遗传算法优化支持向量机的交通流量预测

交通流量预测是智能交通系统中的关键技术,针对当前交通流量预测模型存在不足,提出一种遗传算法优化支持向量机的交通流量预测模型。首先收集交通流量历史数据,并基于混沌理想对其进行相空间重构,然后将训练样本输入到支持向量机中进行学习,并采用遗传算法优化支持向量机参数,建立交通流量预测模型,最后采用测试样本对模型的性能进行测试。结果表明,相对于经典交通流量预测模型,本文模型可以更加准确描述交通流量预测复杂的变化趋势,提高了交通流量的单步和多步预测精度。

Trafficflow prediction model based on support vector machine optimized by genetic algorithm

Traffic flow prediction is key techonogy in intelligent transportation systems, in order to sovle the de_fects of the current traffic flow prediction models, a novel traffic flow predictive model is proposed based on support vector machine optimized by genetic algorithm. Firstly, the historic data of traffic flow is collected and the data are re_constructed based on the chaotic theory, secondly, the training samples are input to support vector machines to learn whihc genetic algorithm is used to optimize the parameters of support vector machine to establish the prediction model of traffic flow, finally the performance is tested by sample. The results show that, compared with the classical predic_tion models of traffic flow, the propsed model can more accurately describe the complex change trend of traffic flow and improve the prediction precision fo single step and multi step traffic flow.