Neural network for evaluating boiler behaviour

Fouling and slagging are some difficulties for the development of biomass as energy potential and to achieve the targets of renewable energy sources utilization. The proper technique to analyze the influence of fouling in a biomass boiler is to monitorize the evolution of heat absorption in heat transfer equipment. Traditional equation-based monitoring techniques have problems to tackle with this complex phenomenon. The objective of this paper is to present the methodology of Neural Network (NN) design and application for a biomass boiler monitoring and point out the advantages of NN in these situations. A combination of traditional methods aided with a NN structure to monitorize the boiler could completely solve the problem. NN monitorizing results show an excellent agreement with real data. It is also concluded that NN is a stronger tool for monitoring than equation-based monitoring. This work will be the basis of a future development in order to control and minimize the effect of fouling in biomass boilers. (c) 2005 Elsevier Ltd. All rights reserved.