基于交替方向乘子法的分布式双线性状态估计

提出了一种计及数据采集与监控(SCADA)系统和相量测量单元(PMU)混合量测的电力系统多区域分布式状态估计的新方法。首先假设量测模型是线性的,以扩展子区域法进行分区,所得到的扩展子区域包含了相邻区域的边界母线。然后采用广泛应用于分布式计算的交替方向乘子法(ADMM)求解,由于各子区域间仅需交换边界母线状态量信息,无需中央协调侧,因而保留了各子区域的独立性。随后基于SCADA系统和PMU混合量测的双线性模型,将传统的非线性加权最小二乘(WLS)估计转化为三阶段问题,第1、第3阶段为线性WLS估计(各子区域由ADMM分布式求解);第2阶段为一步非线性变换(各子区域可独立变换)。双线性理论与ADMM技术的结合,保证了分布式状态估计的收敛性,同时也提高了计算效率。最后,算例测试部分从多个方面论证了所述方法相比于现有方法的优越性。

Distributed Bilinear State Estimation Based on Alternating Direction Method of Multipliers

A new method for power system multi area distributed state estimation considering hybrid supervisory control and data acquisition (SCADA) system and phasor measurement unit (PMU) measurement is presented. First, it is assumed that the measurement model is linear, the network is decomposed by the extended sub areas including the border buses of neighboring sub areas. Then the problem is solved by the alternating direction method of multipliers (ADMM), a method widely used in distributed optimization. Only the states of border buses need to be exchanged among nearby areas without central coordination, thus retaining the independence of each sub area. Afterwards, based on bilinear theory integrating hybrid SCADA system and PMU measurements, the traditional nonlinear state estimation is transformed into a three stage problem. The first and third stage simply solve the linear state estimation problem during which ADMM is applied to solving the distributed problem, the second stage is one step nonlinear transform in which each sub area can perform this nonlinear transformation independently. The combination of bilinear theory and ADMM ensures the convergence of distributed state estimation while improving computational efficiency. Finally, numerical tests have verified the advantage of the proposal over existing methods in many aspects. This work is supported by National Natural Science Foundation of China (No. 51277052, No. 51107032, No. 61104045) and State Grid Corporation of China.