基于样本熵的空中交通流复杂性分析

Complexity analysis of air traffic flow based on sample entropy

空中交通流具有复杂非线性特征,时间序列是研究空中交通流的有效手段。为了定量计算空中交通流的复杂程度,剖析空中交通流的复杂机理,进而为空中交通流的建模、预测和管控提供科学依据,首先阐述了样本熵和多尺度样本熵的计算方法,接着采集三亚01,02,04号扇区连续28天的实际运行数据,构建了空中交通流时间序列,然后基于样本熵计算了时间序列的复杂度,并分析了不同参数的影响。计算结果表明:02号扇区交通流最复杂,04号扇区次之,01号扇区复杂性最小;样本熵和多尺度样本熵可以定量分析空中交通流时间序列的复杂性,且对时间序列长度依赖性小。

Air traffic flow has complex nonlinear characteristics, and time series is an effective way to study air traffic flow. In order to quantify the complexity, analyze the complex mechanism and provide scientific basis for modeling, forecasting and controlling air traffic flow, the calculation method of sample entropy and multi-scale sample entropy is firstly introduced. Then, the actual operation data of No.1, No.2 and No.4 sectors of Sanya were collected consecutively for 28 days. Then, the complexity of time series was calculated based on sample entropy, and the effects of different parameters were analyzed. The calculation results showed that the air traffic flow of No.2 sector was the most complex, followed by No.4 sector, and the of No.2 sector had the least complexity. The calculation results indicate that the sample entropy and the multi-scale sample entropy could quantitatively analyze the time series complexity of air traffic flow, and have little dependence on the length of time series.