Design of Recurrent Neural Network Power System Stabilizer based on Genetic Algorithm

C.-J. Chen and T.-C. Chen (Taiwan)


genetic algorithm, recurrent neural network, power system stabilizer.


This paper presents a novel recurrent neural network power system stabilizer (RNNPSS) based on genetic algorithm (GA) for a multi machines power system. The proposed PSS consists of a recurrent neural network identifier (RNNI) and recurrent neural network controller (RNNC) that identifies the power system and supplies an adaptive signal to the governor and exciter to damp power system oscillations. Both the RNNI and RNNC of the PSS are trained offline by using GA to determine the optimal learning rates. The proposed PSS is simulated for three-generator power system, which results demonstrate the effectiveness and performance of the proposed PSS.

Important Links:

Go Back