Nonlinear Model Predictive Control (NMPC) has become the accepted methodology to solve complex control problems related to process industries. The main motivation behind
explicit
NMPC is that an
state feedback law avoids the need for executing a numerical optimization algorithm in real time. The benefits of an
solution, in addition to the efficient on-line computations, include also verifiability of the implementation and the possibility to design embedded control systems with low software and hardware complexity.
This book considers the multi-parametric Nonlinear Programming (mp-NLP) approaches to
approximate NMPC of constrained
nonlinear
systems, developed by the authors, as well as their applications to various NMPC problem formulations and several case studies. The following types of
systems are considered, resulting in different NMPC problem formulations:
Ø
Nonlinear
systems described by first-principles models and
systems described by black-box models;
-
systems with continuous control inputs and
systems with quantized control inputs;
systems without uncertainty and
systems with uncertainties (polyhedral description of uncertainty and shastic description of uncertainty);
systems, consisting of interconnected
sub-systems.
The proposed mp-NLP approaches are illustrated with applications to several case studies, which are taken from diverse areas such as automotive mechatronics, compressor control, combustion plant control, reactor control, pH maintaining system control, cart and spring system control, and diving computers.