Synchronous machine steady-state stability analysis using an artificial neural network

In the developed artificial neural network, those system variables which play an important role in steady-state stability, such as generator outputs and power system stabilizer parameters, are used as the inputs. The output of the neural net provides the information on steady-state stability. Once the connection weights of the neural network have been learned using a set of training data derived offline, the neural net can be applied to analyze the steady-state stability of the system in real-time situations where the operating conditions change with time. To demonstrate the effectiveness of the proposed neural net, steady-state stability analysis is performed on a synchronous generator connected to a large power system. It is found that the proposed neural net requires much less training time than the multilayer feedforward network with back-propagation-momentum learning algorithm. It is also concluded from test results that correct stability assessment can be achieved by the neural network