Cross Opposition Based Differential Evolution Optimization

Abstract

Differential Evolutionary (DE) Algorithms is one of the most popular metaheuristic approach. For optimization purpose DE is very useful to solve various kind of problems. In addition to that the paper offers a Cross-Opposition Based Differential Evolution (CODE). An impression of Opposition-based learning (OBL) is incorporated in population initialization phase and in step of crossover. The performance of algorithm is analysed for different mutation strategies of DE and various other existing approaches. Results demonstrated that the algorithm outperform in terms of convergence speed, versatile population and dimension size.