各種情報量を用いた混合層の層流―乱流遷移過程の解析

In the laminar-turbulent transition process of a mixing layer formed downstream of a two-dimensional nozzle exit, an analysis was performed based on various information measures. Shannon entropy, permutation entropy and Kullback-Leibler divergence were introduced, and former studies in which they were used in the turbulent analysis were then reviewed. In the present study, the probability distribution of time series of hot-wire output voltage data was obtained, then analyzed. The aim of the investigation was to clarify the effectiveness of the analysis for the transition and turbulent flow. In addition, equations which Shannon entropy must satisfy in the turbulent flow field were derived. The Shannon entropy of the fluctuating velocity changed monotonically in the downstream direction. Thus, it appears to measure the transition process in the mixing layer. The permutation entropy of the fluctuating velocity first increased, then decreased, then increased again, and decreased finally. It reflected the increase of the fluctuating velocity and change of fluctuation manner (from periodic to irregular fluctuation) during the transition process. The Kullback-Leibler divergence based on the probability density function of the fluctuating velocity increased first, then decreased downstream, and thus did not show a monotonic change during the transition process in the mixing layer.