A hybrid neuro-fuzzy static var compensator stabilizer for power system damping improvement in the presence of load parameters uncertainty

Tuning of static var compensator (SVC) stabilizers traditionally assumes th at the system loads are voltage dependent with fixed parameters. However, t he load parameters are generally uncertain. This uncertain behavior of load parameters can de-tune the stabilizer gain-settings, consequently SVC stab ilizer with fixed gains can be adequate for some load parameters but contra rily reduce system damping and contribute to system instability with loads having other parameters. An adaptive network based fuzzy inference system ( ANFIS) for an SVC stabilizer is presented in this paper to improve the damp ing of power systems in the presence of load model parameters uncertainty. Takagi and Sugeno's fuzzy if-then rules and an adaptive feed-forward neural network with supervised learning capability are used in the ANFIS. The pro posed ANFIS is trained over a wide range of typical load parameters in orde r to adapt the gains of the SVC stabilizer. A MATLAB computer simulation is used to show the effectiveness of the proposed ANFIS SVC stabilizer. The s imulation results show that the tuned gains of the SVC stabilizer using the ANFIS can provide better damping than the conventional fixed-gains SVC sta bilizer. (C) 2000 Elsevier Science S.A. All rights reserved.