Discrepancy based control of particulate processes

• New control approach for particulate processes.
• Nonlinear control design for continuous fluidized bed spray granulation with external product classification.
• Stabilization of distributed parameter systems with unstable zero dynamics.

This article deals with a new approach to particulate process control. The model system under investigation, the continuous fluidized bed spray granulation with external product classification, is described by a nonlinear partial integro-differential equation, the population balance equation for the particle size distribution. This process exhibits interesting dynamical behavior, i.e. a change of the stability behavior and the occurrence of limit cycles. In addition, the zero dynamics with respect to moment measurements frequently used in practice are unstable in certain parameter regions. In order to stabilize these types of systems in this contribution the use of a generalized distance measure, the discrepancy, is proposed. Applying, the associated stability theory, i.e. stability theory with respect to two discrepancies, a stabilizing control law can be derived. One of the main advantages of the proposed discrepancy based control method is that no model reduction is required.