Differential evolution using uniform-quasi-opposition for initializing the population

被引:20
作者
Peng L. [1 ,2 ]
Wang Y. [1 ]
机构
[1] College of Computer Science, Huazhong University of Science and Technology, Wuhan
[2] School of Computer, China University of Geosciences, Wuhan
关键词
Adaptive parameter control; Evolutionary algorithms; Opposition-based learning; Optimization; Uniform design method;
D O I
10.3923/itj.2010.1629.1634
中图分类号
学科分类号
摘要
Population initialization is very important to the performance of differential evolution. A good initialization method can help in finding better solutions and improving convergence rate. According to our earlier study, uniform design generation can enhance the quality of initial population. In this study, a Uniform-Quasi-Opposition Differential Evolution (UQODE) algorithm is proposed. It uses a two-population mechanism and incorporates uniform design and quasi-opposition initialization method into differential evolution to accelerate its convergence speed and improve the stability. At the same time, an adaptive parameter control technology is adopted to avoid tuning the parameters of DE. The UQODE is compared with other three algorithms of standard Differential Evolution (DE), Opposition-based Differential Evolution (ODE) and Quasi-Oppositional Differential Evolution (QODE). Experiments have been conducted on 14 benchmark problems of diverse complexities. The results indicate that our approach has the stronger ability to find better solutions than other three algorithms especially for higher dimensional problems, in terms of the quality and stability of the final solutions. © 2010 Asian Network for Scientific Information.
引用
收藏
页码:1629 / 1634
页数:5
相关论文
共 22 条
  • [1] Brest J., Greiner S., Boskovic B., Mernik M., Zumer V., Self-adapting control parameters in differential evolution: A comparative study on numerical benchmark problems, IEEE Trans. Evolutionary Computation, 10, pp. 646-657, (2006)
  • [2] Das S., Konar A., Chakraborty U.K., Two improved differential evolution schemes for faster global search, Proceedings of the Genetic Evolution Computing Conference (GECC O), pp. 991-998, (2005)
  • [3] Fan H.Y., Lampinen J., A trigonometric mutation operation to differential evolution, J. Global Optimization, 27, pp. 105-129, (2003)
  • [4] Gong W.Y., Cai Z.H., Ling L.X., Enhancing the performance of differential evolution using orthogonal design method, Applied Mathematics Computation, 206, pp. 56-69, (2008)
  • [5] Leung Y.W., Wang Y., An orthogonal genetic algorithm with quantization for global numerical optimization, IEEE Trans. Evol. Comput., 5, pp. 41-53, (2001)
  • [6] Liu J., Lampinen J., A fuzzy adaptive differential evolution algorithm, soft computing-A fusion of foundations, Methodologies Appl., 9, pp. 448-642, (2005)
  • [7] Nobakhti A., Wang H., A simple self- adaptive differential evolution algorithm with application on the ALSTOM gasifier, Applied Soft Computing, 8, pp. 350-370, (2008)
  • [8] Noman N., Iba H., A new generation alternation model for differential evolution, Proceedings of the 8th Annual Conference on Genetic Evolution Computer Conference (GECCO 2006), pp. 1265-1272, (2006)
  • [9] Pant M., Ali M., Singh V.P., Differential evolution using quadratic interpolation for initializing the population, Proceedings of the 2009 IEEE International Advance Computing Conference (IACC 2009), pp. 375-380, (2009)
  • [10] Peng L., Wang Y.Z., Dai G.M., Chang Y.M., Chen F.J., Optimization of the earth-moon low energy transfer with differential evolution based on uniform design, Proceedings of the IEEE Congress on Evolutionary Computation, (2010)