ICWLS-SVM在城市道路短时交通流量预测建模中的应用

Application of Improved Cauchy Weighted Least Squares Support Vector Machine in Urban Road Short-Term Traffic Flow Prediction Model

为了提高短时交通流量(STF)预测的精度,提出一种改进的柯西加权最小二乘支持向量机(ICWLS-SVM)建模方法,该模型方法基于改进的柯西分布加权规则,能够对不同的训练样本进行自适应赋权,以合理分配数据样本对模型的贡献.采用遗传粒子群优化(GPSO)算法对模型参数进行优化选择,以获得较优的模型参数.数值仿真实验表明ICWLS-SVM具有较强的鲁棒性,其预测性能优于WLS-SVM和LS-SVM.同时,利用实测交通流量数据对ICWLS-SVM模型进行性能测试,结果表明该建模方法在短时交通流量预测建模方面是有效且可行的.

In order to increase the accuracy of short-time traffic flow prediction, an Improved Cauchy weighted least squares support vector machine （ICWLS-SVM） regression method is proposed. The Genetic particle swarm optimization （GPSO） algorithm is applied to determine the optimal parameters of the ICWLS-SVM. The simula- tion experiment results show that the prediction performance is better than those of weighted least squares support vector machines （WLS-SVM） and least squares support vector machines （LS-SVM） method. Furthermore, the ICWLS-SVM method was applied to build short-term traffic flow forecasting model, and the satisfactory result is obtained.