NICA算法在多类型分布式电源规划中的应用  

Novel Immune Clonal Algorithm for Multi-types Distributed Generators Planning

作　　者：王进[1] 陈加飞[1] 刘娇[1] 许一帆[1] 杨芳华[1] 唐浩 

机构地区：[1]长沙理工大学电气与信息工程学院,长沙410114 [2]湖南省怀化市沅陵县辰州东街凤滩水力发电厂,怀化419699

出　　处：《电力系统及其自动化学报》2015年第9期21-28,共8页Proceedings of the CSU-EPSA

基　　金：国家自然科学基金资助项目(71071025)

摘　　要：在综合多种分布式电源出力与负荷时序特性的基础上,建立了兼顾环境因素、发电成本、网损和静态电压稳定裕度的分布式电源多目标规划模型;提出了应用一种新的免疫克隆算法NICA(novel immune clonal algorithm)来求解该模型。该算法采用整体克隆非支配抗体、非一致性变异和删除帕累托(Pareto)前端密集解的策略来保证算法的收敛速度和解的均匀性。以IEEE33节点配电测试系统为例,与改进非劣分层遗传算法NSGA-II(non-dominated sorting genetic algorithm II)相比,算例分析结果表明,该算法所得最优解在Pareto-前端的分布更宽广、更均匀,验证了该算法的可行性和有效性。A multi-objective programming model of distributed generators（DG） is established, which gives considerations to environmental factors, generation costs, power loss and the stability of steady state voltage on the base of synthesizing timing characteristics of DG daily output and load. And a novel immune clonal algorithm （NICA） is proposed to solve this model. The NICA uses the overall clonal non-dominated ones, non-uniform mutation and removing Paretofront-intensive solutions to ensure the convergence speed and the uniformity of solution. Taking the IEEE33 bus system for example, the analysis results showed that the optimal solution solved by NICA algorithm is more broad and more uniform in Pareto-front distribution comparing with the non-dominated sorting genetic algorithm II（NSGA-II）, and ver- ified the feasibility and effectiveness of NICA.

关 键 词：分布式电源 时序特性 多目标规划 新的免疫克隆算法 帕累托前端 

分 类 号：TM712[电气工程—电力系统及自动化]