Browse by author
Lookup NU author(s): Professor Ehsan Mesbahi
Full text for this publication is not currently held within this repository. Alternative links are provided below where available.
The behaviour of highly nonlinear industrial systems often varies according to different operational conditions. an additional factor degrading controlled system performance can be the high level of uncertainty often experienced during the identification process. Adaptive control methods apply system identification techniques in order to obtain a model of the overall process from input-output knowledge and then use this model within the design of the controller. the main objectives of this article are: to investigate neural controller schemes for adaptive control of time-variant dynamic systems; to study the offline application of recurrent artificial neural networks as a model reference adaptive controller for the mathematical model of high-speed diesel engine operation; and to propose, implement, and investigate the feasibility of an online neural adaptive speed controller for a PERKINS high-speed diesel engine. The proposed neural adaptive controller is implemented in the form of a neuro-governor used to control the speed of a high-speed diesel engine. Variations in the design and training of the controller and associated identifier are discussed, and simulation results are presented in detail.
Author(s): Mesbahi E
Publication type: Article
Publication status: Published
Journal: Control and Intelligent Systems
Year: 2003
Volume: 31
Issue: 3
Pages: 138-153
Print publication date: 01/01/2003
ISSN (print): 1480-1752
ISSN (electronic):
Publisher: ACTA Press
