Fuzzy rules based qualitative modeling
Pei-Hwa Huang, Yu-Shuo Chang · Proceedings of IEEE 5th International Fuzzy Systems · 2002
This paper suggests a fuzzy-model-based procedure for deriving the qualitative description of single-input-single-output systems. From the system numerical input-output data, a fuzzy model is first constructed using the mountain method for structure identification and back-propagation for parameter identification. A procedure consisting of a fuzzy clustering method and a rule processing algorithm is proposed for processing the constructed fuzzy model rules to extract the system qualitative properties. An example is given for illustration.