Optimizing cost and quality of finished cement using near Infra-Red gypsum on line analysis and model predictive control

Abstract

The optimum quantitative addition of gypsum to clinker prior to cement grinding is an optimization opportunity for cement manufacturing, which has not yet been extensively used. Near Infra-Red identifies complex chemical molecular structures of minerals. The drawback of this method is that it measures on the material surface only. Since the gypsum additive is statistically homogeneous, it is a suitable material for Near Infra-Red On Line analysis. This paper presents calibration results of SO3 in gypsum. It also describes how this analysis can be used as input to an optimization algorithm software suite, which is based on model predictive control to arrive at an optimum cement quality within complex plant and cement type specific constraints. This optimization software solution implements closed loop quality control at the minimum overall production costs.