Data-Driven Multi-Objective Optimization Approach to Loaded Meshing Transmission Performances for Aerospace Spiral...

Area of Science:

• Mechanical Engineering
• Aerospace Engineering
• Computational Mechanics

Background:

• Optimizing loaded meshing transmission performance is critical for aerospace spiral bevel gears, focusing on low noise and high strength.
• Existing methods for gear design and manufacturing face challenges in achieving desired performance metrics.
• Data-driven approaches offer potential for enhanced precision and efficiency in gear optimization.

Purpose of the Study:

• To propose an innovative data-driven multi-objective optimization (MOO) method for aerospace spiral bevel gears.
• To improve the accuracy and efficiency of optimizing loaded meshing transmission performances.
• To provide a robust methodology for determining precise machine tool settings.

Main Methods:

• Data-driven tooth surface modeling for curvature analysis of loaded contact points.
• Numerical loaded tooth contact analysis (NLTCA) to establish data-driven relationships between machine tool settings and performance.
• Solving the MOO function using an achievement function approach for accurate output.

Main Results:

• Successful application of data-driven tooth surface modeling and NLTCA.
• Development of accurate relationships between machine tool settings and gear performance evaluations.
• Validation of the proposed methodology through numerical examples.

Conclusions:

• The proposed data-driven MOO method offers higher computational accuracy and efficiency compared to conventional techniques.
• This approach provides a powerful tool for optimizing the loaded meshing transmission performances of aerospace spiral bevel gears.
• The methodology facilitates the achievement of prescribed requirements for low noise and high strength in gear design.