Application of neural modelling in hybrid control model of fluidized bed boiler fired with coal and biomass

The energy diagnosis of power plants and combined heat-and-power plants (CHP) is a very important form in the point of view of rational management of primary energy resources as well as from the economical point of view. The most irreversible component of a power plant is the boiler and the most important factor influencing the energy efficiency is the boiler flue gas temperature. The paper presents two methods of identification of empirical models describing this parameter - regression and neural networks.