ABSTRACT
This chapter presents a brief summary of techniques including expert systems decision trees, artificial neural networks, fuzzy logic systems, and their hybrids. It provides an overview of artificial neural networks and some significant neural learning algorithms. The chapter offers the reader a glimpse of the main set of tools from the emerging area of intelligent systems. The evaluation of intelligent systems has shown that neural-based approaches are most beneficial for power system stability assessment, due to their ability to generalize and "learn" from historical information. The counter propagation network is not as robust as the back propagating techniques, but it can provide quick solutions for applications that cannot tolerate long training sessions. Back propagation artificial neural networks (ANN) are well suited to enhancement of the conventional Transient Energy Function (TEF) procedure for many reasons. The development of the robust ANN involves the selection of an appropriate network architecture to determine how the neurons are connected.
