Skip to main navigation Skip to search Skip to main content

A Data-Driven Multidisciplinary Heterogeneous Distributed Computing Framework Considering Time Consumption Disparity

  • Northwestern Polytechnical University Xian
  • College of Shipbuilding Engineering, Harbin Engineering University

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

While multidisciplinary design optimization (MDO) exhibits theoretical efficacy for complex system design, its practical implementation encounters fundamental challenges due to computationally intensive multidisciplinary analyses. To enhance computational effectiveness and optimization results, this paper presents a multidisciplinary heterogeneous distributed computing framework (MHDCF) that considers the time consumption disparities among different disciplinary evaluations. The framework employs various computational nodes to decouple the multidisciplinary evaluation process into surrogate modeling and distributed disciplinary simulations. Besides, the framework replenishes and allocates samples to different nodes based on time consumption disparity to maximize the utilization of the framework. By applying MHDCF to the MDO of a blended-wing-body underwater glider (BWBUG), the framework’s effectiveness and the necessity for managing time consumption disparities are validated through comparative experiments.

Original languageEnglish
Title of host publicationAdvances in Mechanical Design - Proceedings of the 2025 International Conference on Mechanical Design ICMD 2025
EditorsJianrong Tan, Zhenyu Liu, Weifei Hu
PublisherSpringer Science and Business Media B.V.
Pages631-642
Number of pages12
ISBN (Print)9789819573417
DOIs
StatePublished - 2026
EventInternational Conference on Mechanical Design, ICMD 2025 - Hangzhou, China
Duration: 9 May 202511 May 2025

Publication series

NameMechanisms and Machine Science
Volume204
ISSN (Print)2211-0984
ISSN (Electronic)2211-0992

Conference

ConferenceInternational Conference on Mechanical Design, ICMD 2025
Country/TerritoryChina
CityHangzhou
Period9/05/2511/05/25

Keywords

  • BWBUG
  • Data-driven
  • Distributed computation
  • Multidisciplinary design optimization
  • Time consumption disparity

Fingerprint

Dive into the research topics of 'A Data-Driven Multidisciplinary Heterogeneous Distributed Computing Framework Considering Time Consumption Disparity'. Together they form a unique fingerprint.

Cite this