Enhanced versions of differential evolution: state-of-the-art survey

Over the past few years, differential evolution (DE) is generally considered as a reliable, accurate and robust population-based evolutionary algorithm (EA). It is capable of handling non-differentiable, non-linear, multi-modal and constrained optimisation problems. However, it suffers from slow convergence rate and takes large computational time for optimising the computationally expensive objective functions including problems dimensionality, several local and global optimums. Over the last few years, several attempts have been made to overcome these drawbacks of simple DE by employing the key features of some existing evolutionary algorithms either self-adaptively and have been formed in the forms of enhanced versions of DEs. This paper reviews those efforts and gathered state-of-the-art survey of the DEs that included some novel self-adaptive mechanisms, different ensemble techniques, efficient local search optimisers and various constrained handling techniques.