An ant colony algorithm based on search concentration and dynamic pheromone updating

Abstract

The invention discloses an ant colony algorithm optimization method based on searching concentration degree and dynamic pheromone update, belonging to the artificial intelligence field. Compared withthe classical ant colony algorithm, this method has the following improvements: (1) according to the path found by the previous generation of ant colony, the search concentration factor is introducedinto the selection strategy, which can adjust the range of ant selection node adaptively; (2) In the global pheromone updating phase, the search time is shortened by changing the pheromone increment dynamically; (3) the pheromone rollback mechanism is used to jump out of the local extremum when the optimal solution is not improved after many iterations. Through the above improvements, the invention obviously improves the global searching ability and the running efficiency compared with the basic ant colony algorithm.

Description

technical field [0001] The invention relates to an ant colony algorithm optimization method based on search concentration and dynamic pheromone update, belonging to the field of artificial intelligence. Background technique [0002] With the rapid development of science and technology, the traditional control theory based on the precise model of the object and the optimization algorithm using determinism have encountered great difficulties. So people got inspiration from the intelligent control behavior of simulating human beings, combined artificial intelligence with automatic control theory, and created the intelligent control theory; people were inspired by biological evolution and bionics, and proposed many heuristic intelligent optimization methods. Such as taboo algorithm, neural network algorithm, genetic algorithm, immune algorithm and ant colony algorithm, etc., they provide a new way to solve many complex optimization problems. [0003] Among them, the ant colony ...

AI Technical Summary

Problems solved by technology

But at the same time, the ant colony algorithm also has certain limitations. For example, the lack of pheromones in the initial stage leads to the blindness of most ants' searches. The running cycle of the system is too long and the convergence speed is slow

In addition, with the expansion of the scale of the problem, after the ants search to a certain extent, most of the solutions found by the individual are concentrated on a certain path, and under the effect of positive feedback, the pheromones on these paths will further increase, preventing the search for solutions. Further search of the space, eventually falling into a local optimal solution and stagnation phenomenon

Images