基于改进灰狼优化算法的分布式电源优化配置  

OPTIMAL CONFIGURATION OF DISTRIBUTED GENERATION BASED ON IMPROVED GREY OPTIMIZATION ALGORITHM

作　　者：蔡国伟[1] 刘旭[1] 张旺[1] 孟涛 郑天宇 Cai Guowei;Liu Xu;Zhang Wang;Meng Tao;Zheng Tianyu(College of Electrical Engineering,Northeast Electric Power University,Jilin 132012,China;Electric Power Research Institute,State Grid Jilinsheng Electric Power Supply Company,Changchun 130021,China;Tonghua Electric Supply Company,Tonghua 134000,China)

机构地区：[1]东北电力大学电气工程学院,吉林132012 [2]国网吉林省电力有限公司电力科学研究院,长春130021 [3]通化市供电公司,通化134000

出　　处：《太阳能学报》2019年第1期134-141,共8页Acta Energiae Solaris Sinica

基　　金：国家高技术研究发展(863)计划(SS2014AA052502);长江学者和创新团队发展计划(IRT144);国家自然科学基金(51377017)

摘　　要：计及源荷侧随机波动性的影响,采用机会约束规划方法,建立综合考虑配电网运行费用、电压稳定性和污染气体排放量的分布式电源随机规划模型。考虑各评价指标的数量级不同,为避免过度优化某一评价指标,采用模糊技术建立综合满意度函数。结合半不变量理论和Cornish-Fisher级数展开计算配电网随机潮流,得到节点电压幅值的概率密度曲线,从概率的角度对系统电压水平进行评估。对一种新颖的灰狼优化算法进行改进,采用Tent映射产生的混沌序列代替随机产生的初始种群,同时提出一种非线性收缩因子。PG&E33节点系统的仿真结果验证该文模型和算法的有效性及合理性。Stochastic programming model of distributed generations is proposed based on chance constrained programming method in this paper,which considers operating cost of distribution network,voltage stability,pollution gas emissions and stochastic volatility of generation side and demand side. The quantity level of each evaluation index is different. The fuzzy technology is used to establish comprehensive satisfaction function to avoid excessively optimizing a certain evaluation index. Stochastic power flow is calculated with semi-invariant theory and Cornish-Fisher series expansion in distribution network. System voltage level is evaluated by probability density curve of node voltage amplitude. An improved grey wolf optimization algorithm is proposed,which uses a non-linear shrinkage factor and makes chaotic sequence of Tent map instead of initial population. The simulation results of PG&E33 system verify the validity and rationality of the model and algorithm.

关 键 词：配电网 分布式电源 灰狼优化算法 机会约束规划 Cornish-Fisher级数 

分 类 号：TM614[电气工程—电力系统及自动化]