Improved adaptive ant colony algorithm by incorporating multiple strategies

To address the problems of slow convergence and local optima entrapment in the ant colony algorithm(ACO), this paper proposes an improved ant colony algorithm(IACO) integrating multiple strategies. First, a Latin hypercube strategy based on lens imaging inverse learning is employed for population initialization, which enhances population diversity while improving the quality of initial solutions. Second, dynamic pheromone update rules and adaptive adjustment mechanisms are implemented for pheromone updates, further enhancing global search capability. Third,distribution-based mutation perturbation of pheromone concentration is adopted to strengthen the ability to escape local optima. Finally, comparative simulation experiments are conducted between IACO and the standard ACO as well as other classic swarm intelligence optimization algorithms on multiple standard test functions. Specifically, compa-risons are made with standard ACO, GA(genetic algorithm), and PSO(particle swarm optimization) on three test functions: Rastrigin, Sphere, and Ackley. Additionally, on the Rastrigin function, comparisons are made with the standard ACO by controlling ant population size and heuristic information weight. The experimental results demonstrate that IACO exhibits superior convergence performance and stability.