Fertiliser granulation involves a recycle process. As such the process has an equilibrium. How this equilibrium is obtained, and the effects of any disturbances to such an equilibrium, are of interest to the Plant Operator. He wishes to both predict situations in unknown conditions and also optimise plant performance.
In a recycle process iterative calculation techniques are required, and thus a computer simulation is an effectual method of tackling the problem.
In UKF two computer simulation models were built: one of a CAN-plant at Geleen (NL) and one of an NPK plant at Ince (GB). The basic building blocks of these models are similar. In the paper the methods of modelling the screens, crushers and granulator are described. The model of the granulator is based on an extension of the Sherrington agglomeration equation. The actual calculation procedure involving heat and water balances is also given.
A section comparing actual plant data and computer predictions shows that making good comparisons is difficult, largely through the known difficulties of collecting reliable plant data. The models are considered to give a good qualitative representation if not entirely satisfactory quantitatively.
Theoretical studies using the models are also described. The relative sensitivity of different plant variables is assessed. One conclusion is that performance of the granulator has a much more marked effect than that of screens and crushers.
The benefits and shortcomings of the models are also treated. One of the principal benefits is the plant personnel can gain a far better understanding of how their plant works; other benefits include defining bottlenecks; predicting the effects of changing screen cloths, etc. The major shortcoming of the models lies in the simplified model of the granulator. A better understanding in quantitative terms of the granulation process will be needed before the model can be improved. Other potential improvements lie in a better description of the relative velocities of the solids flows around the recycle loop.
S M P Mutsers, H J M Slangen, H J J Rutten and I K Watson, DSM/UKF Fertilisers Ltd.
56 pages, 24 figures, 4 tables, 8 refs.