Optimal performance of modular and synchronized mechatronic systems

Abstract

Optimizing the configuration and overall performance of synchronized modular systems is considered in this paper. The synchronized modules can be considered as a hybrid system, including continuous-time dynamics of local moving devices, combined with high-level discrete event sequences. The continuous-time trajectories are approximated by the Gauss pseudospectral method, resulting in a nonlinear programming (NLP) problem. The optimal configuration generates the maximal production rate subject to dynamic constraints. A complete design procedure is presented and applied to a case study of a packaging machine, where an alternative optimal configuration is achieved compared to current industrial practices.