基于蚁群优化支持向量机的短时交通流量预测

摘要

为了提高短时交通流量的预测精度,提出一种蚁群算法(ACO)优化支持向量机(SVM)参数的短时交通流量预测模型(ACO-SVM).将SVM参数的选取看作参数的组合优化问题求解,采用鲁棒性较强的ACO来搜索最优解.仿真结果表明,ACO-SVM在预测精度、收敛速度、泛化能力等方面均优于参比模型,更适合于短时交通流量的预测.%In order to improve the prediction accuracy of short-term traffic flow, this paper proposes a short-term traffic flow prediction model based on support vector machine (SVM) optimised by ant colony algorithm (ACO). The parameters selection problem of SVM could be considered as the problem solving with parameters combinatorial optimisation, ACO with fine performance of robustness is applied to search the optimal value of objective function. Simulation results show that the proposed ACO-SVM method is superior to reference models on the aspects of prediction accuracy, convergence time, generalisation ability and so on, and is more suitable for short-term traffic flow prediction.