Comparison of complexity between indoor and outdoor wind speed time series

Quantitatively characterizing wind variation series in different environments is an important problem with significant engineering and industrial applications. We systematically carry out indoor and outdoor experiments separately to record two groups of wind speed time series. Using the Adaptive Optimal Kernel Time-Frequency Representation (AOK TFR) we first investigate the fluctuating behaviors from an energy and frequency point of view. Then we employ two state-of-art complexity measures, i.e., Lempel-Ziv complexity and Permutation Entropy, to discriminate two groups of wind speed time series. The results suggest that due to intrinsic limitations of the AOK TFR, it is difficult to distinguish the different wind field environments. Moreover, the results of the combined method of Lempel-Ziv complexity and Permutation Entropy are sensitive to changes of different environments, which point out to environment-dependent properties of the wind speed signals. The interesting and significant finding shed light on a new perspective which contributes to a better understanding of the complex dynamic behaviors of wind phenomenon.