Application of different simulation approaches to numerically optimize high-frequency mechanical impact (HFMI) post-treatment process

Clément Ernould, Jan Schubnell, Majid Farajian, Andreas Maciolek, David Simunek, Martin Leitner, Michael Stoschka

Research output: Contribution to journalArticle

Abstract

The weld seam is generally the weak point of welded mechanical parts subject to fatigue loading. For this issue, a post-weld mechanical surface process called high-frequency mechanical impact (HFMI) was developed. This process combines both mechanical effects and a weld geometry improvement by generating compressive residual stresses and making a smoother transition between the base plate and the weld. Benefits of the process are statistically proven by numerous fatigue test results. Finite-element method (FEM) based numerical simulations of the process have been developed to estimate the material state after treatment. Good agreements with experimental results were obtained. Until present, rebounds of the pin between each primary impact have not been overlooked by such simulations. To discuss their eventual effects, signal of strain gauges glued on the pin was processed. A typical impact pattern of the pin kinetic during HFMI treatment could be identified and then implemented in a pre-existing FEM model. Numerical simulations were conducted using recently developed non-linear combined isotropic-kinematic hardening law with strain-rate dependency according to Chaboche und Ramaswamy–Stouffer models. These hardening laws were calibrated for S355 J2 mild steel. The simulation procedure was performed for flat specimen and representative butt weld joint.
Original languageEnglish
Pages (from-to)725-738
Number of pages14
JournalWelding in the World
Volume63
Issue number3
DOIs
Publication statusPublished - 2019
Externally publishedYes

Fingerprint Dive into the research topics of 'Application of different simulation approaches to numerically optimize high-frequency mechanical impact (HFMI) post-treatment process'. Together they form a unique fingerprint.

Cite this