Model Predictive Control of hybrid systems with discrete and continuous inputs

Abstract

This paper proposes and discusses a model predictive control approach to hybrid system, with discrete and continuous inputs. The hybrid model of two tank system is translated into mixed logical dynamical (MLD) form, which considering all constraints in the physical plant, such as maximum (flow in the pomp) and maximum level in the tank. The resultant system is well-posed for Model Predictive Control (MPC) that is used as control strategy for the system. Simulations are performed using the Hysdel compiler to illustrate the efficiently of this formalism and prove the influence of increasing the prediction horizon.

Fault-Tolerant Model Predictive Control of Hybrid Systems

“…In fact the addition of continuous auxiliary variable leads to increased inequality constraints as an auxiliary variable can be converted into linear inequalities following 4 equations (16) We can define the MLD fault as follow:…”

“…for more detail see [16]. In order to take into account the flows through the upper valve V4, we define the auxiliary binary variables indicating whether the level in each tank has reached hv: where:…”