Dynamic Operability for the Calculation of Transient Output Constraints for Non-Square Linear Model Predictive Controllers

Fernando Lima and Christos Georgakis
Tufts University


Abstract

This paper introduces a dynamic operability-based approach for the determination of feasible output constraints during transient operation. This approach is based on previously published steady-state operability developments and the concept of output funnels. In this study, high-dimensional non-square systems with more outputs than inputs are of particular interest. Such systems are challenging because it is impossible to control all the outputs at specific set-points when there are fewer degrees of freedom available than the controlled variables. Thus, interval, instead of set-point, control is needed for at least some of the output variables. In order to motivate the new concepts, two non-square case studies are addressed, one illustrative and one industrial - obtained from the control system of a Steam Methane Reformer process. The calculated constraints are validated by running DMCplusTM (AspenTech) closed-loop simulations for the extreme values of the disturbances. These constraints are intended for use online in model-based controllers (e.g., Model Predictive Controllers) to ensure that each of the outputs will remain inside a feasibility envelope during transient operation.