Peridynamic-inspired bonds to address strain incompatibility in modelling rail damage through ratchetting strain accumulation

Incremental accumulation of strain by rail steels in response to repeated compressive shear loading above the shakedown limit determines rail maintenance requirements and is a precursor to the initiation of defects that may develop into safety critical cracks and broken rails. Most modelling techniques struggle, through limitations in scale or on computational expense, to evaluate the long-term performance of rail steels, addressing this by simplifying the rail material response, considering only the ratchetting response and defining failure by a critical value of accumulated strain. However, these simplified models of ratchetting response are unable to capture experimentally observed wear behaviour of rail steels prior to strain exhaustion. Here we demonstrate validation across a range of contact pressures characteristic of rail operation that this behaviour can be represented through introduction of peridynamic-inspired bonds to quantify strain partitioning and resultant failure in the material. This is a crucial step in linking microstructural and metallurgical scale behaviour to macro level wear and damage development for maintenance planning. The newly modelled failure mechanism dominates at lower numbers of contact cycles (<10,000) and under lower strain that acts prior to strain exhaustion and was previously neglected. Through capturing strain partitioning with peridynamic-inspired bonds, inconsistencies between wear experiments and predicted wear rates considering strain exhaustion alone can be addressed. Parameters defining the additional wear mechanism are explored here to understand how they represent the physical failure process, and how they can be combined with existing understanding of material failure through strain exhaustion.