Cuckoo search algorithm based on frog leaping local search and chaos theory

Cuckoo algorithm is a novel optimization algorithm in the field of heuristic intelligence algorithms. Given the strong random leaping in solution space search, careful local searches are susceptible to falling into the local optimum. Thus, the latter phase of the optimization slows down and the accuracy diminishes. To improve the performance of the algorithm, this paper proposes an improved cuckoo search that utilizes chaos theory to enhance the variety of the initial population. Then, this study introduces inertia weight into the Lévy flight random search to improve global searching capability. Finally, it applies the local search mechanism of the frog leaping algorithm to enhance local search and further improve the search speed and convergence precision of the algorithm. Typical test functions are employed to verify the performance of the improved algorithm. Comparison results with other algorithms indicate that the improved algorithm displays strong optimizing accuracy and high speed. Furthermore, this algorithm is confirmed to be convergent.