ANN prediction tool for ReHeater and SuperHeater sprays in boiler performance

Artificial Neural Networks, as a paradigm, is extremely relevant in the present day context where data obtained from processes is plagued by uncertainty and insufficiency. Hybrid prediction techniques for process control systems are the order of the day, which involve a combination of data driven mo...

Full description

Saved in:
Bibliographic Details
Published in2011 3rd International Conference on Electronics Computer Technology Vol. 6; pp. 335 - 337
Main Authors Madhavan, K. S., Prasanna, P., Varman, T., Dhanuskodi, R., Arumugam, S.
Format Conference Proceeding
LanguageEnglish
Published IEEE 01.04.2011
Subjects
Artificial Neural Network
Artificial neural networks
Boilers
General Regression Neural Network
Genetic algorithms
Graphical user interfaces
Hybrid System
Predictive models
ReHeater Spray
Smoothing methods
Soft Computing
SuperHeater Spray
Online AccessGet full text

Cover

More Information
Summary:Artificial Neural Networks, as a paradigm, is extremely relevant in the present day context where data obtained from processes is plagued by uncertainty and insufficiency. Hybrid prediction techniques for process control systems are the order of the day, which involve a combination of data driven models and knowledge driven models. In this paper an Artificial Neural Network prediction tool has been generated with Visual Basic GUI to predict the spray values in a 500 MW boiler within permissible tolerances. The prediction of sprays is done using General Regression Neural Network (GRNN), smoothing factors of which have been generated using a Genetic Algorithm. The General Regression Neural Network predicts the ReHeater Spray and SuperHeater Spray from the input combination of Burner Tilt, Mill Combination, Excess Air Percentage and Load.
ISBN:1424486785
9781424486786
DOI:10.1109/ICECTECH.2011.5942110