• 中国计算机学会会刊
  • 中国科技核心期刊
  • 中文核心期刊

计算机工程与科学

• 人工智能与数据挖掘 • 上一篇    下一篇

区间二型模糊逻辑系统质心降型及加权Nie-Tan算法

陈阳,王涛   

  1. (辽宁工业大学理学院,辽宁 锦州 121001)
  • 收稿日期:2019-07-28 修回日期:2019-10-22 出版日期:2020-04-25 发布日期:2020-04-25
  • 基金资助:

    国家自然科学基金(61973146,61773188,61803189);辽宁省自然科学基金指导项目(20180550056)

Centroid type-reduction of interval type-2 fuzzy
logic systems and three weighted Nie-Tan algorithms

CHEN Yang,WANG Tao   

  1. (College of Science,Liaoning University of Technology,Jinzhou 121001,China)
  • Received:2019-07-28 Revised:2019-10-22 Online:2020-04-25 Published:2020-04-25

摘要:

降型是二型模糊逻辑系统中的核心模块。Nie-Tan (NT) 算法不涉及到迭代过程,可直接得到系统输出,具有减少计算消耗的优势,而连续NT(CNT)算法在最近的研究中被证明为准确的质心降型算法。通过分析离散NT算法中求和运算和连续NT算法中的求积分运算,利用数值积分技术中牛顿-柯斯特求积公式将NT算法扩展成3种不同形式的加权NT(WNT)算法。在取相同的主变量采样率的情况下,3个计算机仿真例子表明了WNT算法比NT算法有更小的绝对误差且计算速度几乎相同,这使3种不同形式的WNT算法在区间二型模糊逻辑系统的实时应用上具有潜在的可行性和有效性。

关键词: 区间二型模糊逻辑系统, 降型, Nie-Tan算法, 加权Nie-Tan算法, 积分

Abstract:

Type-reduction is a kernel block in type-2 fuzzy logic systems. Nie-Tan algorithms that do not involve iterative process can obtain the systems output directly, and have the advantage of saving computational cost. Moreover, the continuous Nie-Tan (CNT) algorithms have been proved to be the accurate centroid type-reduction algorithms in most recent studies. By analyzing the sum operation in discrete NT algorithms and the integral operation version in continuous NT (CNT) algorithms, the paper extends NT algorithms to three different forms of Weighted Nie-Tan (WNT) algorithms according to the formulas of Newton-Cotes quadrature of numerical integration techniques. In case of choosing the same sampling rate of primary variables, three computer simulation examples show that WNT algorithms have smaller absolute errors than NT algorithms and the computation speed is almost the same, so they are potentially feasible and effective on the real-time applications of interval type-2 fuzzy logic systems.
 

Key words: interval type-2 fuzzy logic system, type-reduction, Nie-Tan algorithm, weighted Nie-Tan algorithm, integral