Akademska digitalna zbirka SLovenije - logo
VSE knjižnice (vzajemna bibliografsko-kataložna baza podatkov COBIB.SI)
  • New Robotic Telescope enclosure concept selection and optimisation [Elektronski vir]
    Gradišar, Luka, gradbenik, 1995- ...
    The New Robotic Telescope (NRT) is a robotic and fully autonomous four-metre-class telescope and the first in its size class to utilise a clamshell enclosure. It will be located at the Roque de los ... Muchachos Observatory (ORM) on La Palma, Canary Islands, Spain. Fast slew time, robotic functionality, and reduced dome seeing are the main reasons for the clamshell design, however cost is also an important factor. The greatest opportunity for cost reduction is the movable roof structure of the enclosure, but is the most complex to design and the heaviest part in the initial concept considerations. To solve a complex optimisation problem considering all limitations, conditions and assemblies, a combination of generative design approach and machine learning is used. This enables us to overcome two major obstacles: First, we are able to combine multiple models into one optimisation problem that could analyse multiple states simultaneously, such as the closed and semi-open states. Second, the machine learning-based predictive models are able to run much faster, which allows us to explore many more possible design solutions. Structural optimisation results show more than one optimal solution, which is consistent with multi-objective optimisation, since there are certain trade-offs between mass, capacity utilisation, and hydraulic forces. The goal of the structural optimisation was to explore the possible design alternatives before engaging a construction design partner. This allows for more efficient project development, as the design partner could immediately focus on the construction design work without needing to understand the implications for the telescope.
    Vrsta gradiva - prispevek na konferenci ; neleposlovje za odrasle
    Leto - 2022
    Jezik - angleški
    COBISS.SI-ID - 126971907