On Increasing the Efficiency of a Cement Clinker Kiln Using Machine Learning

Abstract

The production of cement clinker faces many management challenges, particularly in terms of consistently high product quality, efficient energy usage, and stable furnace operation. In this study, a machine learning model based on gradient boosting was developed for the efficient operation modes of the kiln (required quality and low energy consumption). The influence of process parameters on the efficiency of the clinker kiln was investigated. As a result, it was shown that stable kiln feeding improves the quality of the final product. High feeding variation leads to an increase in the dispersion of the entire setup and attempts to maintain it in a stable state by changing the volume of burned gas. When there is high feeder operation variation, the lime saturation factor has a significant impact on the outcome. The obtained results can be used to create a digital assistant for the kiln operator.