基于多分辨率分析的T-S模糊系统

T-S fuzzy system based on multi-resolution analysis

中文关键词  T-S模糊系统  多分辨率分析  B-样条尺度函数  结构辨识

英文关键词  T-S fuzzy system  multi-resolution analysis  B-spline scale function  structure identification

中文摘要

      目前模糊系统缺乏保持辨识精度与模糊语义最佳折中的有效辨识方法,其主要原因在于缺乏系统的优化结构辨识方法.因此,本文从时-频域角度构造出基于多分辨率分析的T-S(Takagi-Sugeno)模糊系统拓扑结构.然后,采用具有多分辨率特点的B-样条尺度函构造模糊隶属函数,根据投影算法和模糊隶属函数相异测度给出了模糊系统结构辨识算法.仿真结果验证了这种模糊系统及其结构辨识算法的有效性.

英文摘要

      The existing methods of fuzzy system identification hardly keep good trade-off between precision and fuzzy meaning.One of the main reasons is that we are short of systematic and optimal structure identification methods.Thus,a novel topology of T-S(Takagi-Sugeno) fuzzy system is constructed based on multi-resolution analysis from the view of time-frequency domain.Then,the fuzzy membership functions are constructed by B-spline scale functions with multi-resolution characteristic;an adaptive learning algorithm for structure identification of fuzzy system is proposed based on projection algorithm and dissimilarity measure of fuzzy membership functions.Simulation results show the efficiency of the fuzzy system and its structure identification algorithm.