Real-Time Model Predictive Control of a Laboratory Liquid Tanks System

This paper deals with real-time implementation of Model Predictive Control (MPC) of a laboratory liquid tanks system using the dSPACE platform. The MPC problem is solved using parametric programming techniques, which allow closed-form solution to the underlying optimization problem to be obtained off-line in a form of a look-up table. Once such a table is calculated, the subsequent implementation reduces to a simple set-membership test, which can be performed very efficiently on-line. In the paper we present a step-by-step description of all steps leading towards the derivation and implementation of such a controller for a laboratory device.