基于子空间方法的航站楼安检队列状态预测  

Prediction of Security Queue Status in Terminal Building Based on Subspace Method

作　　者：秦倩 丁新伟 邢志伟 凌若鸿 QIN Qian;DING Xin-wei;XING Zhi-wei;LING Ruo-hong(Department of International Science and Technology,Capital Airport Group Co.,Ltd,Beijing 100621,China;School of Electronic Information and Automation,Civil Aviation University of China,Tianjin 300300,China)

机构地区：[1]首都机场集团有限公司国际科技部,北京100621 [2]中国民航大学电子信息与自动化学院,天津300300

出　　处：《计算机仿真》2023年第8期72-77,482,共7页Computer Simulation

基　　金：国家重点研发计划项目(2018YFB1601200)。

摘　　要：针对大型繁忙机场旅客安检服务效率低的问题,结合航班时刻及离港旅客属性,提出基于子空间辨识的航站楼安检队列状态预测模型。模型结合安检场景特点,在子空间辨识方法基础上,通过训练大量历史数据设定安检队列初始状态,并根据机场繁忙度与旅客流量特征对安检数据进行分组,对分组数据分别辨识获取不同特征下的系统矩阵,构建多特征模型。使用国内某大型机场安检数据对模型进行检验,结果表明模型在误差允许范围内的预测准确率较高,且有较好的稳定性。Aiming at the problem of low efficiency of passenger security inspection services in large busy airports,an optimized subspace identification method is used to propose a terminal security queue state prediction model based on subspace identification.The model combines the characteristics of the security check scene,and based on the subspace identification method,the initial state of the security check queue is set through training a large amount of historical data.And according to the airport busyness and passenger flow characteristics,the security inspection data is grouped,and the grouped data is identified to obtain the system matrix under different characteristics to construct a multi-feature model.The model was verified with the security inspection data of a large domestic airport,and the results show that the prediction accuracy of the model within the allowable error range is high,and it has good stability.

关 键 词：机场旅客安检 预测模型 安检排队 子空间辨识 队列状态 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]