A novel approach for on-line adaptive tuning of power system stabilizer (PS
S) parameters using radial basis function networks (RBFNs) is presented in
this paper. The proposed RBFN is trained over a wide range of operating con
ditions and system parameter variations in order to re-tune PSS parameters
on-line based on real-time measurements of machine loading conditions. The
orthogonal least squares (OLS) learning algorithm is developed for designin
g an adequate and parsimonious RBFN model. The simulation results of the pr
oposed radial basis function network based power system stabilizer (RBFN PS
S) are compared to those of conventional stabilizers in case of a single ma
chine infinite bus (SMIB) system as well as a multimachine power system (MM
PS). The effect of system parameter variations on the proposed stabilizer p
erformance is also examined. The results show the robustness of the propose
d RBFN PSS and its ability to enhance system damping over a wide range of o
perating conditions and system parameter variations. The major features of
the proposed RBFN PSS are that it is of decentralized nature and does not r
equire on-line model identification for tuning process. These features make
the proposed RBFN PSS easy to tune and install. (C) 1999 Elsevier Science
S.A. All rights reserved.