Title: Online modelling based on Genetic Programming

Authors: Stephan Winkler, Hajrudin Efendic, Luigi Del Re, Michael Affenzeller, Stefan Wagner

Addresses: Institute for Design and Control of Mechatronical Systems, Johannes Kepler University, Linz, Austria. ' Institute for Design and Control of Mechatronical Systems, Johannes Kepler University, Linz, Austria. ' Institute for Design and Control of Mechatronical Systems, Johannes Kepler University, Linz, Austria. ' Department of Software Engineering, Upper Austrian University of Applied Sciences, College of Information Technology at Hagenberg, Austria. ' Department of Software Engineering, Upper Austrian University of Applied Sciences, College of Information Technology at Hagenberg, Austria

Abstract: Genetic Programming (GP), a heuristic optimisation technique based on the theory of Genetic Algorithms (GAs), is a method successfully used to identify non-linear model structures by analysing a system|s measured signals. Mostly, it is used as an offline tool that means that structural analysis is done after collecting all available identification data. In this paper, we propose an enhanced on-line GP approach that is able to adapt its behaviour to new observations while the GP process is executed. Furthermore, an approach using GP for online Fault Diagnosis (FD) is described, and finally test results using measurement data of NOx emissions of a BMW diesel engine are discussed.

Keywords: genetic programming; GP; data driven model identification; self-adaption; machine learning; online modelling; fault diagnosis; automatic learning; real time.

DOI: 10.1504/IJISTA.2007.012487

International Journal of Intelligent Systems Technologies and Applications, 2007 Vol.2 No.2/3, pp.255 - 270

Available online: 19 Feb 2007

Full-text access for editors Access for subscribers Purchase this article Comment on this article