Optimizing CNC turning of AISI D3 tool steel using Al₂O₃/graphene nanofluid and machine learning algorithms

Abstract

Turning AISI (American Iron and Steel Institute) D3 tool steel can be challenging due to a lack of optimal process parameters and proper coolant application to achieve high surface quality and temperature control. Machine learning helps in predicting the optimal parameters, whereas nanofluids enhance cooling efficiency while preserving both the tool and the workpiece. This work intends to utilize advanced machine learning approaches to optimize process parameters with the application of hybrid nanofluids (Al2O3/graphene) during the CNC turning of AISI D3. The Response Surface Methodology (RSM), Back Propagation (BP) neural networks, and Genetic Algorithms (GA) will be utilized to model and predict optimal turning parameters to enhance surface quality and manage tool tip temperature. The experiments ranged the cutting speed, nanofluid concentration, depth of cut, and feed rate from 150 to 180 m/min, 0.3 to 0.9 wt%, 0.5–0.9 mm, and 0.03–0.07 mm/rev. RSM and ANN analyses showed that cutting speed and feed rate had a significant effect on surface quality, contributing 11.5 % and 10.5 %, respectively, whereas the nanofluid affected tool tip temperature by 42.5 %. The GA determined that the optimal cutting speed became 150 m/min, the feed rate was 0.05 mm/rev, the cutting depth was 0.6 mm, and the nanofluid concentration was 0.8 %. At temperatures ranging from 23.01 °C to 28.41 °C, these conditions produced a desirable surface roughness of 0.16–0.45 μm. The findings emphasize the benefits of utilizing Al2O3/graphene nanofluid and machine learning algorithms in CNC turning to improve surface roughness and control temperature.

Keywords

• AISI D3 tool steel
• CNC turning
• Hybrid nanofluid
• Machine learning
• Optimization