摘要
针对机场跑道道面积冰带来的航班起飞降落困难的问题,研究了跑道地温对道面温度的影响,建立了多变量多项式和BP神经网络相结合的道面积冰预测模型,采用BP神经网络的高度非线性逼近的方法预测未来半小时的道面积冰厚度和积冰概率。在道面积冰环境模拟实验平台上采集大量实验数据,对该模型进行训练和仿真并与实验结果对比,结果表明所建的考虑地温的道面积冰预测模型准确、可靠,能够在小雨、中雨以及大雨和不同风速等条件下对道面积冰厚度和概率进行准确预测,对冬季机场跑道道面积冰预测具有一定的参考价值。
In the light of difficulties in taking off and landing on flights caused by icing on airport runway,the effect of ground temperature at different depths on pavement temperature was studied,and a combined model for multivariable polynomials and BP neural networks was established.The highly nonlinear approximation ability of BP neural network was used to obtain the combined model which can predict the ice thickness and the probability of ice accretion in the next half hour.Meanwhile,a large number of experimental data were collected on the experimental platform for simulating pavement icing environments.And the model was trained and simulated based on the experimental data,and compared with the experimental results.The results indicate that the combined model considering ground temperature is more accurate and reliable,and can accurately predict the ice thickness and probability under the conditions of light rain,moderate rain,heavy rain and different wind speeds,which has a certain reference value for ice accretion prediction of airport runway in winter.
作者
王立文
高德伟
陈斌
张洋
WANG Li-wen;GAO De-wei;CHEN Bin;ZHANG Yang(Ground Support Equipment Research Base,Civil Aviation University of China,Tianjin 300300,China;Institute of Electronic Information and Automation of Civil Aviation University of China,Tianjin 300300,China)
出处
《计算机仿真》
北大核心
2019年第12期88-92,共5页
Computer Simulation
基金
国家科技部支撑计划项目(2012BAG04B02)
中央高校基本科研业务费中国民航大学专项基金项目(3122016B004)
关键词
机场跑道道面
组合模型
地温
多变量多项式
模拟实验平台
Airport pavement
Combined model
Ground temperature
Multivariate polynomial
Simulating experimental platform
作者简介
王立文(1962-),男(满族),河北省平泉人,教授,博士研究生导师,主要研究领域为民航装备与系统;高德伟(19744-),男(汉族),山西省应县人,硕士研究生,主研究领域为控制科学与工程;陈斌(1974-),男(汉族),陕西省凤翔人,副研究员,硕士研究生导师,主要研究领域为机场地面运行控制与保障技术;张洋(1991-),男(汉族),河南省信阳人,硕士研究生,主要研究领域为控制工程。
