Application of Graphene Nanofluid/Ultrasonic Atomization MQL System in Micromilling and Development of Optimal Predictive Model for SKH-9 High-Speed Steel Using Fuzzy-Logic-Based Multi-objective Design

Wei-Tai Huang,Fu-I Chou,Jinn-Tsong Tsai,Jyh-Horng Chou
DOI: https://doi.org/10.1007/s40815-020-00930-w
IF: 4.085
2020-09-14
International Journal of Fuzzy Systems
Abstract:This paper focuses on using nanofluid (graphene)/ultrasonic atomization minimum quantity lubrication (MQL) in micromilling for SKH-9 high-speed steel. Utilizing the special properties of graphene, which has excellent thermal conductivity, it is found that it successfully lowers the cutting temperature generated during processing, reduces tool wear, and improves the quality of micromilling products. Using a self-developed ultrasonic atomization system effectively improves the agglomeration of nanoparticles in nanofluids and increases the lubrication efficiency of nanoparticles. The experimental plot is robustly designed, and the <i>L</i><sub>18</sub>(2<sup>1</sup> × 3<sup>7</sup>) orthogonal table is used to find the optimal combination of parameters. The control factors are the average thickness of the nanographene, density of nanofluid, spindle speed, distance of nozzle, feed rate, amount ultrasonic atomization, air pressure, nozzle angle, and using gray correlation analysis with fuzzy inference to find more heavy quality characteristics. Finally, the optimal parameter combination of multi-quality characteristics enhanced by nanofluid (graphene)/ultrasonic atomization MQL is compared with the base fluid/ultrasonic atomization MQL, nanofluid (MWCNTs)/ultrasonic atomization MQL, whereas the differences in micromilling force, temperature, tool wear, and workpiece burr are discussed. The results indicate that the use of nanofluid (graphene)/ultrasonic atomization MQL has better results than other lubrication methods.
computer science, information systems,automation & control systems, artificial intelligence
What problem does this paper attempt to address?