论文部分内容阅读
非参数回归预测方法在交通流短时预测中得到了广泛应用.针对提高搜索速度和关键参数的优化设置两个问题,提出使用KD树作为模式库的存储结构,能够有效提高搜索速度,并且能够在实际运行中不断将新发现的交通流模式实时地加入模式库.提出使用遗传算法对非参数回归中的重要参数进行优化,实验表明能够得到相对较优的参数设置.所得研究结果为实时的交通流短时预测系统提供了一种较好的预测方法.
Nonparametric regression prediction method has been widely used in short-term traffic flow prediction.Aiming at improving the search speed and optimizing the setting of key parameters, it is proposed to use the KD tree as the storage structure of the pattern library, which can effectively improve the search speed, In the actual operation, the newly found traffic flow patterns are added to the pattern library in real time, and the optimization of the important parameters in the non-parametric regression is proposed by using genetic algorithm.Experimental results show that the relatively optimal parameter setting can be obtained.The results obtained in this paper are real-time Short-time traffic flow prediction system provides a better prediction method.