Abstract
This article is concerned with the identification of chaotic behavior from a time serious which is obtained by numerical simulation or experiment. The method of identifying chaos based on the results of continuous wavelet transform of the time history is proposed. With the results of the wavelet transform, the energy distribution is obtained. For a chaotic time history, the distribution will also show the chaotic behavior. With this idea, the chaos index is defined to describe this random like behavior numerically. Then observing the chaos index may allow us to identify the chaotic time history quantitatively. Numerical experiments are provided to demonstrate the proposed method.
