Economic nonlinear model predictive control for mechanical pulping processes

Abstract

In this paper we present an economic model predictive control (econ MPC) strategy for a two-stage (primary and secondary refining) Mechanical Pulping (MP) process. The MP process is a complex multi-input multi-output (MIMO) nonlinear process with strong interactions among the variables. In order to guarantee both stability and convergence of the closed-loop nonlinear MP process, two different econ MPC schemes are proposed: one with penalty on the increment of the input and one with penalty on the offset of the input from its steady-state target. We demonstrate that both econ MPC schemes achieve significant amount of energy reduction in terms of the specific energy consumed by the process while ensuring closed-loop stability and convergence to a nearby steady-state. In addition, we show that more energy reduction can be achieved by using the econ MPC with penalty on the input increment compared with the other scheme. Simulation results also demonstrate the potential benefits of using econ MPC over the standard MPC technique.