This code presents the design of an adaptive Linear Parameter Varying Model Predictive Control (LPV-MPC) scheme for chemical industrial processes. The proposed LPV-MPC method resides in the operation of two consecutive Quadratic Programing (QP) problems: a Moving Horizon Estimator (MHE) and a regular Model Predictive Controller (MPC). The MPC includes a terminal set constraint to ensure recursive feasibility. The terminal set is determined offline through an Linear Matrix Inequalities (LMI) problem.
Simulation results, based on a Solar Thermal (ST) collector is given to illustrate the implementation and performance of such approach.
- At least an i5-3337U [email protected] GHz (2 Cores) with 6 GB of RAM.
- Matlab software R2016b or greater
Bernardi, Emanuel, Hugo Pipino, Carlos A. Cappelletti, and Eduardo J. Adam. "Adaptive Predictive Control for Industrial Processes." In 2021 XIX Workshop on Information Processing and Control (RPIC), pp. 1-6. IEEE, 2021.
Send comments and questions to: