The application of a fuzzy controller to improve the stability of power systems is presented in this paper. To achieve good damping characteristics over a wide range of operating conditions, speed deviation and acceleration of a synchronous generator are chosen as the input signals to the fuzzy controller. The stabilizing signal is determined according to the nonlinear fuzzy membership function depending on the speed and acceleration states of the generator. Simulation and implementation studies have been carried out on a one-machine and two-machine infinite bus system respectively. The simulation and implementation result show that the proposed fuzzy controller is very effective and gives better performance than the conventional power system stabilizer.