Mechanism reliability analysis for multi-support axis seizure with assembly tolerance

Hui Wang, Tianxiang Yu, Huan Pang, Bifeng Song

Research output: Contribution to journalArticlepeer-review

Abstract

The deflection of the supports, which is caused by the external forces or the tolerances in the manufacture and assembly procedure, will induce seizure of the mechanism with multi-support axis. Only few researchers have investigated the seizure reliability of mechanism, and they focused on the simplified models of some simple mechanisms. In recent years, the reliability analysis based on computer simulation technique is widely adopted to improve the reliability of the complex mechanism in practice, and the reliability analysis is fulfilled by introducing the random parameters to the mechanism model in the simulation. In this paper, the reliability analysis for the multi-support axis seizure of a door mechanism is investigated, which is caused by the assembly tolerance. The rigidflexible coupling model of the door mechanism is built by the commercial software LMS Virtual.Lab. The parameterized expression of the assembly tolerance is introduced by the advanced mesh transformation so called mesh morphing technique. The results corresponding to the failure mode are obtained based on plenty of simulations using this parametric model. Then, the approximate model for the limit state function is generated using the support vector regression machine based on the results. Finally, the reliability analysis and evaluation for the multi-support axis seizure of the door mechanism are achieved using the approximate mode.

Original languageEnglish
Pages (from-to)5667-5674
Number of pages8
JournalInformation (Japan)
Volume15
Issue number12 B
StatePublished - Dec 2012

Keywords

  • Mechanism reliability
  • Multi-supports
  • Seizure
  • Simulation
  • Support vector machine

Fingerprint

Dive into the research topics of 'Mechanism reliability analysis for multi-support axis seizure with assembly tolerance'. Together they form a unique fingerprint.

Cite this