A CAPE-Taguchi combined method to optimize a NPK fertilizer plant including population balance modeling of granulation-drying rotary drum reactor

Abstract In this work, a detailed model of a real NPK fertilizer plant has been developed to optimize operation strategies, in order to reduce sensibly the thermal and electrical consumptions, as well as the recycle ratio. Granulation is a very complex process that involves changes in the particle properties and depends on several factors as drum speed and design, binder conditions and material properties. Furthermore, simultaneously to granulation process, a drying process takes place. In consequence, due to its complexity, a novel comprehensive semi-empirical population balance model of an industrial co-current rotary drum reactor has been developed to model the PSD evolution, including granular dynamics and the particle growth by granulation and the size reduction due to the effect of particle-particle and particles-bed attrition. The granulation model is coupled with a drying model, taking into consideration mass and energy balances, including combustion and the bed and reactor heat and mass transfer mechanisms. Granulator model (developed in Matlab) has been coupled to auxiliary equipment modelled in AspenPlus by means of MSExcel intermediate files. Global and local models have been validated with plant data. Additionally, a robust Taguchi Design of Experiments (DoE) has been applied to evaluate the effect of each process parameter in the Key Performance Indicators. From the results of the analysis, it is possible to discriminate what process parameters are critical in the quality of the product and in the use of the resources.