IJSRD - International Journal for Scientific Research & Development| Vol. 3, Issue 08, 2015 | ISSN (online): 2321-0613

Weathering is a continuous process; the best way to deal with weathered coal is to identify the extent of weathering. Based on level of weathering its consumption pattern should be changed, this will help in minimizing the effect of weathering on coke quality. Due to weathering, various technological properties of coal get altered such as as ash, free swelling index, fluidity, plastic range and size degradation. There are various methods to monitor coal weathering. Percentage transmittance determined by alkali extraction test is one of the ways to monitor weathered coal. This paper discusses the use of artificial neural network (ANN) to predict the percentage transmittance. ANN model was developed using the experimental data and R Square values for training, testing and validation are 0.99, 0.95 and 0.98 respectively. It means the developed model can successfully predict the deterioration trend of coal weathering. It will help to take suitable action to minimize the effect of coal weathering on coke quality.