This paper presents the design of a Power System Stabilizer synthesized using an Artificial Neural Network. The patterns used in the network training are sets of controller parameters, previously calculated for several system operation points using the pole-placement method. The trained network presents, as its main characteristic, uniform values for all the stabilizers parameters when the system synchronous machine is generating reactive power, but these same parameters suffer great variations when reactive power is being absorbed by the machine. Simulation tests show very good performance for the proposed Neural PSS, when compared with a fixed-parameter stabilizer.