双输入单输出潜遗传变论域模糊控制算法及其收敛性

Double-input and single-output fuzzy control algorithm with potentially-inherited variable universe and its convergence

研究了一种变论域模糊控制算法及其收敛性问题.首先,针对常见的双输入单输出模糊控制器,提出了一种新型的潜遗传变论域模糊控制算法,通过使用模糊单值转换函数,简化了新算法推理后件的变论域操作,显著地减少了计算量.然后,从理论上证明了新算法的收敛性.最后,为了验证算法的有效性,本文将新算法和已有的两个典型算法做了倒立摆实验对比.结果表明,相对于已有算法,由新算法构成的控制器在控制精度上有明显的提高.

The problems of variable-universe fuzzy control algorithm and its convergence are studied in this paper. First, a novel potentially-inherited variable-universe fuzzy control algorithm is proposed for conventional double-input and single-output fuzzy controllers. By using a fuzzy-singleton transition function, the computation of the algorithm is significantly reduced because its operations are simplified on the variable universe of the consequent part of inference. Then, the convergence of the algorithm is proved theoretically. Finally, to confirm the algorithm＇s effectiveness, comparison experiments on inverted pendulum are done using two typical algorithms and the new one. The results of experiments show that, compared with the two typical algorithms, the controller based on the new algorithm obtains notable improvement in control precision.