An Improved Imperialist Competitive Algorithm using a New Modifications Mechanism

In this paper a new version of Imperialist Competitive Algorithm has been proposed for global optimization in static and continuous environments. In the proposed algorithm, in each iteration, the countries of each colony try to move towards one of the countries in their colony that are in a better position to progress. In addition, a correction mechanism has been used in the proposed algorithm where some of the countries try to improve their position in various aspects by using the position of the colonial countries. By benefiting from this mechanism, the global search ability of the algorithm increases and the algorithm will be able to avoid early convergence. Tests have been carried out on ten benchmark functions that are usually used as metrics for evaluating the efficiency of optimization algorithms. The test results indicate the high efficiency of the proposed.

کلیدواژه ها:

imperialist competitive algorithm ، optimization ، meta-heuristic algorithm ، swarm intelligence