A scheme for a priori generation of the optical patterns for magnetic nanoemulsion based defect detection in carbon steel specimens

Abstract

A numerical scheme is proposed for a priori generation of the optical patterns for magnetic nanoemulsion (MNE) based defect detection in ferromagnetic steel specimens. External static magnetic field (HDC) induced tunable optical contrast in MNE has been exploited for the development of the wide-area non-contact optical sensor, which utilizes the magnetic flux leakage (MFL) from the defect regions to generate visually discernible optical patterns, where the image contrast varies with the severity of the defects. For theoretically generating the optical patterns, a three-step numerical scheme is proposed, where the first step involves the physical characterization of the MNEs. In the presence of a HDC, the MNE droplets exhibit an oriented linear arrangement, where the inter-droplet separation decreases with increasing HDC, resulting in a blue shift of the wavelength corresponding to the maxima of the Bragg's reflection peaks (λmax). Unique λmax-HDC calibration curves are obtained for various types of MNEs depending on the nature of the stabilizing moieties. In the second step, the surface distributions of the MFL are obtained using finite element modelling. Subsequently, in the third stage, using the λmax-HDC calibration curves, the surface distributions of MFL are converted to surface distributions of λmax values, which are then pseudo colour-coded to theoretically generate the optical patterns. The presence of the rectangular slots, double rectangular, cylindrical, and buried defects are clearly discernible from the simulated optical patterns, which are found to be in good agreement with the experimentally recorded patterns. Defect depth-dependent intensity variations and estimated defect widths from the simulated images are in good agreement with the data obtained from the experimental images, thereby validating the efficacy of the proposed scheme for a priori prediction of the optical patterns for MNE-based non-contact wide area defect detection. Further, the proposed scheme will be beneficial for the selection of appropriate MNE-based sensors, reducing inspection time, and developing automated inspection routines with enhanced detection capabilities.