Experimental thermal regulation of an ultra-high precision metrology system by combining Modal Identification Method and Model Predictive Control

A Reduced Model (RM) identified from measured data using the Modal Identification Method (MIM) was combined with a Model Predictive Controller (MPC). The control was applied to minimize the effects of thermal drifts on the principal organ of the cylindricity measurement machine. A parametric study of the MPC was initially conducted to evaluate the robustness of the controller. The association of both RM and MPC allowed significant reduction of thermal drifts generated by the three heating wires, which simulate powers dissipated by the mechanical guiding system. The obtained results can be considered as promising with regard to applications in dimensional metrology.