| In order to provide a better service for the early warning of low temperature disasters on road traffic and mitigate the damage caused by the frozen road against cars,This paper researches the low temperature of highway network in Jiangsu Province during the winter of 2013-2016,the results showed as follows:1.In the study on the temporal and spatial pattern of low temperature occurrence on the road surface in 4a winters,it was found:(1)The occurrence frequency of low temperature below O?,below-2 ? and below-5 ? on the road surface of the whole expressway network in Jiangsu Province showed a distributions of "higher in the north part and lower in the south part".The occurrence frequency of low temperature below 0? decreases from 70%in the north to 20%in the south,which bring the risks of iced road caused by low temperature in all sections of the whole province.(2)Among the 3 months of a winter,the highest occurrence frequency of low temperature appears in January,and the lowest in February.In 4a winters,the average temperature of the road surface showed a rising trend,and the occurrence frequency of 0? low temperature shows a downward trend accordingly.But the occurrence frequency of-2? and-5 ? low temperature decline from 2013 to 2015,and rise again in 2016.(3)In the 5 main road sections,the occurrence frequency of low temperature of the Shanghai-Nanjing Expressway is obviously lower than other road sections.Different road sections showed a similar average diurnal temperature variations in air,road surface and roadbed.The time when 0? appearance dispersed from 06 p.m.to 05 a.m.,but the time when 0? disappearance concentrated at 06-07 a.m..2.In the study on three typical cases of cold wave weathers,it was found:(1)The ranges of low temperature on roads caused by cold wave are different.Severe cold air transportation can bring road surface low temperature below-5 ? in whole Jiangsu Province.Although the road surface temperature in the south is usually higher compared with north,the variation magnitude also can be larger than it in the north.(2)In the 5 main road sections,the magnitude and duration of low temperature in HuNing Expressway caused by cold wave are the least.In the three weather cases,the stations M9355 in Ningsuxu Expressway,M9318 in Suhuaiyan Expressway and M9328 in Yanhai Expressway became the coldest twice during cold wave.(3)Although the macroscopic weather background caused by low temperature in three cases is cold air southward,the time variation and spatial difference of energy budget are important reasons for the spatial and temporal differences in pavement temperature in different road sections.3.Using multiple linear regression,Naive Bayes and support vector machines,statistical modeling and forecasting experiments for low-temperature warning of road surface were carried out.The research results show that:(1)In single station experiment of modeling and forecasting on M9308 at Jinghu Expressway,it is found that thermal variables related to road and relevant meteorological factors which shows significant correlation are both necessary to consider as forecasting independent variables.The independent variables scheme including temperature and its variation of air,road surface and roadbed,relative humidity and U component of wind speed bring the best forecast result.(2)With respect to the whole province's expressway network,the accuracy of the three types of statistical forecast models for the low-temperature prediction of road surface exceeds 75%.Comparing the experimental results of the cold weather forecast of the whole road network,it is found that the multiple linear regression method has the best forecasting effect on the northern road network of Jiangsu Province,and the accuracy of forecasting is mostly above 85%;The support vector machine(SVM)model has the best forecast results for southern Jiangsu Province.The accuracy of low-temperature forecasting for most stations is above 95%. |
PDF Full Text Request