基于GMM聚类的AM-BiLSTM机场安检旅客流量预测  

GMM clustering-based AM-BiLSTM airport security inspection passenger flow prediction

作　　者：李国[1] 钱梦飞 LI Guo;QIAN Mengfei(College of Computer Science and Technology,Civil Aviation University of China,Tianjin 300300,China)

机构地区：[1]中国民航大学计算机科学与技术学院,天津300300

出　　处：《传感器与微系统》2024年第9期11-14,18,共5页Transducer and Microsystem Technologies

基　　金：国家自然科学基金民航联合基金重点资助项目(U2033205)。

摘　　要：针对现有安检旅客流量预测研究大多为正常情况下的预测,未考虑异常突发情况下安检旅客流量的变化趋势,提出一种基于高斯混合模型(GMM)聚类的融合注意力机制的多变量双向长短期记忆(AM-BiLSTM)机场安检旅客流量预测模型。首先,利用GMM聚类算法对原始数据集使用日期特征和延误特征分别进行聚类分析,根据聚类所得的不同日安检旅客流量场景构建不同的AM-BiLSTM旅客流量预测模型。实验结果表明:与现有多种预测方法相比,该方法在不同场景下均能准确预测各时段的安检旅客流量。Aiming at the fact that most of the existing security inspection passenger flow prediction studies are in normal conditions,without considering the change trend of security inspection passenger flow in abnormal emergencies,a Gaussian mixture model(GMM)clustering based AM-BiLSTM airport security inspection passenger flow prediction model is proposed.Firstly,GMM clustering algorithm is used for clustering and analysis on the usage date characteristics and delay characteristics of the original dataset,and different AM-BiLSTM passenger flow prediction models are constructed according to the different daily security inspection passenger flow scenarios obtained by clustering.The experimental results show that,compared with the existing several prediction methods,this method can accurately predict the security check passenger flow in different scenarios.

关 键 词：安检旅客流量 高斯混合模型聚类 长短期记忆网络 

分 类 号：TP39[自动化与计算机技术—计算机应用技术]