Optimal Placement of PMU in Distribution Systems Considering State Estimation Error

The existing optimal phasor measurement unit (PMU) placement (OPP) methods focus on the system observability and neglect to consider the influence of distribution system state estimation on the OPP solutions. This paper proposes a novel OPP approach for power systems where the state estimation error (SEE) is further considered in the OPP optimization problem. The SEE is calculated based on the non-Gaussian measurement noise statistics and the robust maximum likelihood estimator while the existing SEE methods are founded on the Gaussian noise assumption and the well known weighted least squares. Note that the distribution of the PMU measurement noise has been demonstrated to be non-Gaussian. The optimal experiment criterion is exploited to represent SEE constraint and then the constraint can be converted into a linear matrix inequality formula. The simulations on the IEEE 34-bus and 123-bus systems are carried out to verify the effectiveness of the proposed OPP approach.