Department of

Chemical Engineering

Designing molecular technology for the 21st century with biology and chemistry


 


Associate Professor Antonios Armaou | Research




CBMPC: A Nonlinear Predictive Control approach

The issue of nonlinear control is one that has received extensive focus by the chemical engineering community in the past, offering a balance between computational feasibility and optimal with respect to certain criteria control action. State input constraints are widely prevalent in current industrial practice and this motivates a need for controller design methodologies which explicitly address these process constraints. Model predictive control (MPC), also known as receding horizon control, is one such powerful tool for handling these process constraints with an optimal control setting.

The control action in MPC is calculated by solving repeatedly, online, a finite-horizon open-loop optimization problem at each sampling time. As the control action is computed online during process evolution, MPC has the capabilities to suppress the external disturbances and tolerate model inaccuracies (using feedback control) during the course of forcing the system to follow certain optimal path that respects the process constraints. Extensive reviews on various MPC formulations along with their corresponding control-relevant issues such as closed-loop stability, performance and constraint satisfaction can be found in open literature.

Although there are substantial advantages of MPC designs especially to industrial community, the stability guarantees offered in MPC are linked to the feasibility of the optimization problem and moreover the set of initial conditions starting from where a given MPC formulation is guaranteed to be feasible is not explicitly characterized. Our main focus lies in the development of a computationally efficient method to identify the optimal control action with respect to predefined performance criteria. In our approach optimal control problems are reformulated as nonlinear optimization ones with analytically computed sensitivities.

Such problems can be efficiently solved using standard, gradient-based, search algorithms. Our method lies at the interface between collocation and shooting methods, since the states are discretized explicitly in space and time and their sensitivity to the control action is analytically computed, reminiscent of collocation methods, while the states now enter the optimization problem explicitly as a nonlinear function of the control action and are eliminated from the equality constraints, thus reducing the number variables, evocative of shooting methods.


Page created on 02/14/11,   Top of page