Inverse dynamics control of a steam power plant

Abstract

In this paper an artificial neural net is used to control the boiler of a steam power plant. The proposed controller is using a multi-input, multilayer artificial neural network, trained off-line using the error backpropagation algorithm, over a wide range of operating conditions. The data used during the learning phase are calculated by simulating the drum boiler dynamics. The proposed controller has been tested by simulating its online operation, using a medium size boiler model.<>