A hybrid algorithm based on extremal optimization with adaptive levy mutation and differential evolution and application

  • Authors:

  • Xiaogang Fu;Jinshou Yu

  • Affiliations:

  • Shanghai Dianji University, Shanghai, China and Institute of Automation, East China University of Science and Technology, Shanghai, China;Institute of Automation, East China University of Science and Technology, Shanghai, China

  • Venue:
  • ICNC'09 Proceedings of the 5th international conference on Natural computation
  • Year:
  • 2009

Quantified Score

Hi-index 0.00

Visualization

Abstract

A hybrid algorithm based on Extremal Optimization (EO) with adaptive levy mutation and Differential Evolution (HEODE) was proposed in this paper. It applied the idea of combination mechanism of global and local search. In the process of the global search, DE is an evolutionary algorithm based on the difference in group that can quickly approach a approximate optimal solution. During the local search, as a powerful local search capabilities algorithm EO with adaptive lévy mutation helps DE out of local maximum points. Simulation study and its application have proved its capability of strong global search and high immunity against premature convergence. Then HEODE is applied to train artificial neural network to construct a practical soft-sensor of jet fuel endpoint of main fractionator of hydrocracking unit. The obtained results indicate that the new method proposed by this paper is feasible and effective in soft-sensing of jet fuel endpoint.