An Opposition-Based Evolutionary Algorithm for Many-Objective Optimization with Adaptive Clustering Mechanism.

Area of Science:

• Computational Intelligence
• Optimization Algorithms
• Evolutionary Computation

Background:

• Balancing convergence and diversity is crucial in many-objective optimization.
• Evolutionary algorithms face significant challenges with a large number of objectives.
• Existing methods struggle to maintain both solution quality and spread.

Purpose of the Study:

• To propose a novel opposition-based evolutionary algorithm (OBOEA) with an adaptive clustering mechanism.
• To enhance the performance of evolutionary algorithms in many-objective optimization.
• To ensure both convergence and diversity in solving complex optimization problems.

Main Methods:

• Integrating opposition-based learning for initial solution generation.
• Employing a nondominated sorting scheme with a new adaptive clustering mechanism for environmental selection.
• Comparing the proposed OBOEA with nine other evolutionary algorithms on test problems with up to 15 objectives.

Main Results:

• The proposed OBOEA demonstrated superior performance compared to nine other evolutionary algorithms.
• The algorithm effectively balanced convergence and diversity in many-objective optimization.
• Experimental results confirmed the algorithm's competitiveness and effectiveness on real-world engineering problems.

Conclusions:

• The proposed opposition-based evolutionary algorithm with adaptive clustering is effective for many-objective optimization.
• This approach successfully addresses the challenges of balancing convergence and diversity.
• The algorithm shows significant potential for solving complex, real-world multiobjective engineering problems.