Nonlinear MPC using an improved nested partitions algorithm

Non-linear MPC (NMPC) is a technique based on the optimization of a cost function under constraints and this optimization problem is often non-convex. When the problem is non-convex, finding a global solution is required. This paper proposes the improvement of the modified Nested Partitions algorithm to optimize non-convex cost functions applied in NMPC problems. The modifications proposed in the algorithm provides a faster algorithm that is suitable to processes with faster dynamics. The results in a chemical process (CSTR with Vanne de Vusse Reactions) show the efficiency of the proposed method.