Optimal placement of phasor measurement units (PMU) using a novel method based on observability gramian and full observability rank

Abstract

This paper proposes a new methodology based on observability gramian for optimal placement of phasor measurement units (PMUs) to achieve full observability of power system. This would enable better state measurement and state estimation of large scale power systems. An application to Nine-bus system and Two-area system proves that the proposed architecture is unique and provides better results compared to the existing metbod for PMU placements. This energy function approach based on observability gramian is simple in structure but perform better than existing methods for tbe PMU placements. The results illustrates that this method is scalable and the architecture can be implemented in modern power grid.Index Terms---Observability Gramian, Principal component analysis, phasor measurement units (PMUs), optimal placement of sensors. I . INTRODUCTIONModem power systems highly utilize available transnusslOn and generation with large amounts of power interchanged among electric companies and geographical regions, making them highly complex, nonlinear and time varying in nature. With the recent deregulation and increase in the demand, modem power systems becoming more vulnerable much so even if small disturbances occur, maintaining power system stability and security is becoming more and more difficult. I n order to operate power systems more effectively without reduction in system security and quality of supply, new methods of real-time state estimation are extremely critical. This is even more critical in case of contingency conditions such as loss of transmission line or generator units which will occur more frequently under new reforms.Secure operation of power system requires close monitoring of the system operating conditions. This is accomplished using the state estimator which resides in the control center computer and has access to measurement received from numerous substations in monitored system. Studies and research undertaken to analyze root cause of major blackout happened in 2003 revealed inadequacy in traditional methods of state estimation and power system