Study on fluctuation characteristics of 3D high-frequency wind speed time sequence has an important value for revealing the laws of complex wind field flow and evolution. Multifractal detrended fluctuation analysis（ MF—DFA） is applied to analyze fluctuation characteristics of 3D wind speed time sequence measured by highproperty ultrasonic anemometer. Research shows that both horizontal and vertical wind signals exhibit multifractal features

but have different fluctuation structures. Vertical fluctuation structures are much more complex than the horizontal ones. The multifractality in horizontal wind signals is due to the long-range correlations rather than the probability density. While long-range correlations and the probability density both affect on multifractality in vertical wind signal.