基于分变量模糊蕴涵关系的无规则模糊逻辑系统的自适应控制应用

Adaptive control application of fuzzy logic system without any rules based on fuzzy implication relations of each variable

常用的基于模糊if-then规则的模糊逻辑系统在进行控制设计应用时常会遇到规则爆炸问题,为此,提出了一种新的无规则模糊逻辑系统。首先利用从属于系统各分变量论域上开覆盖的单位分解,将模糊知识库中的专家模糊语言值信息进行整合,得到相应的分变量模糊蕴涵关系,再由模糊推理及解模糊化得到系统输出;然后,针对一类非线性不确定系统,先利用伸缩器和饱和器构造出扩展的无规则模糊逻辑系统,再设计被控系统的自适应稳定控制器;最后通过仿真算例验证了该方法的有效性。这种方法的优越性在于新的模糊逻辑系统的构造与自适应律的设计是分离的,在线估计的参数数目明显减少,有效地解决了规则爆炸问题。

The commonly used fuzzy logic systems were mostly based on fuzzy rules in the form of if-then rules, which often encountered the rule explosion problems in control design applications. To deal with this problem, this paper proposed a new fuzzy logic system. Firstly by using the open covering partition of unity subordinated to each system variable domain, the expert fuzzy language knowledge in the fuzzy knowledge base was integrated into fuzzy implication relations, and then adopted the fuzzy reasoning and defuzzifying operators to get the output. Secondly, for a class of nonlinear uncertain system, the extended fuzzy logic system was constructed with scalers and saturators, then designed a stable adaptive controller for the controlled system. Finally, a simulation example illustrated the validity of the proposed method. The superiority of the suggested design scheme is that the process of constructing the new fuzzy logic system and designing the adaptive laws is separated, thus the number of on-line parameters is decreased significantly, which solves the rule explosion problem effectively.