An improved CACO algorithm based on adaptive method and multi-variant strategies

Abstract

Chaotic ant colony optimization (CACO) algorithm is an effective optimization algorithm that simulates the self-organization and chaotic behavior of ants. However, in the research and application of the CACO algorithm for solving complex optimization problems, the CACO algorithm presents some disadvantages. In order to resolve these disadvantages, an improved CACO algorithm based on adaptive multi-variant strategies (CACOAMS) is proposed in this paper. The CACOAMS algorithm takes full advantage of multi-population strategy, the neighborhood comprehensive learning strategy, the fine search strategy, the chaotic optimization strategy, the super excellent ant strategy, the punishment strategy and min---max ant strategy in order to avoid the local optimization solution and stagnation, guarantee learning rate of the different dimensions for each ant and the diversity of the search, eliminate the self-locking trap between environmental boundary and obstacles, improve the search efficiency, search accuracy and robustness of the algorithm. In order to testify to the performance of the CACOAMS algorithm, the CACOAMS algorithm is applied to test the benchmark functions and dynamically adjust the values of PID parameters. The simulation results show that the CACOAMS algorithm takes on the strong flexibility, adaptability and robustness. It can effectively improve system control precision and guarantee feasibility and effectiveness.