Machine-learning based design of nearspherical shells under external pressure

Mitansh Doshi, Xin Ning

Research output: Chapter in Book/Report/Conference proceedingConference contribution


In this paper Extreme Gradient Boost (XGBoost) based machine learning model is designed to predict the critical buckling load of near-spherical composite shells (Icosahedron). Icosahedron is under external pressure, and the effect of changing the geometry or the composite layups on the buckling load is studied in this paper. While designing a composite structure, the engineering design space is often very large. Finding possible combinations to obtain the higher buckling load could be time consuming and computationally expensive. To overcome this problem, a data-driven machine learning model is created in this paper based on the data generated from detailed finite element analyses. Based on the geometry design parameters or material design parameters, the current model predicts the buckling load with excellent accuracy. To verify and test the model an independent test data set is created for each case and then the correlation value (R2 value) or average Root Mean Square Error (RMSE) is computed.

Original languageEnglish (US)
Title of host publicationAIAA Scitech 2021 Forum
PublisherAmerican Institute of Aeronautics and Astronautics Inc. (AIAA)
Number of pages12
ISBN (Print)9781624106095
StatePublished - 2021
Externally publishedYes
EventAIAA Science and Technology Forum and Exposition, AIAA SciTech Forum 2021 - Virtual, Online
Duration: Jan 11 2021Jan 15 2021

Publication series

NameAIAA Scitech 2021 Forum


ConferenceAIAA Science and Technology Forum and Exposition, AIAA SciTech Forum 2021
CityVirtual, Online

ASJC Scopus subject areas

  • Aerospace Engineering


Dive into the research topics of 'Machine-learning based design of nearspherical shells under external pressure'. Together they form a unique fingerprint.

Cite this