par Nielsen, Carl Johan G. 
Président du jury Hendrick, Patrick
Promoteur Kinnaert, Michel
Co-Promoteur Preumont, André
Publication Non publié, 2026-06-12
Président du jury Hendrick, Patrick
Promoteur Kinnaert, Michel
Co-Promoteur Preumont, André
Publication Non publié, 2026-06-12
Thèse de doctorat
| Résumé : | In this work, we test and characterize a multilayer actively controlled polymer space telescope, considered as a possible future paradigm for deployable optical payloads in small-satellite constellations. The demonstrator and breadboard is manufactured in cooperation with MateriaNova (Mons, Belgium) under the Advanced Multilayer Active Thin Shells (AMATS) project funded under the ESA General Support and Technology Programme (GSTP).In order to characterize the reflector, we build a custom implementation of the Software Configurable Optical Testing Method (SCOTS), using an integration method designed for the petal geometry based on a vibration mode expansion of the reflector. Raytracing validates the method to sub-micron residual accuracy. The method is potentially applicable to future on-orbit metrology as well, using a different instrumentation.Finite Element Modelling simulations of the reflector are conducted to estimate the closed loop performance under representative aberrations. We find that for the 36-electrode demonstrator, a reduction of one to two orders of magnitude in rms error can be achieved, depending on the aberration complexity. For the thermal environment, the symmetrically balanced multilayer layup is relatively resistant, but sensitive to the CTE of the clamping mechanism.Finally, we construct a full electromechanical breadboard to measure the influence functions of the reflector, capable of simultaneous control of all electrodes, although with an initial aberration outside control stroke due to an astigmatism error associated with the longitudinal direction of the PEI substrate. We find excellent agreement between measured and simulated influence functions: for a remapped astigmatism, the residual rms error is 0.77 µm and 0.78 µm for measured and simulated actuator responses respectively. We find significant variance in influence function amplitude, which may be explained by variance in the manufacturing or the anisotropy of the Polyetherimide (PEI) substrate. Coupon testing confirms a surface roughness of σ ≈ 14 nm and no significant degradation of piezoelectric performance after 20 krad irradiation, supporting the applicability to space missions in the long-wave infrared. |
