摘要
针对含分布式电源(DG)的有源配电网重构问题,建立考虑DG出力优化的多场景重构模型,提出改进入侵杂草算法(MIWO)求解.该模型从配电系统运行效益和环境效益两个角度,以有功网损、节点电压偏差、负荷均衡度、污染物排放环境成本和弃风弃光成本最小为目标函数.MIWO算法在入侵杂草算法的基础上,设计适用于重构问题的初始种群选择机制,优化了随机产生的初始值;融合Lévy飞行过程和类海明距离判定,保持了种群的多样性,避免陷入局部寻优;提出种子数量调节策略,提高了收敛速度.IEEE 33节点配电系统仿真结果验证了所建模型和所提算法的有效性.与入侵杂草算法、粒子群算法和遗传算法相比,MIWO算法收敛速度更快、寻优能力更强、稳定性更好.
In view of active distribution network with distributed generation （DG） reconfiguration problem, a multi- scene reconfiguration model considering optimizing DG output is established and the modified invasive weed optimi- zation （MIWO） algorithm is proposed to solve the model. Considering operation benefit of distribution system and environmental benefit, the model is built to get reconfiguration objective function to minimize active power loss, node voltage deviation, load balancing, environmental cost of pollutant emission and the costs of abandoning the wind and the light. Based on modified invasive weed optimization algorithm, the proposed algorithm designed initial population selection mechanism for reconfiguration problem to optimize initial value generated randomly, introduced process of L6vy flight and quasi hamming distance determination to maintain population diversity and avoid local optimization, and the proposed seed quantities adjusting strategy to improve convergence speed. The testing results of IEEE 33 node distribution system have confirmed the validity of the proposed model and algorithm. Compared with invasive weed optimization algorithm, particle swarm optimization algorithm and genetic algorithm, MIWO algo- rithm has faster convergence speed, better global search ability and stronger stability.
作者
石季英
袁大玲
薛飞
马丽
杨挺
Shi JiyingI;Yuan DalingI;Xue Fei;Ma LiI;Yang Ting(School of Electrical and Information Engineering,Tiaxljin University,Tiaxljin 300072,China;State Grid Ningxia Electric Power Research Institute,Yinchuaxl 750001,China)
出处
《天津大学学报（自然科学与工程技术版）》
EI
CSCD
北大核心
2018年第8期786-796,共11页
Journal of Tianjin University：Science and Technology
基金
国家高技术研究发展计划(863计划)资助项目(2015AA050202)
国家自然科学基金资助项目(61571324)
天津市自然科学基金资助项目(16JCZDJC30900)
国际科技合作专项资助项目(2013DFA11040)~~
关键词
有源配电网重构
改进入侵杂草算法
分布式电源
环境效益
active network reconfiguration
modified invasive weed optimization algorithm
distributed generation
environmental benefits
