摘要
模糊划分应满足测量准则,这一点常被忽略。采用余弦平方函数作为隶属函数,解决了曲线型隶属函数在模糊化分后满足测量准则问题。T-S模糊模型后件由线性函数构成,以局部线性化方式实现全局非线性,通过分段性逼近辨识曲线,常用于系统辨识中,但其辨识参数多是其主要问题。比例规则后件的T-S模糊系统具有万能逼近特性,且构成参数少,用于非线性系统的辨识中,可以减少辨识参数,提高辨识速度。给出其在系统辨识中应用的方法,并通过仿真实例说明本方法的有效性。
The measure rules will be satisfied in fuzzy partition, which is often ignored. This paper solves the problem that satisfy measure rule after fuzzy partition uxing curve membership function by adapting cosine square function as membership function, The rule consequent of T-S fuzzy model is composed of linear function, ralizes nonlinear through local linearization method, and approach to the identification curve by subsection linearization,which is used in system identification frequently. But the main problem of it is its parameters is too much. T-S fuzzy model with proportion rule consequent has infinite characteristic ,and its parameters is fewer,can reduce the number of identification parameters when used in nonlinear system indetification, A model identification example is given to show the effectiveness of the method,
出处
《火力与指挥控制》
CSCD
北大核心
2005年第4期41-44,共4页
Fire Control & Command Control
基金
国家自然科学基金项目(60474019)
"十五"国防预研基金资助项目
关键词
余弦平方函数划分
模糊辨识
比例规则后件
cosine square function partition ,fuzzy identification ,proportion rule consequent
