大雾能见度估计与预测  

Visibility Estimation and Prediction of Heavy Fog

作　　者：余豪阁 李健[1] 邹安娥 王勇[1] YU Hao-ge;LI Jian;ZOU An-e;WANG Yong(School of Media and Design,Hangzhou Dianzi University,Hangzhou 310018,China)

机构地区：[1]杭州电子科技大学人文艺术与数字媒体学院,浙江杭州310018

出　　处：《数学的实践与认识》2021年第23期240-253,共14页Mathematics in Practice and Theory

摘　　要：雾与霾的自然现象会显著降低能见度,对道路行车与飞机起降造成巨大的安全隐患,因此,根据观测数据对大雾环境中的能见度准确估计与预测的重要性不言而喻.首先,采用因子分析对七个气象因素的观测数据进行降维,并结合辐射雾形成的物理模型确定因子个数,进而拟合出多元非线性关系式;然后,基于暗通道先验与Lambert-Beer定律,结合高速公路监控图像中车道线的距离等信息,建立透射率、距离与消光系数的关系模型,进而求解能见度变化曲线;最后,使用ARIMA时间序列模型预测高速公路能见度达到150m的具体时刻.The natural phenomenon of fog and haze will significantly reduce the visibility,and cause huge safety hazards to road traffic and aircraft landing.Therefore,it is important to accurately estimate and predict the visibility in fog environment according to the observation data.Firstly,the dimension of seven meteorological factors is reduced by factor analysis,and the number of factors is determined by combining with the physical model of radiation fog formation,and then the nonlinear formula is fitted;Then,based on the Dark channel prior theory and Lambert-Beer law,combined with the distance of lane line in the expressway monitoring image,the relationship model among transmittance,distance and extinction coefficient is established,and then the visibility curve is solved;Finally,ARIMA time series model is used to predict the specific time when the visibility of Expressway reaches 150m.

关 键 词：能见度预测 因子分析 暗通道先验 消光系数 ARIMA 

分 类 号：P412.17[天文地球—大气科学及气象学]