Application of regression analysis to optimize hot compaction processing in an indirect solid-state recycling of Mg alloy

An appropriate hot compaction technology is applied not only to obtain high-density blocks from chips, but to consume relatively low energy. In this paper, regression analysis is used to optimize hot compaction processing of machined chips in an indirect solid-state recycling. The nonlinear relation of the temperature, the press and deformation velocity was established according to the rheology of the matrix material using MRA. The lowest energy consumption as criterion is also introduced to further optimize hot compaction parameters in the solid-state recycling. The results based on MRA show that high-density (93%) blocks are obtained according to the built model and that effect of work velocity of the hydrostatic machine on total energy consumption is negligible. During hot compaction, the higher temperature, the more total energy consumption. Besides, void and interface between chips in these hot compaction blocks will be disappeared by extrusion deformation.