Fuzzy modeling of boiler efficiency in power plants

Abstract

Boiler efficiency, an important index that participates in adjusting the combustion process in the boiler, usually cannot be evaluated in real time due to the time-consuming detection process of unburned carbon content in fly ash and cinder. In this paper, we adopt a data-driven soft computing method for measuring the boiler efficiency by incorporating offline fuzzy modeling and online operations. The uncertainty and the non-real-time property of the indirect method are analyzed quantitatively and qualitatively. An improved fuzzy rule extraction method is applied in a 600 MW power plant for boiler efficiency evaluation and real-time control.