Network observability: a solution technique using neural networks

A new method for the network observability solution of the power networks using the neural networks with quickprop as training algorithm is presented in this paper. The network observability problem related to the power network configuration or network topology, called as the topological observability, is taken for the solution. The topological network observability is determined using a neural network model, based on the quickprop algorithm, which uses the second order derivatives of the error function to speed up the learning. This neural network based method has been applied on sample power networks and results are presented.