Predicting loss of synchronism and instability in generators using wide area measurements

In this paper, a new comprehensive framework is proposed to diagnose stability status of power systems and generators. The proposed method predicts rotor angles of generators using an online algorithm in the first place. Since it compensates for the delay in PMUs, rotor angles of generators will be available online with an acceptable accuracy. The proposed approach properly recognizes the power system stability status by calculating the difference between rotor angles of mutually coupled generators using the online data. Obtained value compare with threshold value and when this value exceeds threshold value, warning is given. Accordingly, it identifies the most critical generator using the fault occurrence and fault clearing time data. Moreover, the proposed method predicts the generator stability status using the boundary decision and the previously collected data, 150ms after clearing the fault. Decision boundaries are an inherent characteristic of each generator which can predict the generator stability status with minimum errors. As a result, the proposed method determines critical and unstable generators for generator shedding to prevent blackout in the system. The potential of this approach is tested using IEEE 39-bus (New England system) system.