Load Flow Analysis using Intelligence-based Hopfield Neural Network for Voltage Stability Assessment

This paper presents a novel intelligence-based recurrent hopfield neural network (HNN) for solving the non-linear power flow equations. The proffered method is an energy function-based approach formulated using power residuals of the system. The dynamics associated with the neural networks are minimized by intelligence-based technique to determine the unknown parameters such as voltage magnitude (V) and phase angle (δ) of the system. A hybrid particle swarm optimization-gravitational search algorithm (PSO-GSA) has been used to minimize the dynamics of HNN and its stability is proved in Lyapunov sense of notion. The effectiveness of the method is tested on IEEE 14-bus system and the results obtained are compared to the conventional newton raphson method. Moreover, the stability indices such as voltage stability load index, line stability index, fast voltage stability index and line stability factor pertaining to the assessment of stability under the contingency case of N-1-1-1 was evaluated using the presented load flow analysis technique to study the stability of the system.