Application of neural network observer for on-line estimation of solid-rotor Synchronous Generators' Dynamic Parameters using the operating data

Parameter identification, considering both dynamic performance and energy efficiency is critical for modern control strategies in electrical power systems. This paper presents a novel application of ANN observers in estimating and tracking solid-rotor Synchronous Generator Dynamic Parameters using on-line disturbance measurements. The data for training ANN observers are obtained through off-line simulations of the generators operating in a one-machine-infinite-bus environment. The Levenberg-Marquardt algorithm has been adopted and incorporated into the back-propagation learning algorithm for training feed-forward neural networks. The inputs of ANN are organized in coordination with the results of the observability analysis of synchronous generator dynamic parameters in its dynamic behavior. A collection of ANNs with similar input patterns but different outputs are developed to determine a set of the dynamic parameters. The ANNs are employed and tested to identify the above parameters by the on-line measurements which are carried out within each kind of fault separately. Simulation studies indicate the ANN observer has a great ability to identify the dynamic parameters of the solid-rotor synchronous generators. The results also show that the tests which have given more accurate results in estimation of each parameter can be obtained.