考虑重要输入变量选择的非线性系统模糊辨识

Fuzzy identification of nonlinear system considering the selection of important input variables

中文关键词  模糊系统  输入变量选择  Takagi-Sugeno(T–S)模型  模糊聚类(FCM)算法  Gaussian函数  气动加载系统

英文关键词  fuzzy systems  input variable selection  Takagi-Sugeno (T–S) model  fuzzy C–means (FCM) algorithm  Gaussian function  pneumatic loading system

中文摘要

      针对有数百个可能输入的复杂非线性动态系统模糊建模问题, 本文提出一种新的考虑重要输入变量选择 的模糊辨识方法. 首先采用两阶段模糊曲线方法(TSFC)从大量可选择的输入变量中给出各输入变量与输出之间的 关联度权重, 根据输入变量指标快速选择出重要的输入变量, 然后采用模糊聚类(FCM)和高斯(Gaussian)型隶属函 数确定模糊模型前提参数, 采用递推最小二乘(RLS)辨识出模糊模型结论参数. 最后通过对Mackey-Glass混沌时间 序列和Box-Jenkins煤气炉数据两个国际标准例题模糊建模验证了该方法的有效性, 并将该方法应用到一个实际气 动变载荷加载系统的模糊建模中, 验证了该方法的实用性.

英文摘要

      Aiming at the fuzzy modeling of complex nonlinear dynamic systems with hundreds of possible inputs, this paper proposes a new fuzzy identification method considering the selection of important input variables. First, the two stage fuzzy curves method (TSFC) is used to give the weight of the correlation between each input variable and the output from a large number of selectable input variables, and the important input variable is quickly selected according to the input variable index. Then fuzzy C–means clustering (FCM) and Gaussian membership functions are used to determine the premise parameters of the fuzzy model, and recursive least squares (RLS) are used to identify the conclusion parameters of the fuzzy model. Finally, the effectiveness of the method is verified by fuzzy modeling of two international standard examples of Mackey-Glass Chaotic System and Box-Jenkins system. In order to verify the practicability of this method, this method is applied to the fuzzy modeling of an actual variable load pneumatic loading system.