Advanced Identification and Control for Thermal Systems

Compared to modelling and control for positioning problems, advanced thermal modelling and control has received considerably less attention so far.  The aim of this project is to develop advanced methods for experimental modelling (i.e. identification) and feedback control for thermal problems.

PhD Candidate: ir. Enzo Evers
Supervisor: Bram de Jager, Tom Oomen
Promoter: Maarten Steinbuch
Project Financing: Impulse II
Project Period: October 2016 - September 2020

In recent years several industrial parties have identified the need for further development of the thermal control competence. In academia, but especially in the Eindhoven region, position control has been one of the main research subjects within the mechatronics community. However, many applications within the fields of precision manufacturing, lithography and electron microscopy have experienced significant thermally induced deformations. Due to the advancements in position control these deformations are no longer negligible in the full system performance.

The state of the art for thermal modelling and management is often based on Finite Element Methods to analyze the sensitivity to thermal disturbances and a wait-and-see approach to reach a thermal steady state. The resulting FEM models are often unsuitable for design optimization and control applications due to high number of states, order 10^6 is not uncommon, and mismatch to the true physical system. 

The aim of the current project is to use experimental modelling to construct low-order models that are suitable for control applications by utilizing the existing knowledge on system identification developed for mechanical systems. Important nuances such as excitation signal design, actuator constraints and measurement times will be considered.