(652g) Model Predictive Control of Granule Microstructure

At present, particle size tends to be used as the main product attribute for the purpose of particulate processes control. However, in many particulate products there are other attributes ? especially the internal distribution of formulation ingredients and phases within the particles ? that strongly influence the product end-use properties. The ability to control the microstructure-related attributes is possibly more important than the ability to control particle size: while particle size distribution that deviates from requirements can be ?corrected' in the processing plant by the recycling of fines and the milling of over-size, internal attributes of particles are effectively impossible to revert once formed.

Predictive models of particle microstructure formation (Stepanek and Ansari, 2005) and structured particle dissolution (Stepanek, 2004) have been recently developed and used for the solution of the granule microstructure design problem (Ansari and Stepanek, 2006). In the present work we will address the problem of granule microstructure control. First, the control problem will be formulated and the issue of controllability addressed. By means of parametric sensitivity analysis, a reduced model will be derived, which will be used for model predictive control of granule microstructure in a fluid-bed wet granulation process, where the control objective is the dissolution time of the final granular product, the disturbances occur on the feed material properties space (primary particle size distribution etc.), and the manipulated variables are from the process parameters space (granulation process conditions).

References:
Stepanek, F., Ansari, M.A., 2005. Computer simulation of granule microstructure formation. Chem. Eng. Sci. 60, 4019-4029.
Stepanek, F., 2004. Computer-aided product design: granule dissolution. Chem. Eng. Res. Des. 82, 1458-1466.
Ansari, M.A., Stepanek, F., 2006. Design of granule structure: computational methods and experimental realisation. AIChE J. 52, 3762-3774.

/>/>/>