A Hybrid GRA-TOPSIS-RFR Optimization Approach for Minimizing Burrs in Micro-Milling of Ti-6Al-4V Alloys

Area of Science:

• Materials Science and Engineering
• Manufacturing Processes
• Aerospace Engineering

Background:

• Micro-milling is vital for intricate aerospace components, especially from challenging Ti-6Al-4V alloys.
• Burr formation during micro-milling complicates finishing and degrades part quality.
• Optimizing micro-milling parameters is critical for efficient aerospace manufacturing.

Purpose of the Study:

• To develop and validate a novel hybrid optimization algorithm for minimizing burrs in Ti-6Al-4V micro-milling.
• To investigate the influence of spindle speed, depth of cut, and feed rate on surface roughness and burr width.
• To enhance prediction accuracy for micro-milling outcomes using advanced machine learning.

Main Methods:

• A hybrid-ranking optimization algorithm integrating Grey Relational Analysis (GRA) and Technique for Order Preference by Similarity to Ideal Solution (TOPSIS).
• Incorporation of a Random Forest Regression (RFR) model to capture complex parameter-outcome relationships.
• Systematic investigation of down-milling and up-milling processes with varying machining parameters.

Main Results:

• Depth of cut significantly impacts surface roughness; feed rate per tooth is crucial for burr control.
• The GRA-TOPSIS-RFR hybrid model demonstrated superior performance in optimization and prediction.
• Prediction accuracy improved by 42.6% compared to traditional linear regression models.

Conclusions:

• The GRA-TOPSIS-RFR approach offers a robust method for optimizing micro-milling of titanium alloys.
• Validated experimental results confirm the effectiveness of the proposed optimization strategy.
• This research contributes to enhanced quality, performance, and lifespan of aerospace components.