Open AccessDEVELOPMENT OF SUITABLE MACHINE LEARNING MODEL FOR A CEMENT PLANT CALCINER
Author(s) -
Prateek Sharma,
Mandava V Rao,
B. N. Mohapatra,
Amit Saxena
Publication year - 2021
Publication title -
international journal of engineering applied science and technology
Language(s) - English
Resource type - Journals
ISSN - 2455-2143
DOI - 10.33564/ijeast.2021.v06i03.030
Subject(s) - process engineering , engineering , artificial neural network , process (computing) , waste management , combustion , cement , carbon footprint , computer science , greenhouse gas , artificial intelligence , materials science , ecology , chemistry , organic chemistry , metallurgy , biology , operating system
Cement industry is one of the largest CO2emitters and continuously working to minimize theseemissions. Use of Artificial intelligence (AI) inmanufacturing helps in reducing breakdowns/ failures byavoiding frequency of startups, reducing fuel fluctuationswhich will ultimately reduce carbon footprint. AI offers anew mode of digital manufacturing process that canincrease productivity by optimizing the use of assets at thefraction of cost. Calciner is one of the key equipment of acement plant which dissociates the calcium carbonate intocalcium oxide and carbon dioxide by taking heat input fromfuel combustion. An AI model of a calciner can providevaluable information which can be implemented in realtime to optimize the calciner operation resulting in fuelsavings. In this study, key process parameters of continuousoperation for a period of 3 months were used to train thevarious machine learning models and a best suitable modelwas selected based on metrics like RMSE and R2value. It isfound that Artificial neural network is best fitted model forthe calciner. This model is able to predict the calciner outlettemperature with high degree of accuracy (+/- 2% error)when validated against real world data. This model can beused by industries to estimate the calciner outlettemperature by changing the input parameters as it is notbased on the chemical and physical process taking place inthe calciner but on real world historical data.