Self-tuning feedback linearization controller for power oscillation damping

Power systems exhibit highly nonlinear behavior especially under large disturbances like faults, outages etc. necessitating application of nonlinear control techniques. Nonlinear estimation and control of power oscillations through FACTS devices is illustrated in this paper. A special form of nonlinear neural network compatible with the feedback linearization framework is used. Levenberg-Marquardt (LM) algorithm is adapted to work in sliding window batch mode for online estimation of system oscillatory behavior. At each sampling interval the estimated neural network parameters are used to derive appropriate control using the feedback linearization technique. Use of LM is shown to yield better closed-loop performance compared to conventional recursive least square (RLS) approach. A case study is presented to demonstrate the effectiveness of feedback linearization controller (FBLC), especially, under stressed operating conditions. Its performance is compared against pole-shifting controller (PSC) under different scenarios.