Optimization of a segmented mirror with a global radius of curvature actuation system based on multi-fidelity surrogates

Abstract

Separately-polished segmented mirrors hardly meet the co-phasing surface shape error due to the fabrication difficulties. Therefore, some space-based segmented telescope system such as the James Webb Space Telescope (JWST) uses the global radius of curvature (GRoC) actuation system as an effective solution. A segmented mirror with a GRoC actuation system was optimized in this paper. The high-precision finite element model (FEM) usually has low computational efficiency, and the low-precision finite element model cannot guarantee calculation accuracy. In order to alleviate the conflict between computational cost and calculation accuracy in the optimization of a segmented mirror, multi-fidelity surrogates (MFS) based on sensitivity analysis were proposed. The surrogates were then optimized through the multi-island genetic algorithm (MIGA), and the segmented mirror produces 0.0623λ (λ = 632.8) surface shape error RMS per 1 mm of GRoC is corrected, which met the design requirement. Besides, it can also be deployed to solve other complicated engineering problems.