电动汽车换电站调度优化问题下的约束罚函数  

Penetration penalty functions for optimal dispatch of EV battery swap station

作　　者：张颖[1] 容展鹏 张宇雄[1] 扈海泽 赵军[1] 韦梦[1] 方梦鸽 

机构地区：[1]长沙理工大学电气与信息工程学院,湖南长沙410004

出　　处：《电力自动化设备》2016年第10期24-30,共7页Electric Power Automation Equipment

摘　　要：传统的罚函数约束问题处理方式极易陷入局部最优,难以达到理想的优化效果。针对这一问题,提出了一种新的约束罚函数处理方法,该方法结合粒子群优化算法的原理特性和已有的自适应罚函数方法,组成了分段穿透罚函数。以粒子群优化算法为求解工具,对计及分时电价和可再生能源的电动汽车换电站多目标调度模型进行优化,在MATLAB仿真平台上对比了传统严格递增罚函数、自适应罚函数和分段穿透罚函数方法,验证了所提分段穿透罚函数处理方法具有求解速度快、全局寻优能力强的特点,能有效解决电动汽车换电站调度优化中的高维、高约束问题。A penalty function method for treating the constraints is proposed to avoid the local optimum easily occurred in the traditional methods,which combines the existing method of adaptive penalty function with the principle of PSO（Particle Swarm Optimization） algorithm to develop the segmental penetration pe-nalty functions. With the consideration of time-of-use electricity prices and renewable energy sources,a multi-objective dispatch model of EV（Electric Vehicle） battery swap station is optimized by the PSO algorithm.The rigorously-increasing penalty functions,adaptive penalty functions and segmental penetration penalty functions are compared on the MATLAB platform,and the results show the segmental penetration penalty functions have quicker speed and better global searching ability,effectively solving the high-dimensional and multi-constraint problem of dispatch optimization for EV battery swap station.

关 键 词：电动汽车 换电站 分布式电源 穿透罚函数 粒子群优化算法 调度优化 

分 类 号：TM73[电气工程—电力系统及自动化] U469.72[机械工程—车辆工程]