Online Identification of Self‐Organizing Fuzzy Neural Networks for Modeling Time‐Varying Complex Systems
Girijesh Prasad, Guannan Leng, T.M. McGinnity, Damien Coyle · 2010
The self-organizing fuzzy neural networks (SOFNN) have enhanced ability to identify adaptive models for representing nonlinear and time-varying complex systems. This chapter presents an algorithm for online identification of SOFNN. The SOFNN provides a singleton or Takagi-Sugeno (TS)-type fuzzy model. It therefore facilitates extracting fuzzy rules from the training data. The algorithm guarantees the convergence of both the estimation error and linear network parameters. It generates a fuzzy neural model with high accuracy and compact structure. Superior performance of the algorithm is demonstrated through its applications for function approximation, system identification, and time-series prediction in both industrial and biological systems. The learning process of the SOFNN includes the structure learning and the parameter learning. The structure learning attempts to achieve an economical network size with a self-organizing approach. As a result of this, neurons in the ellipsoidal basis function (EBF) layer are augmented or pruned dynamically in the learning process. Controlled Vocabulary Terms fuzzy neural nets; identification; learning (artificial intelligence); network parameters; time-varying systems