Fuzzy system identification using rule-based models

The purpose of the paper was to examine the methods used to develop fuzzy models of SISO systems which may be thought of as "rule-based" using non-recursive techniques. First, we look at the mathematical concepts used to develop the models. Second, special constraints used in the creation of the fuzzy models are specified. Third, fuzzy models of a simple linear system are developed. The models are used to determine the effects of incorrect or overparameterization on the effects of the model performance. Finally, fuzzy models of a simple non-linear system are developed and their performance examined.<
>