Quickest line outage detection and identification: Measurement placement and system partitioning

This paper addresses the problem of detecting and identifying transmission line outages in near real-time. To this end, we utilize the statistical properties of electricity generation and demand, and apply a quickest change detection algorithm to measurements of voltage phase angles collected using phasor measurement units (PMUs). We propose a procedure for optimally selecting the buses for PMU placement with the goal of minimizing the worst case line outage detection delay. In addition, we propose a method to optimally partition the power system into several areas such that the algorithm for line outage detection can be applied in parallel to each area for faster detection. We illustrate the proposed ideas via case studies involving the IEEE 30-bus test system.