跑道积冰特征因素提取与灰色神经网络预测  

Feature Factor Extraction and Grey Neural Network Prediction Method for Runway Icing

作　　者：陈斌 周冲[1,2] 刘悦 CHEN Bin;ZHOU Chong;LIU Yue(College of Electronic Information and Automation,Civil Aviation University of China,Tianjin 300300,China;Department of Aviation Ground Special Equipment,Civil Aviation University of China,Tianjin 300300,China)

机构地区：[1]中国民航大学电子信息与自动化学院,天津300300 [2]中国民航大学航空地面特种设备研究基地,天津300300

出　　处：《公路工程》2022年第4期192-198,共7页Highway Engineering

基　　金：国家自然科学基金委员会-中国民航局民航联合研究基金资助项目(U1933107)。

摘　　要：跑道积冰预测是机场冬季安全、高效运行的重要保障。针对机场跑道积冰影响因素多且参数之间耦合的复杂问题,采用主成分分析法(Principal Component Analysis,PCA)提取影响积冰厚度的特征因素。通过灰色模型对跑道积冰进行初步预测,然后采用粒子群优化的RBF模型对灰色模型进行误差修正,构建灰色神经网络预测模型。将机场采集的积冰数据代入预测模型进行实例验证,结果表明:灰色神经网络组合预测模型效果优于单一预测模型,相对预测误差保持在5%以内,是一种有效可行的跑道积冰预测方法,能够为冰雪天气下机场除冰设备部署提供科学依据。Aiming at the complex problem of airport runway icing with many factors and the coupling of parameters,principal component analysis(PCA)is used to extract the characteristic factors affecting icing.Preliminary prediction of runway icing was made by the gray model,and then the RBF model optimized by particle swarm was used to correct the error of the gray model,and the gray neural network prediction model was constructed.Substituting icing data into the prediction model for instance verification,the results show that the gray neural network combination prediction model is better than a single prediction model,and the relative prediction error is kept within 5%.It is an effective and feasible runway icing prediction method,the deployment of airport deicing equipment under ice and snow provides a scientific basis.

关 键 词：机场跑道 积冰预测 主成分分析 灰色模型 神经网络 组合模型 

分 类 号：V328[航空宇航科学与技术—人机与环境工程]