基于群智能的建筑配电网状态估计算法研究  

Research on the State Estimation Algorithm of Building Power Distribution Network Based on Swarm Intelligence

作　　者：张玉晗 邢建春 胡浩宇 刘夕广 ZHANG Yuhan;XING Jianchun;HU Haoyu;LIU Xiguang(Army Engineering University of PLA,Nanjing 210007,China)

机构地区：[1]中国人民解放军陆军工程大学,南京市210007

出　　处：《建筑电气》2020年第6期42-48,共7页Building Electricity

基　　金：国家重点研发计划课题,课题名称:新型建筑智能化系统控制与管理应用软件研究,课题编号:2017YFC0704107。

摘　　要：针对现有建筑配电网不良数据检测方法应用在群智能系统架构时出现实时性差、准确性低等方面的不足,提出一种基于群智能系统架构的建筑配电网分布式状态估计算法。将全局估计问题分解为若干个子优化问题,分别在各个群智能节点上进行求解,依据基本电气原理建立约束方程,通过交替方向乘子算法实现各节点之间的协作,从而实现配电网的状态估计。最后,通过对某人防工程配电网算例仿真验证所提方法的有效性。与传统集中式系统架构下的状态估计方法相比,所提出方法的收敛速度和估计精度具有明显的优势。Given that some deficiencies(e.g.poor real⁃time performance and low accuracy)are found when the existing bad data detection method of building power distribution network is applied to the swarm intelligence system architecture,a distributed state estimation method of building power distribution network based on the swarm intelligence system architecture is put forward.The global estimation problem is divided into several sub⁃optimization problems,which is separately solved on each swarm intelligence node,and the constraint equation is established according to the basic electrical principles,and then the alternating direction method of multipliers(ADMM)is used to realize the coordination between various nodes and further realize the state estimation of the power distribution network.Finally,the validity of the proposed method is verified through the example simulation of the power distribution network of a civil air defence works.Compared with the state estimation method in the conventional centralized system architecture,the proposed method has obvious advantages in terms of rate of convergence and estimation accuracy.

关 键 词：抗差状态估计 分布式算法 加权最小绝对值估计法 交替方向乘子算法 建筑配电网 群智能 智能节点 收敛速度 

分 类 号：TU852[建筑科学]