A cascaded neural network and its application to modelling power plant pollutant emission

Pollutant emission in power generation plant, especially NOx emission, has attracted much attention in the last ten years. For an operating plant, development of an advanced control system is essential for pollutant emission reduction. System modelling is an important stage in emission control. In this paper, a cascading neural network is used to model the NOx emission in a coal-fired power generation plant. This type of neural network has more connections than that found in the layered feedforward neural network. Algorithm for training this type of neural network is suggested and then it is used to build a NOx emission model for a coal-fired power generation plant. Simulation results show the merits of this type of neural network.