基于贝叶斯估计的短时空域扇区交通流量预测  

Short-Term Traffic Flow Prediction of Airspace Sectors Based on Bayesian Estimation Theory

作　　者：陈丹[1] 胡明华[1] 张洪海[1] 尹嘉男[1] 

机构地区：[1]南京航空航天大学民航学院,江苏南京211106

出　　处：《西南交通大学学报》2016年第4期807-814,共8页Journal of Southwest Jiaotong University

基　　金：国家自然科学基金资助项目(U1333202);国家科技重大支撑计划资助项目(2011BAH24B08);江苏省普通高校研究生科研创新计划资助项目(KYLX_0290)

摘　　要：为准确把握空域扇区流量分布态势及未来变化趋势,提出了一种基于贝叶斯估计的短时空域扇区交通流量预测方法.首先,通过解析空域系统内航空器原始雷达数据,提取各扇区历史运行信息,建立了多扇区聚合交通流模型;其次,采用贝叶斯估计理论对模型参数进行最优估计和动态更新,预测了空域扇区交通流量的未来演变趋势及其不确定范围;最后,选取国内5个典型繁忙扇区为例,以5 min为时间段,以未来1 h为预测范围,对所提预测方法进行了验证.研究结果表明:85%以上时段交通流量预测结果的绝对误差在3架以内,平均绝对误差均在2架次以内,预测结果的稳定性较好,可充分反映各空域扇区之间短时交通流的动态性和不确定性,符合空中交通的实际情况.To accurately forecast the air traffic flow distribution in airspace sectors and its development trend in the future, a short-term traffic flow prediction method based on Bayesian estimation theory is proposed. First, the operational history data of various sectors in the airspace system are extracted by parsing raw radar data of the aircraft within the airspace system. On this basis, an aggregate multi- sector traffic flow model is established. Then, Bayesian estimation theory is adopted to predict the future trend of airspace sector traffic flow and its uncertainty intervals by estimating and updating the optimal parameter of the aggregate muhi-sector traffic flow model dynamically. Finally, the proposed method is verified on a set of operational history data of five air route sectors, taking 5 vain as one time step to predict the short-term air traffic flow in the next one hour. The results show that the absolute error of the predicted results of more than 85% time steps is less than 3, the average absolute error is less than 2, and the stability of the predicted results is well. The proposed method can adequately reflect the dynamics and uncertainty in the airspace system operation, and hence is well in line with the practice.

关 键 词：空中交通管制 短时流量预测 多扇区 贝叶斯估计 不确定性 雷达数据 

分 类 号：V355.2[航空宇航科学与技术—人机与环境工程]