State estimation and transient stability analysis in power systems using artificial neural networks

Power system state estimation and security assessment have recently become two major issues in the operation of power systems due to the increasing stress on power system networks. Utility operators must be properly informed of the operating condition or state of the power system in order to achieve a more secure and economical operation of today's complicated power systems. The state of a power system is described by a collection of voltage vectors for a given network topology and parameters. In this work, we applied Artificial Neural Networks (ANN) to estimate the state of a power system. State filtering and forecasting techniques were used to build Time Delay Neural Network (TDNN) and Functional Link Network (FLN) to capture the dynamic of a power system. Security assessment is the evaluation of a power system's ability to withstand disturbances while maintaining the quality of service. Many different techniques have been proposed for stability analysis in power systems. We focused on using neural networks as a fast and accurate alternative to security assessment. We developed an ANN-based tool to identify stable and unstable conditions of a power system after fault clearing. The hybrid method employing neural networks was used to successfully evaluate the Transient Energy Function (TEF) as a security index.