On-line security monitoring and analysis using Levenberg-Marquardt algorithm-based Neural Network

Abstract

Due to open access in the restructured power system, the events of bus voltage limit violation and transmission line overloading are occurring frequently. These events are mainly responsible for several incidents of major network collapses leading to partial or even complete blackouts and due to this, security monitoring and analysis has become a challenging task to be performed at energy control centre. A fast and accurate method of Power Flow (PF) study may be able to investigate the system security by determining the power system static states, i.e. voltage magnitude and voltage angle at each bus. In this paper, a Levenberg-Marquardt algorithm-based Neural Network (LMNN) has been proposed which provides a fast learning to the multi-layer neural network. The effectiveness of the proposed LMNN-based approach for security monitoring and analysis has been demonstrated by computation of bus voltage magnitudes and voltage angles for line-outage contingencies at different loading conditions in IEEE 14-bus system.