less sampling points. Then a BP neural network was used for the prediction extension of nonlinear vibration signals and a novel parameter identification method was proposed based on the prediction results. Finally, through the numerical simulation of two nonlinear vibration models, the method was proved to be effective in the anti-noise property and identification of nonlinear model parameters.