Inner loops with model predictive control controllers

Abstract

This chapter focuses on the derivation of the mathematical modeling of closed-loop systems that includes optimal controllers in the form of an MPC itself. Here, we formulate the MPC-based reference governor strategy as a bilevel optimization problem, which takes into account the optimization which takes place in closed loops. We cover two approaches that reformulate the bilevel problem to a mixed-integer optimization problem and can be solved in a tractable fashion.