Have confidence in your coronagraph: statistical analysis of high-contrast coronagraph dynamics error budgets

Abstract

We have combined our Excel-based coronagraph dynamics error budget spreadsheets with DAKOTA scripts to perform statistical analyses of the predicted dark-hole contrast. Whereas in the past we have reported the expected contrast level for an input set of allocated parameters, we now generate confidence intervals for the predicted contrast. Further, we explore the sensitivity to individual or groups of parameters and model uncertainty factors through aleatory-epistemic simulations based on a surrogate model fitted to the error budget. We show example results for a generic high-contrast coronagraph.

Numerical modeling of the proposed WFIRST-AFTA coronagraphs and their predicted performances

“…The resulting change in the electric field at the image plane (a complex value) is recorded for each. These fields are used in an error budgeting program 23 that allocates contrast performance contributions to various system properties (position stability of each optic, pointing offsets, etc.). The impact of a position error in one optic, for example, can be estimated by calculating via ray tracing the resulting low order aberrations, reading in the corresponding fields, adding them together and converting to contrast.…”