基于PN和CNN-LSTM-ATT的航班延误预测  

Flight Delay Prediction Based on Petri Net and CNN-LSTM-ATT

作　　者：吴涔 叶宁 王甦[1] 季翔宇 WU Cen;YE Ning;WANG Su;JI Xiang-yu(School of Computer Science,Nanjing University of Posts and Telecommunications,Nanjing 210023,China;Jiangsu High Technology Research Key Laboratory for Wireless Sensor Networks,Nanjing 210023,China)

机构地区：[1]南京邮电大学计算机学院、软件学院、网络空间安全学院,江苏南京210023 [2]江苏省无线传感网高技术研究重点实验室,江苏南京210023

出　　处：《计算机技术与发展》2023年第4期213-220,共8页Computer Technology and Development

基　　金：江苏省科技重点研发计划(社会发展)(BE2020713)。

摘　　要：航班延误预测对提高机场地面保障效率具有重要参考意义。针对目前航班地面保障流程复杂多变以及航班过站延误预测精度不高的问题,提出了一种基于Petri Net和融合预测模型CNN-LSTM-ATT的航班延误预测模型。首先,根据机场航班实际地面保障流程抽象构建离港航班地面保障作业Petri Net模型,获取保障流程中的关键作业时长成为动态特征;其次,将动态特征、航班信息、延误信息和天气信息输入CNN-LSTM-ATT模型中进行特征提取和分类预测,模型中引入注意力机制,通过注意力权重突出关键数据信息的影响,进一步挖掘重要特征之间的内部规律。实验结果显示,该融合模型准确率相比独立模型提升了6%,达到98.1%。通过对不同模型的对比表明该模型能较好地应对场面流程变化并且具备较好的延误预测能力。Flight delay prediction is an important reference to improve the efficiency of airport ground service.Nowadays,the flight ground service process is complex and changeable,and the prediction accuracy of flight transit delay is not high,thus a flight delay prediction model based on Petri net and fusion prediction model CNN-LSTM-ATT is designed.Firstly,the Petri Net model of the departing flights ground service process was abstractly constructed according to the actual ground service process of flights,and the critical service operation duration became the dynamic feature.Secondly,the dynamic feature,flight information,delay information and weather information are input into the CNN-LSTM-ATT model for feature extraction and classification prediction.The attention mechanism is introduced to highlight the influence of key data through attention weight and further mine the internal laws of important features.The results show that the fusion model improves the accuracy by 6%compared with the independent model,and the final accuracy of the experiment reaches 98.1%.The comparison of different models shows that the model can better cope with the changes of scene and has better delay prediction ability.

关 键 词：航班地面保障 保障流程分析 延误预测 Petri Net CNN-LSTM-ATT 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程]