抄録
The functional reasoning and the simplified reasoning, which are special cases of the so-called min-max-centroidal method, have been applied to construct a fuzzy controller or to model a system. It is not necessary known, however, for control engineers to rationally and easily design the conclusion part or any parameters in the conclusion part. In this paper, a new fuzzy reasoning is proposed so that the conclusion consists of a function of mean-values on each membership function in the antecedent. It is then shown that any parameters except for the mean-values in the conclusion can be rationally and easily designed as a slope of stable switching plane (or line) which is basically independent of the control rule, if the conclusion is formally regarded as a VSS (variable structure system) controller with the meanvalues in the antecedent as states. Thus, the laber required for designing the conclusion part can be reduced drastically by using the proposed method, compared with the traditional methods. The effectiveness of the proposed method is illustrated by simulating a fuzzy controller with two inputs-single output reasoning or a fuzzy controller with three inputs-single output reasoning.
