ANALYZING THE NONLINEAR TIME SERIES OF TURBULENT FLOWS WITH KERNEL INTERVAL REGRESSION MACHINE
Abstract
The turbulent flow consists of coherent time- and space-organized vortical structures with a particular formation and instability cycle. Research has already shown that some dynamic systems and experimental models still cannot provide a good nonlinear analysis of turbulent time series because in the real turbulent flow there exist very complicated nonlinear behaviors affected by many vague factors. An approach of fuzzy piecewise regression analysis has been proposed to predict the nonlinear time series of turbulent flows. In this paper, we propose kernel interval regression machine for nonlinear time series analysis of turbulent flows. The proposed method applies the kernel trick to interval regression analysis in terms of the inner products of input vectors. In order to indicate the performance of this method, we present an example of predicting the near-wall turbulence time series as a verifiable model. Furthermore, we compare this method with fuzzy piecewise regression model. Experimental results show that the proposed method is very attractive for the turbulence time series in nonlinear analysis and in fuzzy environments.
