The Giant Magellan Telescope Project relies on a comprehensive integrated modeling (IM) tool to evaluate Observatory Performance Modes (OPM), ranging from Seeing Limited to Adaptive Optics. The development of the integrated model is driven by the need to accurately estimate errors that affect the science instrument data products and mitigate technological risks associated with the telescope. The IM end-to-end simulation models combine structural dynamics, optics, and control models seamlessly in a unified framework. Computational fluid dynamics analysis produces a set of time series representing most of the disturbance sources affecting the telescope performance (namely, dome seeing, wind loads, and structural thermal deformations) under different boundary conditions. Conceiving and managing such a tool imposes several challenges. Firstly, due to the wide range of scientific and engineering expertise required. Furthermore, developing a realistic system representation while dealing with the computational aspects is critical, particularly in adaptive optics OPMs, where the system complexity (vast number of degrees of freedom combining slow and fast dynamic behaviors demanding high sampling rates) can make simulations impractically long. This paper presents the architecture of the GMT integrated model tailored for the Natural Guide Star and Laser Tomography Adaptive Optics OPMs. The features of the computing framework that integrates the domain-specific models into a unified model are also approached. We also show end-to-end simulation results illustrating the interaction between the control loops composing those adaptive optics modes.
|