A modified dynamic non-singleton fuzzy logic system for nonlinear modeling

This paper presents a modified version of dynamic non-singleton fuzzy logic systems (NSFLS). Mouzourise and Mendel (1997) originally introduced the NSFLS technique. This technique suffers from the lack of good convergence property. To remedy this and to improve the convergence rate of the NSFLS method, we introduce a fuzzy based algorithm. In order to show how this technique outperforms the dynamic NSFLS, gas turbine has been selected as a case study. Both methods, the NSFLS with the original learning algorithm and with the proposed technique, were applied to this problem and simulation results are very promising