Decomposition and adaptive weight adjustment method with biogeography/complex algorithm for many-objective...

Area of Science:

• Multi-objective optimization
• Computational intelligence
• Engineering optimization

Background:

• Evolutionary multi-objective optimization (EMO) algorithms struggle with balancing convergence and diversity in many objective optimization problems (MaOPs).
• Decomposition-based EMO algorithms are effective for MaOPs, and the biogeography based optimization for complex systems (BBO/Complex) offers low complexity.

Purpose of the Study:

• To propose a novel decomposition and adaptive weight adjustment based BBO/Complex algorithm (DAWA-BBO/Complex) tailored for MaOPs.
• To improve the balance between convergence and diversity in solving MaOPs.

Main Methods:

• A new crowding distance-based method for generating uniformly distributed weight vectors.
• An adaptive weight adjustment strategy to handle complex Pareto optimal fronts.
• A novel selection strategy within subsystem spaces to find non-dominated solutions.

Main Results:

• The DAWA-BBO/Complex algorithm demonstrated superior performance in convergence and diversity compared to other algorithms on DTLZ benchmark problems.
• The algorithm was successfully applied to optimize numerical control (NC) cutting parameters.

Conclusions:

• The proposed DAWA-BBO/Complex algorithm effectively addresses the challenges of MaOPs, improving both convergence and diversity.
• The algorithm provides a practical solution for real-world optimization problems, as evidenced by its application to NC cutting parameter optimization.