Identification of a high efficiency boiler based on neural networks with locally distributed dynamics

This paper considers the application of neural network with distributed dynamics to the identification of nonlinear systems. The main intention is to provide a simulation tool to the development engineer. The identification of the thermal plant is accomplished without any a priori knowledge about the nonlinear structure or the dynamics of the plant. Identification results of two different neural nets are shown and compared. The application area is boilers for building energy management systems.