BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Memento EPFL//
BEGIN:VEVENT
SUMMARY:MEchanics GAthering –MEGA- Seminar: Enabling Robust and Integrat
 ed Design of Gas-Bearing Supported Turbocompressors Through Neural Network
  Ensemble Surrogate Models
DTSTART:20240418T161500
DTEND:20240418T171500
DTSTAMP:20261024T002632Z
UID:72fe1aa933b6a31df1c801877c060d85ce86524159454b9fc86159a6
CATEGORIES:Conferences - Seminars
DESCRIPTION:Soheyl Massoudi (LAMD\, EPFL)\nAbstract: Modern engineering 
 challenges in the design of small-scale turbocompressors call for a multi-
 disciplinary approach\, necessitating the consideration of robustness agai
 nst manufacturing imperfections alongside traditional performance metrics.
  This presentation introduces a framework utilizing ensembles of neural ne
 tworks to create surrogate models\, significantly enhancing the robustness
  and integration in the design process of gas-bearing supported turbomachi
 nery.\nWe examine the limitations of conventional design optimization\, wh
 ich focuses on component-level metrics rather than holistic system robustn
 ess. Our work demonstrates that maximizing nominal performance metrics oft
 en fails to account for the manufacturing realities\, potentially compromi
 sing system reliability and safety margins.\nIn response\, we propose an a
 utomated\, multi-objective optimization framework that incorporates robust
 ness as a core design objective\, evaluated against geometrical and operat
 ional deviations. This approach is empowered by the speed and accuracy of 
 neural network surrogates\, which replace time-intensive high-fidelity mod
 els\, expediting the design process while maintaining precision.\nCase stu
 dies on micro-turbocompressor applications reveal clear trade-offs between
  efficiency\, operational range\, and robustness. The application of our m
 ethodology to herringbone grooved journal bearings underlines the importan
 ce of integrated design\, offering practical design guidelines and an inte
 ractive design and simulation tool\, DARTS-NETGAB\, which harnesses real-t
 ime analysis and 3D construction capabilities for engineers and researcher
 s.\nThis research embodies a substantial stride forward in the domain of t
 urbocompressor design\, advocating for a paradigm shift towards integrated
  optimization and robustness. The seminar will discuss the implications a
 nd applications of the proposed design framework in enhancing the resilien
 ce and performance of turbocompressors.\n\nBio:Soheyl Massoudi is a Ph.D. 
 candidate at the Laboratory for Applied Mechanical Design (LAMD)\, EPFL. H
 e joined the lab after doing his B.Sc. and M.Sc. specializing in fluid dyn
 amics with a minor in energy at the EPFL. His research focuses on design m
 ethodology via surrogate-model assisted optimization using artificial neur
 al networks and evolutionary algorithms. The target application is gas-bea
 ring supported turbocompressors for heat pumps and fuel cells.
LOCATION:MED 0 1418 https://plan.epfl.ch/?room==MED%200%201418 https://epf
 l.zoom.us/j/67041786969?pwd=a1lXa3lsVGpvL1VpN2RDR2l4clg0QT09
STATUS:CONFIRMED
END:VEVENT
END:VCALENDAR