摘要
针对航班延误后航班运行态势状况预测的问题,提出一种融合动态关联规则的多级输入层人工神经网络预测模型,将通过动态关联规则产生的强关联关系融合进多级输入层人工神经网络中,用于处理天气预报,达到对航班运行态势实时在线预测的目的。通过实验对比,该模型可以用于处理海量数据和标称变量,相比其它模型,该模型在预测准确率、内存占用率、运行效率以及召回率等性能的表现上都有较大提升。
Aiming at the problem of flight situation prediction after flight delay,a multi-level input layer artificial neural network prediction model integrating dynamic association rules was proposed.The strong association relationship generated by dynamic association rules was integrated into the multi-level input layer artificial neural network and it was used to process weather forecast to achieve the purpose of real-time online prediction of flight operation situation.Through experimental comparison,the model can not only be used to deal with massive data and nominal variables,but has a great improvement in predicting accuracy,memory occupancy,operation efficiency and recall rate compared with other models.
作者
丁建立
王辰
DING Jian-li;WANG Chen(College of Computer Science and Technology,Civil Aviation University of China,Tianjin 300300,China)
出处
《计算机工程与设计》
北大核心
2023年第4期1097-1103,共7页
Computer Engineering and Design
基金
国家自然科学重点基金项目(U2033205)。
关键词
融合
关联规则
多级输入
神经网络
航班态势
航班延误
预测模型
integration
association rules
multi level input
neural network
flight situation
flight delay
prediction model