Probabilistic Transient Stability Assessment of Power Systems Using Artificial Neural Network

Abstract

With the evolution of renewable energy sources, such as wind and photovoltaic generators (PVGs), the present electric power systems will transform into systems, possessing various amounts of uncertainties. These systems will be dominated by an unparalleled combination of a variety of generation and power transmission technologies, accompanied by flexible load and storage devices, and possessing spatial and temporal uncertainties. In this regard, the significance of probabilistic transient stability cannot be overlooked. The inherent methods to determine power system transient stability, such as Lyapunov direct method (based on transient energy function), and time-domain simulation method (based on numerical integration and algebraic-differential equations), have proven to be very computationally intensive. Novel soft computing techniques, such as machine learning and neural networks, provide promising results for tackling such kind of issues. Therefore, this paper aims to describe and discuss the framework for probabilistic transient stability in electric power systems and the application of artificial neural network to enhance its evaluation process. Moreover, the approach of probabilistic transient stability is demonstrated using the standard IEEE 39-bus test system. Finally, a direction for future research in this growing area is identified.