Estimation of Nox Emissions in Thermal Power Plants Using Eng-Genes Neural Networks

Abstract

This paper investigates neural network based estimation of NO x emissions in a thermal power plant, fed with both oil and methane fuels. Two types of neural network namely a novel 'eng-genes' architecture and a Multilayer Perceptron (MLP) have been developed, both being optimised using genetic algorithms. Due to the local nature of the NO x generation process, operational information on the burner cells of the combustion chamber has been considered. Neural networks, with different numbers of hidden nodes have been tested on a set of three-dimensional data of the simulated combustion chamber. It is shown that, the proposed 'eng-genes' neural network can produce accurate estimations with better generalisation performance than MLP.

Model based NOx emission monitoring in natural gas fired hot water boilers

“…(Ikonen et al, 2000) used neuro-fuzzy modelling for fluidised-bed combustion (FBC) process. Eng-genes and MLP neural networks were applied to a power plant fed with oil and methane (Li et al, 2005). An ensemble model consisting of fuzzy clustering, least squares support vector machine (LSSVM) and partial least squares (PLS) were applied by (Lv et al, 2013) to a coal boiler.…”

Control of the Boiler Swing Rate for NO Emission Minimization

“…An estimation of NO x emissions in a thermal power plant that is based on neural network was studied. 11 Two types of neural networks, namely, the multilayer perceptron (MLP) and a network with different numbers of hidden nodes, have been trained and validated.…”

“…With regard to the modeling of the prediction of NO x formation from fuel-fired boilers, the number of methods can be found in the literature. The neural networks are a commonly used tool. − Parametric data, gathered from previous boiler testing, were used to train several different neural networks for NO x and unit heat rate . Computer experiments were conducted to determine the best means of maximizing the ability of the neural network to predict boiler output responses when given different input data.…”

“…The NO x and heat rate neural networks produced accurate results in predicting boiler response values at full-load conditions. An estimation of NO x emissions in a thermal power plant that is based on neural network was studied . Two types of neural networks, namely, the multilayer perceptron (MLP) and a network with different numbers of hidden nodes, have been trained and validated.…”