摘要
通过建立遗传神经网络的模型 ,综合考虑各种影响因素 ,对道路交通量作较准确的预测 ,实验比较发现 ,预测结果优于常规时间序列模型ARMA的预测结果 ,是一种智能化程度较高的预测方法 ,该方法具有很强的学习能力和自适应性 。
Through the establishment of hereditary neural net model, various influential factors have been well considered and the forecast on road volume of traffic has been accurately made. Compared with experimental results, this forecast is better than that made by ARMA. The forecast on volume of traffic is an approach with high intelligence, which is equipped with learning ability and self adaptability. Thus it has good value for application.
出处
《广西交通科技》
2003年第2期24-26,共3页
Guangxi Communication Science & Technology