Cuckoo search algorithm with onlooker bee search for modeling PEMFCs using T2FNN.

The accurate mathematical model plays an important role in the simulation and analysis of proton exchange membrane fuel cells (PEMFCs). This paper proposes the nonlinear modeling approach using type-2 fuzzy neural network (T2FNN) for PEMFCs. For the optimal tuning of the T2FNN, a novel cuckoo search algorithm, the cuckoo search algorithm with onlooker bee search (ObCS) is proposed. In ObCS, the opposition based learning strategy is used to make the initial population distribution more uniform. A modified local search strategy on onlooker bee search is designed to improve the local search ability. Numerical experiments with two groups of test functions show that the ObCS has a higher quality of solutions in comparison with the basic CS, the two improved CSs, and the other state-of-the-art optimization algorithms. Finally, the ObCS is applied to optimize the parameters of the T2FNN for modeling a PEMFC. The experimental results demonstrate that the ObCS and RLS based T2FNN is a more efficient technique by comparing with the CS and RLS based T2FNN.