Cuckoo Search Algorithm With Ensemble Strategy for Continuous Optimization Problems

Abstract

Cuckoo search (CS) algorithm is a simple and effective optimization technique. However, CS algorithm may encounter the issue of premature convergence as the complexity of the problem increases. To address this challenge, a cuckoo search algorithm with ensemble strategy, called CSES, is presented in this paper. Specifically, three new search strategies with diverse properties are designed to boost the competitiveness. After that, according to the idea of selective ensemble, a priority roulette method is employed to select the appropriate search strategy at distinct phases of the evolution process, so as to produce more promising results. Furthermore, the effectiveness evaluation of CSES algorithm is carried out on 58 benchmark functions from CEC 2013 and CEC 2017 test suites and several real-world problems including chaotic time series prediction and transformer fault classification. Simulation outcomes illustrate that the introduced CSES is superior to five recently developed CS variants in terms of search accuracy and robustness, for example it provides 10 and 12 better performance improvements on $30D$ and $50D$ optimization of the CEC 2013 benchmarks, and produces 19 and 11 better performance improvements on $30D$ and $50D$ optimization of the CEC 2017 benchmarks, respectively. Moreover, CSES also exhibits more superiority compared to several other advanced evolutionary methods, including butterfly optimization algorithm (BOA), dung beetle optimizer (DBO), electric eel foraging optimization (EEFO), jellyfish search (JS) and wild horse optimizer (WHO), and yields 25 better performance improvements on $30D$ optimization of the CEC 2013 benchmarks.