Application of fuzzy logic to control of gas composition in a combustion process

Abstract

A fuzzy logic-based multi-input/single-output controller has been designed for controlling the gas composition in a combustion process. The combustion process pyrolyzes alumina fiber for utilization in a metal matrix composite. The exothermic reaction, which affects the quality of the alumina fiber, can be indirectly controlled through control of the oxygen concentration. Mass flow controllers which inject and exhaust various gas species to the furnace are used to control the gas composition. The process is controlled thermally through the use of heating elements. Because of the complexity of the combustion process, the use of a fuzzy controller was investigated for control of the gas composition. Steady-state fluid flow models were utilized to define the appropriate membership functions and control rules. The performance of the controller is demonstrated using online process data of the combustion process.<>