摘要
道路结冰是一种导致危险驾驶条件并严重影响道路交通的恶劣天气现象,一种遗传微粒群算法支持向量机集成道路结冰预测系统被提出,提出改进的遗传微粒群算法用于自适应选择合适的气象因子并优化支持向量机参数,提高预报模型的准确率,采用此系统基于武汉市和十堰市1 980年至2006年期间多个时间段的历史气象数据,武汉市和十堰市的道路结冰预报模型被分别建立,利用两市2007年至2008年实例数据对所建立模型分别进行验证,结果证实了集成系统的可行性和有效性.
The road icing is an adverse weather condition leads to dangerous driving conditions with consequential effects on road transportation. A numerical road icing predication approach is employed for automatic prediction of road icing conditions. The approach is derived from the support vector machine （SVM）. To improve the classification accuracy for road icing prediction, a modified genetic particle swarm optimization （GP- SO） is employed to simultaneously select features and optimize the SVM parameters. With the data from ]980 to 2006, using the proposed approach, the road icing models for Wuhan City and Shiyan City are created, which have been tested by the data of the both cities from 2007 to 2008. The results have shown feasibility and effectiveness of the forecast approach.
出处
《华中师范大学学报（自然科学版）》
CAS
CSCD
北大核心
2010年第3期392-396,共5页
Journal of Central China Normal University：Natural Sciences
基金
2009年度湖北省高等学校优秀中青年团队计划项目(T200904)
关键词
道路结冰
支持向量机
微粒群算法
road icing
support vector machine
particle swarm optimization
