NEURAL-NETWORK IDENTIFICATION OF POWER-SYSTEM TRANSFER-FUNCTIONS

This paper describes an investigation into the use of a multilayered n eural network for measuring the transfer function of a power system fo r use in power system stabilizer (PSS) tuning and assessing PSS dampin g. The objectives are to quickly and accurately measure the transfer f unction relating the electric power output to the AVR PSS reference vo ltage input of a system with the plant operating under normal conditio ns. In addition, the excitation signal used in the identification proc edure is such that it will not adversely affect the terminal voltage o r the system frequency. This research emphasized the development of a neural network that is easily trained and robust to changing system co nditions. Performance studies of the trained neural network are descri bed. Simulation studies suggest the practical feasibility of the algor ithm as a stand-alone identification package and as a portion of a sel f-tuning algorithm requiring identification in the strategy. The same technique applied to a forward modelling scheme can be used to test th e damping contribution from different control strategies.