Online Assessment of Voltage Stability Region using an Artificial Neural Network

As the voltage instability disturbs the adequate operation of power system, incessant observing of the system status is needed. Artificial Neural networks can be used for voltage stability monitoring due to the non-linear nature of the voltage stability prediction problem. In this work, a Multi-layered Feed Forward Neural Network (MLFFNN) with less amount of neurons is used. It estimates the voltage stability index of all the buses by using VCPI (Voltage Collapse point indicator) under different loading conditions. Test results indicate that the proposed Multi-Layer Feed Forward neural network-based approach gives exact assessment of Voltage Stability Index values for different loading conditions. Since the method is much quicker it can evidence that, it is easily adopted for on-line applications as compared to conventional power flow methods. The standard IEEE 14 bus test system is tested with the proposed method.