ABSTRACT
The present work demonstrates the real time predictions of the performance characteristics of a 210 MW boiler using artificial neural network (ANN) approach. ANNs have been successfully applied to model the different characteristics such as power, outlet steam pressure, and efficiency of a coal-fired water tube boiler. The training and testing data to develop the ANN were obtained through a number of experiments from the boiler of a thermal power station. Back propagation algorithm was utilized to train the ANN. The inputs considered were coal flow rate, percentage of load, and steam flow rate. The application of the developed model exposed better results in terms of regression value (R2), accuracy, percentage error, and mean square error. The co-efficient of multiple determination value was 0.99874 for the developed model which is very close to 1. It is perceptible that the developed model could be a better tool in predicting the performance of the boiler. In addition, modeling was also done based on radial basis network and compared with the back propagation network. Comparison of results indicates that the back propagation algorithm network performs better than radial basis network.
Correction Statement
This article has been republished with minor changes. These changes do not impact the academic content of the article.