<p>Computer Numerical Control milling is a critical technique in advanced manufacturing, yet ensuring consistent surface quality remains challenging due to dynamic variations in temperature, force, and other uncertain factors such as material properties and environmental noise. Robust optimization offers a promising approach to mitigate these uncertainties, enhancing product quality and precision, which holds significant potential for advancing machining technologies of Computer Numerical Control. In this study, 7075 aluminum alloy was selected as the workpiece material, and a comprehensive set of 25 × 4 face milling experiments was conducted. Take the spindle speed (n), feed per tooth (<InlineEquation ID="IEq1"> <EquationSource Format="TEX">${\text{f}}_{\text{z}}$</EquationSource> <EquationSource Format="MATHML"><math> <msub> <mtext>f</mtext> <mtext>z</mtext> </msub> </math></EquationSource> </InlineEquation>), milling depth (<InlineEquation ID="IEq2"> <EquationSource Format="TEX">${\text{a}}_{\text{p}}$</EquationSource> <EquationSource Format="MATHML"><math> <msub> <mtext>a</mtext> <mtext>p</mtext> </msub> </math></EquationSource> </InlineEquation>) and milling width (<InlineEquation ID="IEq3"> <EquationSource Format="TEX">${\text{a}}_{\text{e}}$</EquationSource> <EquationSource Format="MATHML"><math> <msub> <mtext>a</mtext> <mtext>e</mtext> </msub> </math></EquationSource> </InlineEquation>) as input variables, and the mean and variance of surface roughness as output variables. Use stochastic kriging methods to reasonably characterize the relationship between input and output. To achieve a combination of milling parameters that ensures both minimal surface roughness and high robustness, the Mean-Square Error criterion was utilized to integrate the mean and variance models constructed through stochastic kriging. Additionally, the hunter-prey optimization algorithm was enhanced by introducing Latin hypercube sampling and a nonlinear dynamic update strategy. Then, the improved algorithm was applied to the established Mean-Square Error model, and the milling parameter combination with better surface roughness and robustness was obtained. The findings of this study not only provide a practical framework for optimizing machining parameters of Computer Numerical Control but also contribute to the broader field of robust optimization in manufacturing. By addressing the challenges posed by uncertainties, this research supports the development of more reliable and efficient machining processes, aligning with the goals of Industry 4.0 and smart manufacturing. The proposed model is not limited to manufacturing; its robust optimization framework is adaptable to diverse fields such as aerospace, energy systems, and biomedical engineering, offering reliable solutions under uncertainty.</p> Graphical Abstract <p></p>

Robust parameter optimization of 7075 aluminum alloy milling based on improved hunter-prey optimizer algorithm

期刊论文
发表日期：

2025年7月17日

查看全文

Yang Yang,
Jie Hong,
Zetong Li,
Chen Su,
Xinbei Wei

摘要

Computer Numerical Control milling is a critical technique in advanced manufacturing, yet ensuring consistent surface quality remains challenging due to dynamic variations in temperature, force, and other uncertain factors such as material properties and environmental noise. Robust optimization offers a promising approach to mitigate these uncertainties, enhancing product quality and precision, which holds significant potential for advancing machining technologies of Computer Numerical Control. In this study, 7075 aluminum alloy was selected as the workpiece material, and a comprehensive set of 25 × 4 face milling experiments was conducted. Take the spindle speed (n), feed per tooth ( ${\text{f}}_{\text{z}}$ $f_{z}$ ), milling depth ( ${\text{a}}_{\text{p}}$ $a_{p}$ ) and milling width ( ${\text{a}}_{\text{e}}$ $a_{e}$ ) as input variables, and the mean and variance of surface roughness as output variables. Use stochastic kriging methods to reasonably characterize the relationship between input and output. To achieve a combination of milling parameters that ensures both minimal surface roughness and high robustness, the Mean-Square Error criterion was utilized to integrate the mean and variance models constructed through stochastic kriging. Additionally, the hunter-prey optimization algorithm was enhanced by introducing Latin hypercube sampling and a nonlinear dynamic update strategy. Then, the improved algorithm was applied to the established Mean-Square Error model, and the milling parameter combination with better surface roughness and robustness was obtained. The findings of this study not only provide a practical framework for optimizing machining parameters of Computer Numerical Control but also contribute to the broader field of robust optimization in manufacturing. By addressing the challenges posed by uncertainties, this research supports the development of more reliable and efficient machining processes, aligning with the goals of Industry 4.0 and smart manufacturing. The proposed model is not limited to manufacturing; its robust optimization framework is adaptable to diverse fields such as aerospace, energy systems, and biomedical engineering, offering reliable solutions under uncertainty.

Graphical Abstract