Recent advances in differential evolution: a survey and experimental analysis Ferrante Neri · Ville Tirronen Published online: 27 October 2009 © Springer Science+Business Media B.V. 2009 Differential Evolution (DE) is a simple and efficient optimizer, especially for continuous optimization. It has been widely used for solving various engineering problems. However, the DE structure has some limitations in the search logic, as it contains too narrow a set of exploration moves. This has inspired many researchers to improve DE by proposing modifications to the original algorithm. This paper presents a survey on DE and its recent advances. The DE modifications are classified into two macro-groups: (1) algorithms that integrate additional components within the DE structure, and (2) algorithms that employ a modified DE structure. For each macro-group, four algorithms representative of the state-of-the-art in DE are selected for an in-depth description of their working principles. These eight algorithms are tested on a set of benchmark problems. Experiments are repeated for both low and high dimensional cases. The working principles, differences, and similarities of these recently proposed DE-based algorithms are highlighted. Although it is unclear whether one algorithm is superior to others, some conclusions can be drawn. To improve DE performance, modifications that include additional and alternative search moves are necessary. These extra moves should assist the DE framework in detecting new promising search directions. A limited employment of these alternative moves appears to be the best option. Successful extra moves are obtained through an increase in exploitative pressure and the introduction of some randomization. However, randomization should not be excessive, as it may jeopardize the search. A proper increase in randomization is crucial for significant improvements in DE functioning. Numerical results show that the most efficient additional components in a DE framework are population size reduction and scale factor local search. Regarding modified DE structures, global and local neighborhood search and self-adaptive control parameter schemes are the most promising modifications. Keywords: Differential Evolution · Survey · Comparative Analysis · Self-Adaptation · Continuous OptimizationRecent advances in differential evolution: a survey and experimental analysis Ferrante Neri · Ville Tirronen Published online: 27 October 2009 © Springer Science+Business Media B.V. 2009 Differential Evolution (DE) is a simple and efficient optimizer, especially for continuous optimization. It has been widely used for solving various engineering problems. However, the DE structure has some limitations in the search logic, as it contains too narrow a set of exploration moves. This has inspired many researchers to improve DE by proposing modifications to the original algorithm. This paper presents a survey on DE and its recent advances. The DE modifications are classified into two macro-groups: (1) algorithms that integrate additional components within the DE structure, and (2) algorithms that employ a modified DE structure. For each macro-group, four algorithms representative of the state-of-the-art in DE are selected for an in-depth description of their working principles. These eight algorithms are tested on a set of benchmark problems. Experiments are repeated for both low and high dimensional cases. The working principles, differences, and similarities of these recently proposed DE-based algorithms are highlighted. Although it is unclear whether one algorithm is superior to others, some conclusions can be drawn. To improve DE performance, modifications that include additional and alternative search moves are necessary. These extra moves should assist the DE framework in detecting new promising search directions. A limited employment of these alternative moves appears to be the best option. Successful extra moves are obtained through an increase in exploitative pressure and the introduction of some randomization. However, randomization should not be excessive, as it may jeopardize the search. A proper increase in randomization is crucial for significant improvements in DE functioning. Numerical results show that the most efficient additional components in a DE framework are population size reduction and scale factor local search. Regarding modified DE structures, global and local neighborhood search and self-adaptive control parameter schemes are the most promising modifications. Keywords: Differential Evolution · Survey · Comparative Analysis · Self-Adaptation · Continuous Optimization