Dynamic Safety Constraints by Scenario Based Economic Model Predictive Control

Abstract

This paper studies use of scenario based model predictive control (MPC) to control a plant into a fault-tolerant state set. For some systems it is not always sufficient that the controller is reconfigured after a fault, it is also necessary that the system is in a given state set when the fault occurs in order to guarantee that the constraints are not violated. This is often achieved by using static safety constraints. However, it may be hard to find such constraints, and they may be conservative. This article presents a method for implementing dynamic safety constraints based on fault scenarios with an economic MPC. The controller is tested using closed-loop simulations.