神经网络模型在短期交通流预测领域应用综述

Review of Application of Neural Network Based Models in Short-Term Traffic Flow Forecasting

对基于神经网络的预测模型和方法的研究进行了综述,基于神经网络模型用于短期交通流预测的优点和固有缺陷,认为多种神经网络相结合的混合模型比单一的神经网络模型的预测效果要好,而将神经网络模型与其他领域的研究相结合的综合模型的预测效果要好于混合模型。因此,神经网络与各相关学科的人工智能技术有机结合将会形成强大的综合优势,更有效地用于短期交通流预测研究。

The short-term traffic flow forecasting is the basis for the Dynamic Route Guidance Systems (DRGS).A lot of prediction models have been put forward in the past.Recently,Neural Networks based models have been studied extensively.The models based on neural networks are summed up,analyzed and evaluated.Based on the merits and demerits of the Neural Networks,the mixed neural networks models are better than the single models while the integration of neural networks and the other field study is better than the mixed models.In the near future,the integration of the neural networks and the artificial intelligence will be widely studied so as to be applied more effectively in short-term traffic flow forecasting.