A New On-Line Structure and Parameter Learning Architecture for Fuzzy Modeling, Based on Neural and Fuzzy Techniques

Function al reasoning or the Takagi-Sugeno-Kang model is a fuzzy reasoning method aiming at numerical accuracy and has found wide use in fuzzy modeling. ln this method, each rule consists of a fuzzy implication and a functional consequence part. ln this work, a new, online identification method for such a system is presented, for supervised learning tasks. Structure identification is executed by a fuzzy ART learning module, following the procedure_of splitting fuzzy rules that tend to produce high output error, Fuzzy rules are also added wherever the error exceeds a threshold. Ali parameters are fine tuned by the d rule, a basic learning technique in neural networks. Computer simulation exhibits the potentials of this approach, which is tested with weil known benchmarks, yielding excellent results.