Multiobjective design of a fuzzy controller for a gas turbine aero-engine

Abstract

This paper describes the design of a fuzzy logic controller for a gas turbine aero-engine based on multiobjective genetic algorithms. The design is for a single manoeuvre around a 50% operating point at sea-level static conditions. The input-output relationship of the original controller is observed for the manoeuvre and a fuzzy controller constructed to approximate this relationship and verified on a nonlinear thermodynamic of the model engine. The fuzzy rule-base and membership functions are then optimized for a representative set of performance and operational design criteria directly on the nonlinear model and compared with the original controller