A fuzzy basis function network based power system stabilizer for generatorexcitation control

A fuzzy basis function network (FBFN) based power system stabilizer (PSS) i s presented in this paper. The proposed FBFN-based PSS provides a natural f ramework for combining numerical and linguistic information in a uniform fa shion. The proposed FBFN is trained over a wide range of operating conditio ns in order to retune the PSS parameters in real-time, based on machine loa ding conditions. The orthogonal least squares (OLS) learning algorithm is d eveloped for designing an adequate and parsimonious FBFN model. Time domain simulations of a synchronous machine equipped with the proposed stabilizer subject to major disturbances are investigated. The performance of the pro posed FBFN PSS is compared with that of a conventional power system stabili zer (CPSS) to demonstrate the superiority of the proposed stabilizer. The e ffect of parameter changes on the proposed stabilizer performance is also e xamined. The results show the robustness of the proposed FBFN PSS and its c apability to enhance system damping over a wide range of operating conditio ns. (C) 1999 Elsevier Science S.A. All rights reserved.