Experimental comparative analysis of adaptive fuzzy logic controllers

Abstract

Conventional control depends on the mathematical model of the plant being controlled. When this model is uncertain, intelligent controllers, like fuzzy logic controllers (FLCs), promise better performance. FLC requires expertise knowledge of the process operation for FLC parameter setting, and the controller can be only as good as the expertise involved in the design. To make the controller less dependent on the quality of the expert knowledge, we investigate different adaptation schemes to compensate for this deficiency and propose practical adaptive fuzzy logic controllers (AFLCs). While most intelligent controller's effectiveness is proven only using simulation, we aim in this paper to compare the conventional control to FLC and AFLC experimentally. This can be achieved by constructing a hardware station comprising a plant and implementing different control algorithms for the same load conditions or disturbances.