Development of identification based fuzzy supervisory control for pressure tank system

Abstract

This paper presents the development of identification based fuzzy supervisory control for application to pressure process. The system identification is carried out using an integral based method that reformulates the differentials with an integral, that provides robustness to noise. A theoretical analysis of the identification method is also provided. Pressure data was obtained through the use of a pressure guage which is connected to a workstation via a DAQ. The collected data was then identified using the developed identification algorithm. Appropriate PID gains are firstly selected by pole placement methodology, where they are later modified with fuzzy rules according to the desired responses. Results show that the designed fuzzy supervisory control outperformed the constant PID gains which was designed by pole placements.