Scholarlink

<p>This paper presents a new algorithm for state-space type-2 evolving fuzzy identification. Important aspects related to adopted methodology, in correspondence to its originality, include: the use of particle swarm algorithm for the online optimal estimation of the footprint of uncertainty for handling uncertainties inherent to the experimental data, the multi-objective optimization associated with the minimal width and maximum coverage of the prediction interval of the experimental data, a filtering-based Markov parameter estimation recursive algorithm for ensuring unbiased estimation from experimental data corrupted by non-white noise. The efficiency and applicability of the algorithm are demonstrated through the adaptive identification of a benchmark nonlinear system and a two-degrees-of-freedom helicopter.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

State-Space Type-2 Fuzzy Adaptive Identification Approach Based on Evolving Data Clustering

  • 期刊论文

  • 发表日期：

    2025年8月21日

查看全文
  • Luís Miguel Magalhães Torres,
  • Ginalber Luiz de Oliveira Serra

摘要

This paper presents a new algorithm for state-space type-2 evolving fuzzy identification. Important aspects related to adopted methodology, in correspondence to its originality, include: the use of particle swarm algorithm for the online optimal estimation of the footprint of uncertainty for handling uncertainties inherent to the experimental data, the multi-objective optimization associated with the minimal width and maximum coverage of the prediction interval of the experimental data, a filtering-based Markov parameter estimation recursive algorithm for ensuring unbiased estimation from experimental data corrupted by non-white noise. The efficiency and applicability of the algorithm are demonstrated through the adaptive identification of a benchmark nonlinear system and a two-degrees-of-freedom helicopter.