History based learning artificial bee colony algorithm for electromagnetic inverse problems

Xiu Zhang, Xin Zhang, Weinong Fu, S. X. Nu

Research output: Chapter in book / Conference proceedingConference article published in proceeding or bookAcademic researchpeer-review


Electromagnetic (EM) inverse problems are usually modeled as simulation-based optimization problems with multimodal and non-differentiable properties. These properties justify the invention of more effective optimization algorithms. Artificial bee colony (ABC) has already been applied to tackle EM inverse problems. This paper applies a tree structure to record all search moves of ABC. A history based learning ABC is proposed to assist the finding of promising search directions. Novel variation formula is invented to guide the move of honey bees. The proposed algorithm is applied to tackle a loudspeaker design problem. Simulation results show that the proposed algorithm is more effective and reliable than the compared algorithms.
Original languageEnglish
Title of host publicationIEEE CEFC 2016 - 17th Biennial Conference on Electromagnetic Field Computation
ISBN (Electronic)9781509010325
Publication statusPublished - 12 Jan 2017
Event17th Biennial IEEE Conference on Electromagnetic Field Computation, IEEE CEFC 2016 - Hotel Hilton Miami Downtown, Miami, United States
Duration: 13 Nov 201616 Nov 2016


Conference17th Biennial IEEE Conference on Electromagnetic Field Computation, IEEE CEFC 2016
Country/TerritoryUnited States


  • Artificial bee colony
  • Electromagnetic device design
  • Electromagnetic inverse problem
  • History based learning

ASJC Scopus subject areas

  • Computational Mathematics
  • Instrumentation
  • Electrical and Electronic Engineering

Cite this