This paper describes an on-line identification technique for modelling a turbogenerator system. The dynamics of a single turbogenerator infinite bus system are modelled using an adaptive artificial neural network identifier (AANNI) based on continual online training (COT). This paper goes further to show that multilayered perceptrons with deviation signals as inputs and outputs trained using the standard backpropagation algorithm retain past learned information despite COT. Simulation and practical results are presented.

Meeting Name

International Conference on Electric Power Engineering, 1999. PowerTech Budapest 99


Electrical and Computer Engineering

Keywords and Phrases

Adaptive Neural Network Identifier; Backpropagation; Backpropagation Algorithm; Continual Online Training; Deviation Signals; Dynamics Modelling; Identification; Inputs; Machine Control; Multilayer Perceptrons; Multilayered Perceptrons; On-Line Identification Technique; Outputs; Power Engineering Computing; Turbogenerator Infinite Bus System Dynamics; Turbogenerator System Modelling; Turbogenerators; Turbogenerators Control

Document Type

Article - Conference proceedings

Document Version

Final Version

File Type





© 1999 Institute of Electrical and Electronics Engineers (IEEE), All rights reserved.

Publication Date

01 Jan 1999