Modeling of a Nonlinear System Using Fuzzy Neural Networks, Application to Concentration Identification in a Chemical Reactor

Abstract

In this work, we study the application of fuzzy neural networks to the identification of uncertain, nonlinear systems. This approach was investigated using two different architectures:  ANFIS (Adaptive Network Fuzzy Inference System), based on the Takagi-Sugeno model.  FLP (Fuzzy Logic Processor), based on the Mamdani model. Each architecture is implemented through two methods: a series-parallel (one step ehead) model and a parallel (long range) model. An example of simulation is used, identification of concentration in a chemical reactor to demonstrate the effectiveness of these new approaches and to validate the results obtained