The Influence of Post-Necking Strain Hardening Behaviour on Fatigue Lifetime Prediction of Cold-Formed High Strength Steel.

High strength steels (HSS) have proven to be indispensable for technological advances in mechanical engineering. Modern design strategies are mainly focussed on increasing the durability and efficiency of components through smart material selection and Finite Elements Analysis (FEA). In this regard, the adopted material models can have a direct influence on the resulting design approach. Since newly developed HSS grades are progressively being used for many performance–critical components and applications, the importance of representative material models has grown. Especially in the case of mobile lifting equipment, where load-bearing components consist of hollow rectangular sections, derived from secondary forming processes. In addition, these applications are often subjected to cyclic loading conditions, causing fatigue crack initiation at bent regions of the cold-formed sections. Consequently, accurate strain hardening behaviour and fatigue data of HSS are of major importance. To tackle this problem a cold-formed lab-scale specimen was modelled and fatigue tested. To enable a prediction of the residual plastic strains after bending and the strain amplitude of fatigue loading, a modelling strategy was developed in ABAQUS v19. The resulting outputs are then used to predict the fatigue life according to a critical plane approach. However, several uncertainties in material properties can directly influence the reliability of this prediction. For instance, inverse identification procedures can bring about certain scatter on material parameters, the true plastic behaviour lies within a range of possible post-necking strain hardening curves. The current study is focussed on investigating the influence of this scatter on numerically predicted fatigue life. This is performed by means of a mixed numerical-experimental approach, where a cold-formed fatigue specimen, made from 4 mm-thick S500MCL, is investigated. This method allows for the construction of a numerically derived SN-curve that can be compared and validated with an experimental test campaign.

Jaar van publicatie:2020

Toegankelijkheid:Open