Efficient Application of Genetic Algorithms for Robust Control System Design Problems

To effectively apply genetic algorithms for robust control system design problems, we combine genetic algorithms with local optimization methods, proposing a Dimension-Reduced Sweeping Adaptive Multi-Objective Genetic Algorithm (DRSA-MOGA). The algorithm implementation incorporates fitness function normalization techniques, dimension-reduced scanning based on optimal Pareto solution set search, and adaptive fitness function adjustment methods. These enhancements significantly improve global optimization performance and local search capability. Simulation results demonstrate that DRSA-MOGA achieves high approximation accuracy without compromising solution uniformity, where the fitness evaluation function systematically handles multiple optimization objectives through weighted aggregation.