基于PSO的轨道交通列车节能控制优化研究

Study on Energy-saving Control Optimization for Rail Transit Train Based on PSO

针对当前列车节能控制智能优化研究中广泛采用遗传算法,但存在算子多、收敛速度不快的问题,采用粒子群算法结合罚函数法思想进行研究。通过对列车运行过程及列车牵引、巡航、惰行和制动阶段能耗状态的分析,给出列车节能控制的多约束优化数学模型和牵引能耗的计算方法。基于罚函数将该问题转化为无约束优化问题,并采用粒子群算法寻优求解,从而得出可使列车运行能耗最小的工况转换点。仿真结果表明,与改进的遗传退火算法相比,算法节能效果更优,且具有良好的收敛速度。

Currently, the genetic algorithms are widely adopted in the research of intelligent optimization of train energy-saving control. Since the genetic algorithms have many operators and may converge slowly, the particle swarm optimization （PSO） algorithm combined with the penalty function method is used in this paper. Through the analysis on the train operation process and the energy consumption of the train during the four stages, i.e. traction, cruise, coasting and braking, a multi-constraint optimization mathematical model of train energy-saving control is established and the calculation method of traction energy consumption is also given. The problem is converted into an unconstrained one by using penalty functions and then solved by the PSO algorithm to obtain the operation mode transition points which can minimize the operation energy-consumption of the train. The simulation shows that the proposed method gains a better energy-saving effect compared to the modified genetic algorithm-simulated annealing method （MGASA） and has a fast rate of convergence.