计及坏数据辨识的微网群三相分布式状态估计方法

Three-phase Distributed State Estimation Method of Microgrid Group Considering Bad Data Identification

为了满足微网群数据分布式管理的实际需求,微网群状态估计需要采用分布式模式。首先提出了一种分布式坏数据辨识方法,实现对微网群系统中不良数据的检测和辨识,然后提出基于增广拉格朗日交替方向非精确牛顿(augmented Lagrangian alternating direction inexact Newton,ALADIN)法的双层分布式状态估计算法。在数据交换量方面,每个分区分别进行独立的状态估计计算,分区之间仅仅通过交换相邻分区的边界耦合信息,采用共轭梯度(conjugate gradient,CG)算法更新每个分区的拉格朗日乘子,避免了信息隐私的暴露。基于5节点和21节点微网群系统的算例结果表明,所提分布式状态估计算法能够分布式辨识杠杆量测和有效处理坏数据,具有分区间信息交换少和收敛性好的特征,且能够达到集中式计算的求解精度。

In order to meet the actual demands of distributed management of microgrid group data,the distributed mode needs to be adopted in the state estimation of microgrid group.In this paper,a distributed bad data identification method is firstly proposed to realize the detection and identification of bad data in microgrid group system,and a two-layer distributed state estimation algorithm is proposed based on the augmented Lagrangian alternating direction inexact newton(ALADIN)method.In terms of the amount of data exchange,independent state estimation is performed in each partition,boundary coupling information of adjacent partitions is only exchanged between partitions,and the Conjugate Gradient(CG)algorithm is utilized to update the Lagrange multiplier of each partition,thus avoiding information privacy exposure.The results of calculation examples based on 5-node and 21-node microgrid cluster systems show that the distributed state estimation algorithm in this paper can identify leverage measurement in a distributed manner and effectively process bad data.It has the characteristics of less information exchange between partitions and good convergence,and is able to achieve the solution accuracy of centralized calculation.