Voltage stability factor prediction in power system using artificial neural network / Norlida Musa

This thesis presents the applications of artificial neural network (ANN) to predict the voltage stability level of power system network. Two types of neural network have been used i.e. Back-propagation and Radial Basis Function Network. Both ANN models developed have three layers i.e. input layer, hidden layer and output layer. To determine the level of voltage stability, it is measured by using voltage stability factor, i.e. L-fector, developed by Jasmon. In both networks, the same sets of data have been used in the training and the same other sets of data for testing process. All those sets of data are generated by the Second Order Newton Raphson (SONR) load flow simulation. Real and reactive power have been used as input nodes and L-factor values as output node. Tests are carried out and the results are compared on the basis of learning rate, momentum and number of hidden node. From the results, it shows that the artificial neural network can be used to predict voltage stability level for power system, with an advantage using Radial Bassis Function Network since it performed better than Back-propagation Network.