Towards a Digital Twin for Thermal Processes: Control-centric approach

Manufacturing processes modelling can be complicated, since processes ar in their majority spatially distributed systems. The case of thermal processes such as laser processes is also in this category. Furtermore, specific issues that increase this complexity can be non-linearities (in particular phase change). Herein, under thre framework of creating a control-centric digital twin for thermal (laser-based) processes being able to adapt through measurements and respond to whti-if scenarios, switched dynamic systems are considered. Thus, the Processes are much easier to be controlled through optimal/adaptive/robust controllers. Various modelling issues are discussed and it is shown that dynamics depend highly on the operation phase. Furthermore, closed-loop systems efficency is studied regarding time-related criteria pertaining to tracking. Finally, the possibility of automated decision making on controllers' use is discussed