基于Pareto多目标遗传算法的单列车定时节能研究

Study on timing energy saving of single train based on Paretomulti-objective genetic algorithm

针对地铁列车定时节能运行的复杂多目标优化问题,提出一种基于Pareto多目标遗传算法来定时优化列车的区间运行能耗。首先以列车的能耗指标和准时指标为优化目标,同时将列车运行的安全指标、精准停车指标和舒适度指标作为约束条件,然后采用基于NSGA-II的Pareto多目标遗传算法进行求解,得到一组非支配解集,再从非支配解集中选择最符合定时运行要求的运行策略。最后通过广州地铁七号线的实际测量数据进行仿真验证,结果表明,定时节能优化相比节时策略能耗节约16.54%,而运行时间只比运营计划多了0.01 s,符合定时运行的要求。该方法具有较高的可行性,可以应用于地铁列车节能。

Aiming at the complex multi-objective optimization problem of timing energy saving of subway trains, a Pareto multi-objective genetic algorithm is proposed to optimize the energy consumption of the train. First, the energy consumption index and punctuality index are taken as the optimization targets while the safety indicator, precise parking indicator and comfort indicator are taken as the constraints. Second, a Pareto multi-objective genetic algorithm based on NSGA-II is used to solve the problem, and a set of non-dominated solutions is obtained. Then the most reasonable strategy which satisfying the timing requirements is selected from the non-dominated solutions. Finally, the simulation results are verified by actual measurement data of Guangzhou metro line 7. The results show that the timing energy saving strategy saves 16. 54% energy compared with the time saving strategy while the running time is only 0.01s more than the operation plan, which meets the timing requirements. The method is feasible and can be applied to the energy saving of the subway train.