Comparative surface texture analysis of EN31 steel under different sustainable turning environments: A digital image processing-based study

Abstract

Industrial cutting fluids significantly contribute to pollution, prompting the adoption of sustainable machining processes like dry and minimum quantity lubrication (MQL) machining to reduce their usage. This study analyzes surface texture in EN31 steel turning to assess surface quality, detect machining defects, and evaluate cooling and lubrication effects under sustainable conditions. A comparative investigation is conducted using coated tool inserts in dry, MQL with water-soluble mineral oil, and NFMQL with aluminum oxide nanofluid, focusing on surface roughness and texture analysis. The study found that NFMQL machining significantly reduced surface roughness (R_a), achieving a 39.54% reduction compared to dry machining and a 24% reduction compared to MQL machining. Image-based texture analysis using digital image processing revealed machining patterns, including repetitive feed marks and burnout areas on machined surfaces, with a significant reduction in burnout surfaces under NFMQL conditions. Threshold-based and edge-based segmentation, combined with morphological processing, enhances the identification of machined surface features such as deep feed marks, tool marks, burnout areas, and surface irregularities like plowing effects and scratches. The refined surface profile in NFMQL conditions, characterized by reduced linear groove depth variation and increased spacing between deep feed marks, indicates lower surface roughness variability and a more widely spaced texture due to enhanced cooling and lubrication. The steeper image histogram (high mean, low variance, negative skewness, and higher kurtosis) under NFMQL conditions compared to dry and MQL indicates a smoother, more uniform texture surface with dominant brighter regions and consistent surface brightness.