Fuzzy modeling of a gas turbine engine using clustering and multi-objective optimisation

Using gas turbine engine test data, a fuzzy model is obtained using cluster estimation and multi-objective optimisation. Cluster estimation provides the mechanism for calculating the density of the data distribution and for extracting useful information that can be used as the centres of the fuzzy membership functions. In addition, the multi-objective optimisation approach to the design of a fuzzy model has clear advantages over the conventional single-objective optimisation approach. It provides an understanding of the trade-off relationship between two objectives model fidelity and the number of fuzzy rules used.