Self-learning adaptive-network-based fuzzy logic power system stabilizer

An adaptive network-based fuzzy logic power system stabilizer (ANF PSS) with self-learning capability is presented in this paper. This method combines the advantages of artificial neural networks (ANNs) and fuzzy logic control schemes to design a new PSS, without resorting to another existing controller. In this approach, two ANFs are employed, one functions as power plant model, the other as controller. The error signal at the output of the plant is backpropagated through different stages in time to train the controller. The improvement of the dynamic performance of the power system is demonstrated by simulation studies for different operating conditions and disturbances