Investigating the Accuracy of Eddy Current Simulation and Evaluation for Rail Testing

Abstract

Eddy current testing is a non-destructive evaluation (NDE) technique that is widely used for testing railway rails with a strong focus on detecting cracks. In this paper, we investigate the accuracy of eddy current simulation and eddy current evaluation in the context of rail testing. We focus on analyzing the numerical accuracy of BEM simulations depending on the mesh size relative to the skin depth, and on estimating the influence of simplifying assumptions for cracks. Specifically, we consider the defect width, surface geometry, testing train velocity and material parameter change. We provide relative error estimates and we propose a regime for systematic rail defect simulation.