Economic model predictive control with transient average constraints

In this paper, we propose an economic model predictive control algorithm which ensures satisfaction of transient average constraints, i.e., constraints on input and state variables averaged over some finite time period. Compared to asymptotic average constraints, this stricter form might be of independent interest in various applications such as the operation of a chemical reactor, where e.g. the amount of inflow during each one hour period must not exceed a certain value. We show that compared to asymptotic average constraints, a different constraint formulation within the optimization problem solved at each time instant is needed, but that certain results such as closed-loop average performance and convergence established for the former setting can be transferred to the case of transient average constraints.