A NEW ALGORITHM BASED ON THE CUCKOO SEARCH WITH DYNAMIC ADAPTATION OF PARAMETERS USING FUZZY SYSTEMS

A NEW ALGORITHM BASED ON THE CUCKOO SEARCH WITH DYNAMIC ADAPTATION OF PARAMETERS USING FUZZY SYSTEMS

In this work we studied of the parameters of the Cuckoo Search Algorithm via Levy Flights (CS). The main goal of the paper is designing a novel hybrid approach for modifying the Cuckoo Search Algorithm using a Fuzzy Inference System of the Mamdani type for calculating the optimal parameter values independent of the benchmark problem, which we are calling Fuzzy Cuckoo Search (for its acronyms FCS). In this paper different variants of the FCS are presented and the difference is the number of parameters adjusted by the fuzzy control system and the number of rules. Results show that the FCS outperforms the original version of the CS algorithms and the OCS variant of the algorithm proposed by Zhao. The statistical test shows that using a type - 1 fuzzy system in conjunction with the cuckoo search algorithms provides the best solutions