蚁群与小波粒子群算法结合优化配电网重构

2019-01-10 01:48:14 现代电子技术2019年1期

李世光 孟凡涛 赵沙沙 高正中 程建军

关键词: 配电网; 重构; 蚁群算法; 小波变异粒子群算法; 有功损耗; 节点电压

中图分类号: TN911.1?34; TM732                   文献标识码: A                  文章编号: 1004?373X(2019)01?0124?05

Abstract: An ACO?IPSOWM algorithm combining ant colony optimization (ACO) and improved wavelet mutation particle swarm optimization (IPSOWM) is proposed to reconstruct the power distribution network after failure or adding DG efficiently and stably. The minimum active power loss and minimum node voltage deviation of the power distribution network are taken as the objective function, and converted into single?objective problem after weighting and normalization. The binary coded switching states and topological correction strategy are used to check the radiation of the power distribution network. The algorithm is preliminarily optimized by using ant colony optimization algorithm, and the wavelet mutation is used to extend the effective population space to avoid that the algorithm falls into the local optimum. The power distribution network reconfiguration is simulated after the fault of IEEE33 node system and DG output of system node. The experimental results show that the ACO?IPSOWM algorithm can combine the advantages of ACO and IPSOWM after selecting the appropriate parameters, and the reconstructed performance is better.

Keywords: power distribution network; reconfiguration; ant colony algorithm; wavelet mutation particle swarm optimization; active power loss; node voltage

0  引  言

配电网重构是指在维持配电网辐射结构同时满足系统约束条件下,通断配电网中的开关来改变其网络结构,从而使配电网正常运行,最大程度上保证配电网络供电的可靠性、安全性、稳定性要求。目前解决重构问题的方法有支路交换法[1]、最优流模式法[2]、遗传算法、免疫算法、模拟退火算法、差分进化算法等,均能较好地达到配电网重构的效果。文献[3]提出一种基于协同进化蚁群算法的含光伏发电的配电网重构方法,但单一的蚁群算法处理时间较长。文献[4]提出一种改进的二进制量子粒子群算法,对含DG的重构模型进行求解,引入遗传算法的交叉操作和变异操作来避免早熟,提高了算法的全局搜索能力。文献[5]采用基于改进粒子群算法的含分布式电源(DG)配电网的优化重构,由此可知重构时DG的接入有利于减少网络损耗,提高节点的电压水平。粒子群算法虽然迭代次数少,但计算时容易早熟使寻优结果不佳;蚁群算法不易早熟,但其迭代次数较多,收敛速度较慢。本文将蚁群算法与改进的小波变异粒子群算法结合使用,改进的小波变异可为粒子进化提供更好的方向,避免其陷入局部最优,分别通过故障后重构与接入DG后重构两个案例证明了算法的可行性。

加入DG后采用算例1中ACO?IPSOWM算法,在实验取值时将网络损耗和电压偏差纳入一个数量级,并按单目标问题处理,在大量实验中选取几组具有代表性的加权优化方案,如表2所示。

由表2可知,当网络损耗的权重系数[α]大时,配电网最小网络损耗相对较小;反之,节点电压偏差的权重系数[β]大时,最大节点电压的偏差相对较小。由于配电网是带DG正常工作,故重构优化目标函数更偏重网络损耗更小,故两权重系数选为[α=0.8,β=0.2]。

选取合适的权重系数后,对含DG和不含DG应用本文算法进行Matlab仿真实验,算例1已知不含DG时重构后网损值为138.526 kW,含DG调和重构后网损值为73.415 kW;图5分别为含DG和不含DG时配电网重构后的节点电压偏差值,不含DG重构后节点电压最大偏差值约为0.052 p.u.,接入DG重构后节点电压最大偏差值为0.036,由此可见接入DG出力优化对减少配电网网络损耗、提高电压质量有很大帮助,但是由图5可以看出一个问题,各个节点电压之间存在明显波动性,电压不稳定,这时可接入无功补偿装置弥补无功的缺失。

5  结  论

模拟仿真实验结果表明,ACO?IPSOWM能够结合蚁群算法和小波粒子群算法两种算法的优势,对于配电网故障后重构与接入DG重构后网络损耗更小,节点电压偏差更小,且算法较单一的粒子群算法和蚁群算法能够更好的寻优,但算法参数选取需要不断修正调和,这是下一步需要改进的地方。

参考文献

[1] 刘秋源,宫诗玖,袁琦.基于支路交换法的配网重构方法分析[J].电气开关,2014,52(5):55?58.

LIU Qiuyuan, GONG Shijiu, YUAN Qi. Analysis of the distribution network reconstitution method based on branding exchange method [J]. Electric switchgear, 2014, 52(5): 55?58.

[2] 原亚飞.含分布式电源的配电网重构研究[D].天津:天津理工大学,2016.

YUAN Yafei. Research on reconfiguration of distribution network with distributed generation [D]. Tianjin: Tianjin University of Technology, 2016.

[3] 刘科研,盛万兴,贾东梨,等.基于协同进化蚁群算法的含光伏发电的配电网重构[J].可再生能源,2017,35(5):702?708.

LIU Keyan, SHENG Wanxing, JIA Dongli, et al. Distribution network reconfiguration with photovoltaic generation based on co?evolutionary ant colony algorithm [J].  Renewable energy resources, 2017, 35(5): 702?708.

[4] 张涛,史苏怡,徐雪琴.基于二进制量子粒子群算法的含分布式电源配电网重构[J].电力系统保护与控制,2016(4):22?28.

ZHANG Tao, SHI Suyi, XU Xueqin. Distribution network reconfiguration with distributed generation based on improved quantum binary particle swarm optimization [J] Power system protection and control, 2016(4): 22?28.

[5] 陈丹阳,张雪霞.基于AMOPSO考虑分布式电源的配电网重构[J].太阳能学报,2017,38(8):2195?2203.

CHEN Danyang, ZHANG Xuexia. Distribution network reconfiguration of distributed generation based on AMOPSO algorithm [J]. Acta energiae solaris Sinica, 2017, 38(8): 2195?2203.

[6] TIAN Y, GAO D, LI X. Improved particle swarm optimization with wavelet?based mutation operation [C]// 2012 International Conference on Advances in Swarm Intelligence. Berlin: Springer?Verlag, 2012: 116?124.

[7] LING S H, IU H H, CHAN K Y, et al. Hybrid particle swarm optimization with wavelet mutation and its industrial applications [J]. IEEE transactions on systems, man & cybernetics Part B, 2008, 38(3): 743?763.

[8] 米文博,王宇,刘庆瑞.基于改进前推回代法的配电网的潮流计算与分析[J].科技展望,2017,27(5):87?88.

MI Wenbo, WANG Yu, LIU Qingrui. Power flow calculation and analysis based on improved forward?backward generation method [J]. Technology outlook, 2017, 27(5): 87?88.

[9] 王守相,王成山.现代配电系统分析[M].2版.北京:高等教育出版社,2014.

WANG Shouxiang, WANG Chengshan. Analysis of modern distribution network system [M]. 2nd ed. Beijing: Higher Education Press, 2014.

[10] 李國清.分布式电源对重要用户供电可靠性的影响研究[J].科技创新与应用,2017(27):171.

LI Guoqing. Research on the influence of distributed power on reliability of power supply to important users [J]. Technology innovation and application, 2017(27): 171.