TP 78 An Interactive Genetic Algorithm and Machine Learning approach for validating the effect of preheated air combustion on fuel consumption of oil-fired industrial furnace

This paper presents an interactive approach of Genetic Algorithm and Machine Learning being used as tool for optimizing the Fuel Consumption of oil-fired rotary furnace operating with 20% preheated excess air. The author has attempted to solve a crucial problem of optimizing fuel consumption in ferrous foundries. Based on experiments carried on a self-designed and developed 200 kg rotary furnace in a foundry, it was found that crucial operating parameters like preheated excess air not only affects the flame temperature and melting rate but also significantly the fuel consumption An attempt has been made to establish a correlation between fuel consumption as output parameter and all others as input parameters It is expected that this model may prove practically beneficial for industry by estimating the effect of input process parameters for predicting the fuel consumption The results of Genetic Algorithm and Machine approach correlates with the experimental results