基于自适应粒子群优化算法的交通量短时预测模型

A Short-time Traffic Flow Prediction Model Based on Adaptive Particle Swarm Optimization Algorithm

在考虑交通量短时变化的时空特性和波动性基础上,建立了非线性交通量短时预测模型.根据我国城市道路交通流非线性、时变性、随机性等特点,提出自适应粒子群优化算法对非线性交通量短时预测模型进行在线修正.该自适应粒子群优化算法采用两步优化策略,对算法参数进行调整,避免算法早熟收敛,有效提高了算法的运算精度和效率.利用城市道路的实测数据,通过Mat-lab软件工具箱对该模型进行计算机仿真验证.

The paper proposed a new nonlinear short-time traffic flow prediction model based on space-time and wave attributes of the urban traffic flow, which effectively overcame nonlinearity, time-variation and random characteristics of the urban traffic flow. Then this paper put forward a new Adaptive Particle Swarm Optimization （APSO） algorithm to the model. This APSO algorithm introduced two-steps feedback tactics to amend parameters of the algorithm, in order to improve the accuracy and efficiency for the algorithm. Finally,the model and its algorithm were operated on Matlab.