Ant colony algorithm with Stackelberg game and multi-strategy fusion

摘要

Aiming at the disadvantages of the ant colony algorithm, such as slow convergence speed and easy to fall into local optimum, this paper proposes an ant colony algorithm with Stackelberg game and multi-strategy fusion. Firstly, Stackelberg game is established between ant colonies, and the population with the excellent performance is taken as the leader to increase the influence of excellent ant colony. Secondly, a multi-strategy fusion system is proposed, which is composed of three strategies: One is the pheromone fusion strategy, which selects the population whose entropy is less than the threshold value and the population with the highest similarity for pheromone fusion to increase the diversity of the algorithm. The second is the elite ant learning strategy, which speeds up the convergence rate by learning the elite ants of the elite population; The third is the pheromone recombination strategy, which helps the algorithm jump out of the local optimum. The simulation experiments of multiple cases in TSPLIB show that the improved algorithm balances diversity and the convergence speed, and effectively improves the quality of the solution.