An Approach to Optimal Control of the Combustion System in a Reverberatory Furnace

Abstract

In this work an optimal control technique is applied to control the excess air in the combustion process of a reverberatory furnace by the monitoring of O2 percentage in the stack, the controller is designed based on a nonlinear model estimated by artificial neural networks and a data base is used for the previous training; the data base has two subsets one for training and other to validate the net.