基于LS-SVM的交通流组合预测模型

Traffic Flow Combining Forecast Model Based on Least Squares Support Vector Machine

智能交通系统是目前世界上公认的解决城市交通拥堵问题的最佳措施,而实时准确地交通流量预测则是实现智能交通系统和智能交通诱导控制的重要依据.针对城市交通"智能运输系统"和交通流的特性,在多元线性回归、支持向量机和改进的BP神经网络等三种预测模型的基础上,提出了基于最小二乘支持向量机方法的交通流组合预测模型.实验预测结果表明该组合预测模型具有较高的预测精度,为交通流量提供了一个更好的预测模型.

At present,intelligent transportation system is considered as the best practice to solve the problem of urban traffic congestion.Timely and accurate forecast of traffic flow is an important basis to achieve intelligent transportation system and intelligent transportation guidance and control.According to the city intelligent transportation system and the characteristics of traffic flow,the combining forecast model is set up based on least squares support vector machine after support vector machine,back propagation neural network modified by support vector machines and multiple linear regression which are applied to establish the model of traffic flow prediction respectively.Experiments show better effect and higher precision of forecast by the combining model of LS-SVM and this model is a better traffic flow forecast one.