基于自回归差分滑动平均的云南高速公路冬季路温短临预测模型研究  

Study on Short-impending Prediction Model for Winter Pavement Temperature of Yunnan Expressway Based on Autoregressive Integrated Moving Average

作　　者：丁宇超 苏宇 姜波 熊昌安 DING Yu-chao;SU Yu;JIANG Bo;XIONG Chang-an(National Engineering Laboratory for Surface Transportation Weather Impacts Prevention,Broadvision Engineering Consultants,Kunming 650200,China;Key Laboratory of Transportation Meteorology,CMA,Nanjing 210008,China)

机构地区：[1]云南省交通规划设计研究院有限公司陆地交通气象灾害防治技术国家工程实验室,昆明650200 [2]中国气象局交通气象重点实验室,南京210008

出　　处：《价值工程》2023年第10期118-120,共3页Value Engineering

基　　金：云南省交通运输厅科技项目(2019303、云交科教便[2021]90-2);南京气象科技创新研究院北极阁开放研究基金(BJG202101)。

摘　　要：文章基于云南麻昭高速公路沿线分钟级交通气象站数据,应用自回归差分滑动平均(Autoregressive Integrated Moving Average,ARIMA)方法建立了冬季路温短临预测模型,可对未来3h路温变化进行分钟级预测,并验证了不同路面状态背景下模型预测准确度。结果表明:模型对未来第1h的路温预测效果最好,在1.0℃误差范围内,模型总体预测准确率为89.22%,尤其对于特定路面状况背景有较好的预测能力,在路面结冰状态下,未来1h在1.0℃误差范围内,预测准确率可达94.42%。模型建模容易,输入参数易于获取,便于工程应用,模型可用于路面结冰短时临近预测,以及时发现潜在结冰点,为管理部门提前部署路面结冰养护处置工作提供支持。Based on the minute-level traffic meteorological monitoring station data along the Ma-Zhao Expressway in Yunnan Province,this article applies the Autoregressive Integrated Moving Average(ARIMA)method to develop a short-impending prediction model for winter pavement temperature,which can conduct minute-level prediction of pavement temperature changes in the next 3 hours.The prediction accuracy of the model was further verified under different pavement condition backgrounds.The results show that the model is the best in predicting pavement temperature changes in the first hour of the future,with an overall prediction accuracy of 89.22%within an error range of 1.0°C.In particular,the model has better prediction capability for specific pavement condition backgrounds,and the prediction accuracy can reach 94.42%within an error range of 1.0°C in the first hour of the future under icing pavement conditions.The model input parameters are convenient to obtain for engineering applications.The model can be used for short term proximity prediction of pavement icing and timely detection of potential icing points to provide support for management departments to deploy preparation work against road icing in advance.

关 键 词：ARIMA 路温 路面结冰 预测模型 

分 类 号：U416.1[交通运输工程—道路与铁道工程]