CIPSO算法在城市有轨电车控制策略中的应用研究

在保证电车安全的前提下,轨道交通中的城市有轨电车控制策略优化问题实质上是多目标优化问题,主要是针对节能、正点、停靠准确和乘客舒适度优化等方面的复杂问题,以电车运动学方程为基础,针对粒子群优化算法在离散优化问题中处理不佳,容易陷入局部最优的问题,采用混沌Tent映射初始化粒子群,建立其多目标优化模型。而后采用免疫接种和免疫选择的方法提高PSO优化算法的优化能力,对模型进行求解。以广州市海珠区环岛新型有轨电车试验段数据为对象进行实例仿真,结果表明,混沌免疫微粒群优化算法较传统微粒群优化算法可获得更好的控制策略,能更有效的解决电车运行多目标优化问题。

Application of CIPSO Algorithm in City Tram Control Strategy of Rail Transit

Under the premise that the safety of train is guaranteed,the optimization of city tram control strategy is actually a multi-objective optimization issue,and it is mainly aimed at the complex problems about energy saving,punctuality,stopping accuracy and the improvement on passenger comfort and etc.Particle swarm optimization algorithm does not perform well in discrete optimization,because it is likely to result in local optimum. Therefore,chaos Tent mapping is adopted to initialize particle swarm and to establish its multi-objective optimization model on the basis of the train kinematic equation. Then immune vaccination and immune selection are used to improve the optimization ability of PSO optimization algorithm and to solve the model. Simulation is conducted on the data from the test section of the new city tram around Haizhu district of Guangzhou. The results show that the chaotic immune particle swarm optimization algorithm can obtain better control strategy than the traditional particle swarm optimization algorithm,and it is more effective in solving the multi target optimization problem in train operation.