An Adaptive Cuckoo Search Algorithm with Optional External Archive for Global Numerical Optimization
6 Pages Posted: 12 Nov 2018
Date Written: July 1, 2018
In this study, an adaptive cuckoo search algorithm with optional external archive (ACS-OEA) is proposed to improve the optimization performance of the original cuckoo search (CS). The proposed ACS-OEA mainly consists of the following three algorithmic components, i.e., a modified global random walk strategy, optional external archive and the parameter adaptation scheme. The modified global random walk strategy is presented with probabilistic mutation by incorporating a set of recently explored inferior solutions rather than the best solution obtained before. The archive operation is introduced to store the recently explored inferior solutions. Besides, the parameter adaptation scheme dynamically updates the control parameters based on a Cauchy distribution and the Lehmer mean at each iteration process. Moreover, in order to verify the performance of ACS-OEA, experiments are conducted on 48 benchmark functions from two popular test suites. The experimental results demonstrate that ACS-OEA shows an significant improvement in effectiveness and efficiency, and is superior to the original CS and several CS variants. At last, ACS-OEA is applied to solve an application problem of parameter identification of uncertain fractional-order chaotic systems. Based on the observations and results analysis, the proposed algorithm can be regarded as an efficient and promising tool for solving the real-world complex optimization problems besides the benchmark problems.
Suggested Citation: Suggested Citation