作者: Igor Škrjanc , Sašo Blažič , Osvaldo Agamennoni
DOI: 10.1016/J.AUTOMATICA.2004.09.010
关键词:
摘要: In this paper we present a new method of interval fuzzy model identification. The combines identification methodology with some ideas from linear programming theory. On finite set measured data, an optimality criterion that minimizes the maximal estimation error between data and proposed output is used. idea then extended to modelling optimal lower upper bound functions define band contains all measurement values. This results in or parameters. called (INFUMO). can be used when describing family uncertain nonlinear systems physical parameters are observed. We believe very efficiently used, especially fault detection robust control design.