An Effective Constraint-Handling Improved Cuckoo Search Algorithm and Its Application in Aerodynamic Shape Optimization

This research develops an effective constraint-handling method to improve meta-heuristic algorithms' performance when solving constrained optimization problems. With the cuckoo search (CS) as the basic optimization algorithm, a constraint-handling improved cuckoo search algorithm (CICS) is presented and applied to an airfoil aerodynamic shape optimization. First, the newly developed constraint-handling (CH) method is compared to five types of traditional techniques by incorporating the cuckoo search algorithm, particle swarm optimization (PSO), and genetic algorithm (GA), respectively, on ten benchmark analytical test problems and four engineering design optimization problems. Results indicate that the present method is effective and robust, and outperforms the other constraint-handling methods, not only for cuckoo search but also for particle swarm optimization and genetic algorithm. Next, the presented constraint-handling improved cuckoo search algorithm is compared to nine types of state-of-the-art meta-heuristic algorithms. It shows better performance when solving the aforementioned constrained problems. Finally, the CICS algorithm is successfully applied to a strongly constrained aerodynamic drag minimization design problem. It also shows its feasibility to aerodynamic design optimization and superiority to the original cuckoo search algorithm.