Download PDFOpen PDF in browser

A Hybrid Methodology for the Determination of System Level Sensitivities Employing Multi-Body Co-Simulations of Mechatronic Systems

EasyChair Preprint no. 13601

2 pagesDate: June 7, 2024

Abstract

Over the years, it has been established in both research and in industry, that the design and modelling of mechatronic systems is a complex process. Therefore, many commercial software tools have been developed that enable accurate and rapid modelling of these systems. Using these system models, it is possible for the designers to simulate the behaviour of the system beforehand without the need for accurate prototypes, which reduces the cost to market. However, the increasing demand for optimized designs of complex mechatronic systems entails a need for novel modelling and optimisation tools to aid the designers in the design process. This work proposes an approach to enable design optimisation and parameter identification, through a novel method to compute system level sensitivities. Determining these system level sensitivities for mechatronic systems is no simple task, as mechatronic systems, by definition, consist of multiple different types of system models where it is common that a single design parameter influences different engineering domains included in the mechatronic system. In such a case, it is needed to obtain the local sensitivities from each of the engineering domains or submodels and combine them into a system level sensitivity that represent the influence of this design parameter on the full mechatronic system.

Keyphrases: Mechatronics, multibody co-simulation, sensitivity analysis

BibTeX entry
BibTeX does not have the right entry for preprints. This is a hack for producing the correct reference:
@Booklet{EasyChair:13601,
  author = {Daan Bortels and Simon Vanpaemel and Frank Naets},
  title = {A Hybrid Methodology for the Determination of System Level Sensitivities Employing Multi-Body Co-Simulations of Mechatronic Systems},
  howpublished = {EasyChair Preprint no. 13601},

  year = {EasyChair, 2024}}
Download PDFOpen PDF in browser