中国科学院新疆天文台机构知识库

针对模糊控制器内部工作机理未知的问题,提出一种区间二型(IT2)模糊PI/PD控制器,并推导了该IT2模糊控制器的解析结构。该模糊控制器采用Mamdani型规则形式,为每个输入变量仅分配两个IT2模糊集合,具有规则数量少、规则不确定性影响小、控制输出平滑等优点。根据IT2模糊控制器的降型特点,将整个输入空间划分为15个分区,进而推导出在每个分区上控制输出的数学表达式,证明出该IT2模糊控制器等效为具有变增益的非线性比例积分(PI)或比例微分(PD)控制器。最后,在仿真和实验中与相对应的一型模糊控制器和传统PI控制器的控制效果进行对比,验证了该IT2模糊PI/PD控制器的有效性。

Aiming at the problem of unknown internal working mechanism for the fuzzy controller，a class of interval type-2 ( IT2) fuzzy PI /PD controller was proposed，and the analytical structure of this IT2 fuzzy controller was derived． Mamdani rule form was used in fuzzy controller and each input variable only has two IT2 fuzzy sets． The fuzzy controller has the advantages of fewer rules，smaller uncertainty effects of fuzzy rules and smoothing control output． The proposed IT2 fuzzy controller is proved to be equivalent to a nonlinear proportional integral ( PI) or proportional differential ( PD) controller with variable gains by dividing the input space into 15 partitions and calculating the mathematical expression of the control signal on each partition． Finally，the effectiveness of the IT2 fuzzy controller were verified by comparing the control effects with its type-1 fuzzy controller and traditional PI controller in the simulation and experiment．