ANFIS modeling of turning Al7075 hybrid nanocomposites under compressed air cooling

Introduction. Hybrid metal matrix composites (HMMCs) are increasingly used in the aviation and automotive industries due to their low density, high stiffness, and exceptional specific strength. Among aluminum MMCs, Al7075-based composites are gaining wider acceptance. Continuous research and development in this field focuses on improving the durability and performance of these advanced materials. Purpose of the work. Machinability of Al7075 is a significant challenge because the abrasive reinforcement phase causes rapid tool deterioration, increased machining forces, and a poor surface finish. Moreover, the industrial focus on green manufacturing has led to a shift from traditional coolant-based machining to sustainable alternatives. In this context, researchers have optimized machining performance using advanced technological advancements and techniques. However, limited work is reported on modeling the machining performance of Al7075 nanocomposites during turning under compressed air cooling. Methods of investigation. Manufacturers can gain a better understanding of increasing the effectiveness of turning processes for Al7075 nanocomposites by creating a comprehensive model. Therefore, this work models the machining performance of hybrid Al7075 nanocomposites during turning under compressed air-cooling conditions with an artificial neuro-fuzzy inference system (ANFIS) to predict tool wear (TW), surface roughness (Ra), and cutting force (Fc) as a function of process parameters. Results and discussion. In this work, an ANFIS model was developed to predict the machining performance considering the effect of process parameters such as cutting speed, feed rate, and depth of cut for different Al7075-based nanocomposites. These nanocomposites were prepared using silicon carbide (30–50 nm) and graphene (5–10 nm) nanoparticles as reinforcements by the stir casting process. Reinforcement materials affect the mechanical and physical properties of composites. For engineering applications, SiC and graphene are preferred reinforcements with distinctive features. ANFIS models were developed to predict Ra, Fc, and TW based on the experimental results. The Sugino method was used to represent fuzzy rules and membership functions, as it utilizes weighted averages in the defuzzification process and offers better processing efficiency. The MATLAB ANFIS toolbox was used to design and tune fuzzy inference systems. The developed ANFIS model predicts machining responses effectively and offers a practical approach for optimizing process parameters with high reliability. The results of this research show good agreement between the experimental results and the predicted ANFIS outcomes, with an average prediction error below 8%. Specifically, the ANFIS model yielded errors of 5.1% for Ra, 13.45% for Fc, and 7.92% for TW. The model exhibited excellent agreement with experimental data, demonstrating high prediction accuracy and generalization capability. 3-D graphs are plotted for a better understanding of the effect of process parameters on Fc, Ra, and TW for different nanocomposites. The findings affirm the efficacy of compressed air cooling in improving machinability while minimizing environmental impact. Furthermore, the developed ANFIS model serves as a reliable tool for optimizing turning parameters for Al7075 composites, supporting the advancement of green manufacturing strategies. This research warrants further investigation into the application of ANFIS in machining processes, specifically exploring various metal matrix composite types and rigorously assessing the long-term effects of compressed air cooling on both environmental sustainability and tool life.