Enhancing the performance of a multivariable fuzzy controller by means of multiobjective genetic programming and statistical analysis

Bica, B., Chipperfield, A.J., Fleming, P.J. and MacKenzie, S. (2000) Enhancing the performance of a multivariable fuzzy controller by means of multiobjective genetic programming and statistical analysis. In IECON Proceedings (Industrial Electronics Conference). IEEE. 6 pp . (doi:10.1109/IECON.2000.973210).

Record type: Conference or Workshop Item (Paper)

Abstract

The paper addresses the issue of performance optimisation of a MIMO fuzzy controller for a gas turbine aero-engine. The proposed method attempts to improve the performance of the controller by looking at the accuracy of the input-output mapping of the control parameters. A multiobjective genetic programming approach is utilized to search for suitable input-output structures, able to satisfy the rigorous performance criteria imposed on military engines and simultaneously to ensure the accuracy of the output surfaces. The effectiveness of the approach is verified by performing statistical tests of significance on the design data. In an effort to reduce the computational burden associated with controller design via optimisation, a response surface method is also considered.

This record has no associated files available for download.