采用辅助问题原理的多分区并行无功优化算法  

Multi-subarea Parallel Reactive Power Optimization Based on APP

作　　者：刘宝英[1] 杨仁刚[1] 

机构地区：[1]中国农业大学信息与电气工程学院,北京市海淀区100083

出　　处：《中国电机工程学报》2009年第7期47-51,共5页Proceedings of the CSEE

基　　金："十一五"国家科技支撑计划重大项目(2006BAJ04B06)。~~

摘　　要：针对大规模电网集中式串行无功优化计算所面临的计算问题和瓶颈问题,基于分布式并行计算思想建立一种多分区并行无功优化模型,并采用辅助问题原理综合考虑?–变量提出一种附加函数,有效地将全网的优化问题完全分解为多个子区独立的优化问题,同时解决了分区计算所引起的多平衡点问题。该方法实现了完全的分布式优化,解决了数据集中上传的瓶颈问题,有效降低了优化问题的求解规模,大大缩短了总运算时间,提高了计算与控制的实时性和灵活性。仿真结果表明:该方法有效可行,计算速度快,收敛性好。To solve the problems in computation and bottle-neck problem with centralized serial reactive power optimization of large scale power system, the multi-subareas parallel model of reactive power optimization based on distributed parallel computing was proposed. An auxiliary function based on auxiliary problem principle （APP） and A-variables was constructed to solve the problem of multi-slack bus resulted from subarea division. In this way, the distributed optimization mode is absolutely achieved to solve the bottle-neck problem. And the optimization problem could be effectively divided into several completely independent sub-problems to decrease the scale of problem solved, shorten computing time and improve real-timeliness. The simulation shows the algorithm proposed has good feasibility, fast speed and good convergence.

关 键 词：并行无功优化 辅助问题原理 附加函数 △-变量 分区 分布式并行计算 

分 类 号：TM714[电气工程—电力系统及自动化]