Abstract: Methods and apparatus for determining the NOx content in a combustion flue gas. Several parameters of the combustion process are measured in the combustion furnace. These parameters are then input to a neural network, which then, based upon the input data, generates output data representative of the concentration of NOx in the flue gas. This result is determined without the use of physical NOx sensors.

Abstract: A method and apparatus for predicting the NOx concentration in flue gas produced by combustion in a furnace. This prediction does not physically require any NOx sensors, and the only measurements required are inputs to a neural network. The required measurements are of combustion pressure, combustion or flue gas temperature, and some datum which is representative of the nitrogen concentration in the flue gas. This method and apparatus results in less expensive, immediate NOx value estimates, in contrast to more expensive, delayed measurements from physical NOx sensors. By limiting the number of input data to 5 or fewer, the noise resulting form too large a number of data is limited. This method and apparatus is applicable to furnaces fueled with fuel oil, natural gas, or a fuel oil and natural gas mixture, as well as to furnaces which use oxygen or oxygen-enriched air as an oxidant.