基于多目标差分进化算法的列车惰行控制  

Coast control of urban train based on multi-objective differential evolution algorithm

作　　者：韩蕙心 吴鹏[1] 吴杰[1] 李金键[1] 

机构地区：[1]西南交通大学电气工程学院,成都610031

出　　处：《计算机应用》2013年第A02期286-289,共4页journal of Computer Applications

基　　金：国家科技支撑计划项目(2009BAG12A01-A04-1)

摘　　要：针对地铁列车准点节能优化,在惰行控制基础上,引入多目标差分进化算法(DE),寻求满足安全运行的列车准点节能控制工况序列。首先建立了基于多目标的列车运行控制模型,在应用差分进化算法时,对区间按照起伏坡道划分成多个子区间,以区间列车工况序列作为染色体;通过预设列车运行防护曲线对种群进行修正,提高可行解的比例,从而确保快速有效地获得列车最优控制方案。基于Visual Studio开发环境编制了城轨列车牵引计算软件,最后以西安地铁二号线为算例,验证了模型与算法的正确性。In this paper an approach was proposed for optimization of speed profiles in energy saving of urban train under the premise of ensuring safety and punctuality. It deals with the problem by a multi-objective Differential Evolution （DE） algorithm based on coast control. In the established multi-objective train control model, one interval was divided into several sub-intervals according to the slope of the line, and the interval sequential list of operation mode was represented as a chromosome. After modifying the chromosome under the constraints of train protection curve to increase the proportion of possible solution, the fast and efficient access to optimal control plan could be ensured. A software of traction calculation simulation of urban train has been completed based on Visual Studio and from the case analysis of Xi＇an Metro Line 2, and the feasibility and performance of the designed model and algorithm were tested and verified.

关 键 词：多目标 差分进化 地铁列车 惰行控制 

分 类 号：TP391.9[自动化与计算机技术—计算机应用技术]