Differential Evolution (DE) is a well known optimization approach to solve nonlinear and complex problems. But, DE, like other probabilistic optimization algorithms, sometimes exhibits premature convergence and stagnation. DE exploration and exploitation capabilities depend on the two processes namely mutation process and crossover process. In these two processes exploration and exploitation are balanced using the fine tuning of scale factor F and crossover probability CR. In the solution search process of DE, there is a enough chance to skip the true solution due to large step size. Therefore, in this paper, to balance the diversity and convergence capability of DE, fitness based self adaptive F and CR are proposed. The proposed strategy is named as Fitness based Self Adaptive DE (FSADE). The experiments on 16 well known test problems of different complexities show that the proposed strategy outperforms the basic DE and recent variants of DE, namely Self-adaptive DE (SaDE) and Scale Factor Local Search DE (SFLSDE) in most of the experiments. © 2014 Springer International Publishing Switzerland.
CITATION STYLE
Sharma, H., Shrivastava, P., Bansal, J. C., & Tiwari, R. (2014). Fitness based self adaptive differential evolution. Studies in Computational Intelligence, 512, 71–84. https://doi.org/10.1007/978-3-319-01692-4_6
Mendeley helps you to discover research relevant for your work.