基于贝叶斯估计的短时空域扇区交通流量预测

发布时间：2018-04-05 17:12

本文选题：空中交通管制　切入点：短时流量预测　出处：《西南交通大学学报》2016年04期

【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.
[Abstract]:In order to accurately understand the spatial sector traffic distribution situation and future trends, a short-term spatial sector traffic flow prediction method based on Bayesian estimation is proposed.Firstly, by analyzing the original radar data of aircraft in the airspace system, extracting the historical operation information of each sector, the multi-sector aggregate traffic flow model is established. Secondly, the Bayesian estimation theory is used to estimate and dynamically update the parameters of the model.The future evolution trend and uncertain range of traffic flow in airspace sector are predicted. Finally, five typical busy sectors in China are selected as examples, 5 min as time period and 1 hour as prediction range to verify the proposed prediction method.The results show that the absolute error of the traffic flow forecast results is less than 3 vehicles and the average absolute error is less than 2 vehicles during the period of more than 85%, and the stability of the forecast results is good.It can fully reflect the dynamics and uncertainty of short-term traffic flow between different spatial sectors, which is in line with the actual situation of air traffic.
【作者单位】：南京航空航天大学民航学院;
【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)
【分类号】：V355

【相似文献】