조건부적인 퍼지 클러스터링을 이용한 온-라인 적응 뉴로-퍼지 제어

On-line Adaptive Neuro-Fuzzy Control using Conditional Fuzzy Clustering

  • 신동철 (충북대학교 전기공학과 대학원) ;
  • 곽근창 (충북대학교 전기공학과 대학원) ;
  • 전병석 (충북대학교 전기공학과 대학원) ;
  • 김종근 (충북대학교 전기공학과 대학원) ;
  • 유정웅 (충북대학교 전기공학과 대학원)
  • Shin, D.C. (Dept. of Electrical Engineering, Chungbuk National University) ;
  • Kwak, K.C. (Dept. of Electrical Engineering, Chungbuk National University) ;
  • Jeun, B.S. (Dept. of Electrical Engineering, Chungbuk National University) ;
  • Kim, J.G. (Dept. of Electrical Engineering, Chungbuk National University) ;
  • Ryu, J.W. (Dept. of Electrical Engineering, Chungbuk National University)
  • 발행 : 1999.07.19

초록

The main idea of the proposed neuro-fuzzy system is conditional clustering whose main objective is to develop clusters preserving homogeneity of the clustered patterns with regard to their similarity in the input space as well as their respective values assumed in the output space. In the proposed neuro-fuzzy system, the structure identification is used with conditional fuzzy clustering, the parameter identification carried out by the hybrid learning scheme using back-propagation and total least squares.

키워드