Estimation of ash and calorine value of coal with machine learning

There are two important parameters that determine the quality of coal in coal beds, production processes and usage stages. One of them is the thermal (calorific) value and the other is the ash value released as a result of combustion. Very expensive tools and methods are needed to determine these values. For this reason, Machine Learning method, which is used in many areas, can be used in estimation of calorific and Ash values. Using machine learning, predictions can be made and possible outcomes can be predicted before operations occur. With the help of the predicted results, the lower thermal (LCV) and ash values of the coal are measured more efficiently and quickly, eliminating the occurrence of unexpected results and the system is carried out in a healthy, fast and trouble-free manner with high added value. In this study, the LCV and ash value of lignite coal is estimated by using machine learning method of lignite coal used in thermal power plants. In the study, moisture ratio of coal, original ash, sulfur, volatile matter values were selected as independent variables for the input of the artificial neural network, while LCV and dry ash values were used as dependent variables. In this study, if the lower heat and ash value of lignite coal determined in the laboratory environment is estimated with the developed model, there will be no need for laboratory studies. Therefore, economical, time and labor costs for laboratory studies will be eliminated. With this study, 99% successful results were obtained in estimating the ash and lower thermal (LCV) values. Keywords: Artificial Intelligence, Machine Learning, Deep Learning, Data Mining, Coal Lower Calorific Value