Application of Back Propagation Neural Network to Drum Level Control in Thermal Power Plants

The paper describes the development and testing of a neural network based drum level controller for sub-critical thermal power plant boilers. Experimental data obtained from an operational coal fired power plant (500MW Thermal Power Station, Korba, India) is used to train the neural network. This model proposes a simple training algorithm for a class of nonlinear systems, which enables the neural network to be trained with the output errors of the controlled plant. The only a priori knowledge of the controlled plant is the direction of its output response. Due to its simple structure and algorithm, and good performance, the proposed controller has high potential for handling difficult problems in process-control systems. The Artificial neural networks (ANN) modeling can significantly reduce the frequency of deviations and the degree of deviation of the water level in the drum. The ANN model to be applied for the boiler feed system in the power plant will not only increase the efficiency of the system but also shall considerably reduce the tripping of the power plant.