Goal-Pareto based NSGA Optimization Algorithm

Goal-Pareto 기반의 NSGA 최적화 알고리즘

  • Park, Jun-Su (School of Electronic Engineering, Soongsil University) ;
  • Park, Soon-Kyu (School of Electronic Engineering, Soongsil University) ;
  • Shin, Yo-An (School of Electronic Engineering, Soongsil University) ;
  • Yoo, Myung-Sik (School of Electronic Engineering, Soongsil University) ;
  • Lee, Won-Cheol (School of Electronic Engineering, Soongsil University)
  • 박준수 (숭실대학교 정보통신전자공학부) ;
  • 박순규 (숭실대학교 정보통신전자공학부) ;
  • 신요안 (숭실대학교 정보통신전자공학부) ;
  • 유명식 (숭실대학교 정보통신전자공학부) ;
  • 이원철 (숭실대학교 정보통신전자공학부)
  • Published : 2007.03.25

Abstract

This paper proposes a new optimization algorithm prescribed by GBNSGA(Goal-Pareto Based Non-dominated Sorting Genetic Algorithm) whose result satisfies the user's needs and goals to enhance the performance of optimization. Typically, lots of real-world engineering problems encounter simultaneous optimization subject to satisfying prescribed multiple objectives. Unfortunately, since these objectives might be mutually competitive, it is hardly to find a unique solution satisfying every objectives. Instead, many researches have been investigated in order to obtain an optimal solution with sacrificing more than one objectives. This paper introduces a novel optimization scheme named by GBNSGA obeying both goals as well as objectives as possible as it can via allocating candidated solutions on Pareto front, which enhances the performance of Pareto based optimization. The performance of the proposed GBNSGA will be compared with that of the conventional NSGA and weighted-sum approach.

본 논문에서는 최적화 알고리즘의 속도를 향상시킬 수 있는 방안으로 설계자가 원하는 목적함수들의 수렴 범위를 Goal로 설정하여 최적화를 수행하는 GBNSGA(Goal-Pareto based Non-dominated Sorting Genetic Algorithm)를 제안한다. 많은 공학문제들은 하나의 목표치를 충족하는 해를 찾는 것이 아니라 다수 목적함수들을 충족하는 해를 찾는 것이 일반적이다 특히, 이러한 목적함수들은 서로 상충적인 관계를 갖는 경우가 대부분이기 때문에 모든 목적함수들을 만족하는 유일해를 찾는 것은 거의 불가능하다. 그 대안으로 일부 목적을 희생하며 설계에 부합되는 최적해를 찾는 파레토(Pareto) 방식의 최적화 알고리즘들에 대한 많은 연구가 진행되었다. 본 논문에서는 이러한 파레토 기반의 최적화 알고리즘들의 성능 향상을 도모하기 위하여 설계자의 목적을 파레토 할당에 반영하는 GBNSGA를 제안하고, 그 성능을 NSGA와 weighted-sum 접근 방식과의 비교를 통해 그 우수성을 검증하였다.

Keywords

References

  1. A. Osyczka, Multicriteria optimization for engineering design, Design Optimization (J.S. Gero, ed.), pp. 193-227, Academic Press, 1985
  2. J. Andersson, A survey of multiobjective optimization in engineering design, Technical report LiTH-IKP-R-1097, Dept. of Mechanical Engg., Linkping Univ., Linkping, Sweden, 2000
  3. J. M. III, Cognitive radio : An Integrated Agent Architecture for Software Defined Radio, Ph. D. Thesis, Royal Institute of Tech., Sweden, May 2000
  4. C. J. Rieser, Biologically Inspired Cognitive Radio Engine Model Utilizing Distributed Genetic Algorithms for Secure and Robust Wireless Communications and Networking, Ph. D. Dissertation, Virginia Tech., Blaksburg, Aug. 2004
  5. T. W. Rondeau, C. J. Rieser, and C. W. Bostian, Cognitive radios with genetic algorithms : intelligent control of software defined radios, Proc. SDR Forum Technical Conference, Phoenix, pp. C-3 - C-8, Nov. 2004
  6. R. L. Haupt, and S. E. Haupt, Practical Genetic Algorithms, 2nd edition, John Wiley & Sons, 2004
  7. E. Zitzler, Evolutionary algorithms for multiobjective optimization : Methods and applications, Ph. D. Dissertation, Swiss Federal Inst. Tech. (ETH), Zurich, Switzerland, 1999
  8. D. F. Jones, S. K. Mirrazavi, and M. Tamiz, Multiobjective meta-heuristics : an overview of the current state-of-the-art, European Journal of operational research, vol. 137, no. 1, pp. 1-9, 2002 https://doi.org/10.1016/S0377-2217(01)00123-0
  9. N. Srinivas, and K. Deb, Multiobjective optimization using nondominated sorting in genetic algorithms, Evolutionary Computation, vol. 2, no. 3, pp. 221-248, Aut. 1994 https://doi.org/10.1162/evco.1994.2.3.221
  10. K. Deb, Non-linear goal programming using multi-objective genetic algorithms, Technical Report No. CI-60/98, Dept. of Computer Science/XI, Univ. of Dortmund, Germany, pp. 269-308, Oct. 1998