Comparative Performance of Improved Shrinking Span Fuzzy Logic Controller

Abstract

A comparative study of shrinking span membership function (SSMF) fuzzy logic controller (FLC) relative to conventional FLCs has been presented in this work. The fuzzy logic controller provides a means of converting a linguistic control strategy based on expert knowledge into an automatic control strategy. The SSMFs have different spans for various term set elements in the universe of discourse. This research work has three objectives, first one, is to show the effect of different rule bases on the performance of FLC with consideration of simple rule base, initial rule base, Brae Rutherford rule base and Macvikar Whelan rule base. Second objective is to show the effect of shrinking factors on the performance of SSMF - FLC. Third one addresses the comparative study of SSMF - FLC relative to conventional FLC. SSMF-FLC was found to yield better response due to flexible design feature of span adjustment in formulating membership function. Results also indicate that SSMF-FLC is not robust and give unacceptable steady state errors for some test cases based on simulation studies done using C++ and MATLAB. A modification was proposed in implementing FLCs by introducing a gain block in feedback path to improve the steady state error. Simulation results shows that with the additional designed parameter, required refinement in performance can be achieved