The Improved Cuckoo Search Algorithms And Their Application Researches

Cuckoo search algorithm (Cuckoo Search, CS) is a new intelligent optimization algorithm given by YANG Xin-she and DEB Suash in 2009. The algorithm has the advantages of fewer parameters, but there are still existing problems like the search speed is not fast enough, precision calculation is not high enough. This paper makes some improvement to the original cuckoo search algorithm, and it’s applied to solve function optimization problems. This paper made the following results:1. Due to the calculation accuracy and the convergence issues of the original CS algorithm are not so high and fast enough respectively, on the basis of the original algorithms, we use step control vector dynamic grouping strategy to adjust the step length control vector, and a dynamic adaptive grouping scheme cuckoo algorithm (DGCS) is given. The comparison between DGCS algorithm and CS algorithm is done with four classic function performance, and the experimental results show that, DGCS algorithm can effectively increase the diversity of the population, compared with CS algorithm, DGCS algorithm has the advantages of converging faster and higher accuracy.2. As the applications for cuckoo algorithm is not big enough and the importance of multi-modal optimization problems, we give a multimodal function optimization algorithm for adaptive niche cuckoo. With four classic test functions, the comparison between the performances of solving multimodal function optimization algorithm adaptive niche cuckoo and several other classical algorithm is done. Finally, experiments show that for multimodal functional optimization problems, the algorithm has the advantages of fast solving speed and less computation time than other algorithms. In this paper, the algorithm has certain effect in solving multimodal function optimization problems.