Skip to Main Content
Damping electromechanical oscillations in power systems using feedback signals from remote sensors is likely to be affected by occasional low bandwidth availability due to increasing use of shared communication in future. In this paper, a predictor corrector (PC) strategy is applied to deal with situations of low-feedback data rate (bandwidth), where conventional feedback (CF) would suffer. Knowledge of nominal system dynamics is used to approximate (predict) the actual system behavior during intervals when data from remote sensors are not available. Recent samples of the states from a reduced observer at the remote location are used to periodically reset (correct) the nominal dynamics. The closed-loop performance deteriorates as the actual operating condition drifts away from the nominal dynamics. Nonetheless, significantly better performance compared to CF is obtained under low-bandwidth situations. The analytical criterion for closed-loop stability of the overall system is validated through a simulation study. It is demonstrated that even for reasonably low data rates the closed-loop stability is usually ensured for a typical power system application confirming the effectiveness of this approach. The deterioration in performance is also quantified in terms of the difference between the nominal and off-nominal dynamics.