Identification of fuzzy relation models using hierarchical fair competition-based parallel genetic algorithms and information granulation

摘要: The paper is concerned with a hybrid optimization of fuzzy inference systems based on hierarchical fair competition-based parallel genetic algorithms (HFCGA) and information granulation. process granulation realized the aid C-Means clustering. HFCGA being multi-population (PGA) exploited here to realize structure carry out parameter estimation models. becomes helpful in context models as it restricts premature convergence encountered quite often problems. It concerns set parameters model including among others number input variables be used, specific subset variables, membership functions. In process, two general mechanisms are explored. structural development via C-Means, whereas deal parametric we proceed standard least square method use technique. A suite comparative studies demonstrates that proposed algorithm leads whose performance superior comparison some other constructs commonly used modeling.