摘要
为提高短时交通流的预测速度和精度,提出一种基于基因搜索算法的短时交通流预测模型,对所有相同连续时间段的历史交通流数据进行编码,按照编码值的大小顺序排列,建立短时交通流预测基因库,根据当前连续时间段的交通流数据编码,快速搜索基因库中最相似的染色体种群.计算该种群中偏差值最小的3条染色体,组合解码后对当前交通流状况做出预测.实验结果表明,该模型可通过调整连续时间段的交通流数据个数以及短时交通流预测基因库的大小来提高预测精度.
A short-term traffic flow prediction model based on the genetic searching algorithm was proposed in order to improve the prediction speed and accuracy of short-term traffic flow.Firstly,all the previous traffic flow dates in the fixed continuous period were encoded,and ordered by the size to develop the gene pool of the short-term traffic flow forecasting.Secondly,according to the code of current traffic flow dates,the most similar chromosome populations were quickly searched in the gene pool.Finally,three chromosomes with the minimum deviation in the population were calculated and combined to predict the current traffic flow conditions.The experiment results show that,the proposed method is faster and owns better precision.Moreover,the accuracy of the prediction can be improved by adjusting the quantity of the traffic flow data of the same continuous period and the size of the gene pool of the short-term traffic flow forecasting.
作者
李杰
贺莹莹
杨矿利
LI Jie;HE Yingying;YANG Kuangli(School of Electrical and Mechanical Engineering,Pingdingshan University,Pingdingshan,Henan 467036,China;Henan Pinggao Electric Co.,Ltd.,Pingdingshan,Henan 467001,China)
出处
《平顶山学院学报》
2023年第2期47-52,共6页
Journal of Pingdingshan University
关键词
智能交通系统
编码
基因搜索
偏差值计算
解码预测
intelligent traffic system
coding
genetic searching
deviation calculation
decoding prediction
