基于GP-LVM和LS-SVM航班延误等级预测研究  

Study on flight delay grade prediction based on GP-LVM and LS-SVM

作　　者：陈超 Chen Chao(College of Aviation Engineering,Civil Aviation University of China,Tianjin 300300,China)

机构地区：[1]中国民航大学航空工程学院,天津300300

出　　处：《无线互联科技》2020年第6期10-11,14,共3页Wireless Internet Technology

摘　　要：针对气象条件导致航班延误难以准确预测的问题,文章提出基于GP-LVM和LS-SVM的航班延误等级预测算法。通过GP-LVM对经过预处理的气象数据非线性降维,得到影响航班延误的显著变量;对航班延误进行LS-SVM的延误等级预测。仿真结果表明,文章提出的组合预测方法相对于单一SVM模型,能够提高航班延误等级预测准确率。This paper establishes a combined flight delay prediction model based on GP-LVM and LS-SVM according to the difficulty in accurately predicting flight delays due to meteorological conditions.GP-LVM is used to reduce the dimension of preprocessed meteorological data nonlinearly,and the significant variables affecting flight delay are obtained.According to the classification verification results of LS-SVM,the next dimension reduction iteration was carried out.The simulation results showed that,the prediction accuracy of combined prediction method was improved compared with SVM.

关 键 词：航班延误 高斯过程隐变量模型 非线性降维 最小二乘支持向量机 

分 类 号：TP181[自动化与计算机技术—控制理论与控制工程] V35[自动化与计算机技术—控制科学与工程]