改进GAAA算法在多目标电网规划中的应用  

Application of Improved GAAA Hybrid Algorithm in Multi-objective Power Network Planning

作　　者：孔祥聪[1] 周步祥[1] 汝锐锐[1] 肖贤[1] 叶宗斌[1] 

机构地区：[1]四川大学电气信息学院,成都610065

出　　处：《电力系统及其自动化学报》2013年第6期112-116,共5页Proceedings of the CSU-EPSA

摘　　要：针对多目标电网规划中多维变量问题,该文采用GAAA算法对多目标电网规划的应用进行建模。建立了能够描述电网经济性、可靠性、N-1静态安全的多维动态规划模型。为改善GAAA算法的性能,对该算法分别进行了种群初始化、信息素更新精英策略、变量离散化和越界处理等措施的改进,提高了GAAA算法的收敛性和寻优效率。试验中,通过对一个46节点系统进行测试,该计算结果证明了改进的GAAA算法能有效地解决电网规划含离散性变量的多维组合优化问题。To solve the multidimensional variables problems of multi-objective power network planning, the paper carries on the modelling from applying the GAAA hybrid algorithm in muhi-objective power network planning. A mul- tidimensional dynamic programming model is established to describe the economic efficiency, reliability, and N-1 static security of the power network. To improve the performance of the GAAA, the paper respectively makes better the follwing measures： population initialization, pheromone updating with elitist strategy, the variable diseretization and cross-border processing, which can enhance the convergence and optimization efficiency of the GAAA. At last, the pa- per tests a 46-bus system and draws a conclusion, which shows that the improved algorithm can effectively solve the multi-dimensional combination problems with discrete variable optimization of power network planning.

关 键 词：多目标电网规划 全局寻优 收敛性 离散性变量 GAAA 

分 类 号：TM715[电气工程—电力系统及自动化]