Second-order consensus predictive control for 360MN extrusion machine

Abstract A new consensus predictive control protocol is proposed in this paper. It significantly speeds up the convergence rate comparing to traditional consensus mechanism and has been proved working well in simulation platform. A mathematical model is built by simplifying the hydraulic cylinders to single-DOF agents distributed evenly around 360 MN extrusion machine׳s main table. With this model, a theoretical analysis of the mechanism is given and the validity of the consensus predictive control protocol for strong coupling multi-agent system on ADAMS–EASY5 simulation platform is illustrated. The machine׳s main table balance problem is well solved and the convergence rate is increased. These results are meaningful for increasing productivity.