### Abstract

This paper describes a novel idea for designing a fuzzy-neural network for modeling of nonlinear system dynamics. The presented approach assumes a state-space representation in order to obtain a more compact form of the model, without statement of a great number of parameters needed to represent a nonlinear behavior. To increase the flexibility of the network, simple Takagi-Sugeno inferences are used to estimate the current system states, by a set of a multiple local linear state estimators. Afterwards, the output of the network is defined, as function of the current and estimated system parameters. A simple learning algorithm based on two step Gradient descent procedure to adjust the network parameters, is applied. The potentials of the proposed modeling network are demonstrated by simulation experiments to model an oscillating pendulum and a nonlinear drying plant.

Original language | English |
---|---|

Title of host publication | 2014 IEEE International Symposium on Innovations in Intelligent Systems and Applications (INISTA) |

Publisher | Institute of Electrical and Electronic Engineers IEEE |

Pages | 212-217 |

ISBN (Electronic) | 978-1-4799-3020-3 |

ISBN (Print) | 978-1-4799-3019-7 |

DOIs | |

Publication status | Published - 1 Jan 2014 |

MoE publication type | A4 Article in a conference publication |

Event | 2014 IEEE International Symposium on Innovations in Intelligent Systems and Applications, INISTA 2014 - Alberobello, Italy Duration: 23 Jun 2014 → 25 Jun 2014 |

### Conference

Conference | 2014 IEEE International Symposium on Innovations in Intelligent Systems and Applications, INISTA 2014 |
---|---|

Country | Italy |

City | Alberobello |

Period | 23/06/14 → 25/06/14 |

### Fingerprint

### Keywords

- fuzzy-neural systems
- Gradient descent
- State-space
- Takagi-Sugeno

### Cite this

*2014 IEEE International Symposium on Innovations in Intelligent Systems and Applications (INISTA)*(pp. 212-217). [6873620] Institute of Electrical and Electronic Engineers IEEE. https://doi.org/10.1109/INISTA.2014.6873620

}

*2014 IEEE International Symposium on Innovations in Intelligent Systems and Applications (INISTA).*, 6873620, Institute of Electrical and Electronic Engineers IEEE, pp. 212-217, 2014 IEEE International Symposium on Innovations in Intelligent Systems and Applications, INISTA 2014, Alberobello, Italy, 23/06/14. https://doi.org/10.1109/INISTA.2014.6873620

**State-space fuzzy-neural network for modeling of nonlinear dynamics.** / Todorov, Yancho; Terziyska, Margarita.

Research output: Chapter in Book/Report/Conference proceeding › Conference article in proceedings › Scientific › peer-review

TY - GEN

T1 - State-space fuzzy-neural network for modeling of nonlinear dynamics

AU - Todorov, Yancho

AU - Terziyska, Margarita

PY - 2014/1/1

Y1 - 2014/1/1

N2 - This paper describes a novel idea for designing a fuzzy-neural network for modeling of nonlinear system dynamics. The presented approach assumes a state-space representation in order to obtain a more compact form of the model, without statement of a great number of parameters needed to represent a nonlinear behavior. To increase the flexibility of the network, simple Takagi-Sugeno inferences are used to estimate the current system states, by a set of a multiple local linear state estimators. Afterwards, the output of the network is defined, as function of the current and estimated system parameters. A simple learning algorithm based on two step Gradient descent procedure to adjust the network parameters, is applied. The potentials of the proposed modeling network are demonstrated by simulation experiments to model an oscillating pendulum and a nonlinear drying plant.

AB - This paper describes a novel idea for designing a fuzzy-neural network for modeling of nonlinear system dynamics. The presented approach assumes a state-space representation in order to obtain a more compact form of the model, without statement of a great number of parameters needed to represent a nonlinear behavior. To increase the flexibility of the network, simple Takagi-Sugeno inferences are used to estimate the current system states, by a set of a multiple local linear state estimators. Afterwards, the output of the network is defined, as function of the current and estimated system parameters. A simple learning algorithm based on two step Gradient descent procedure to adjust the network parameters, is applied. The potentials of the proposed modeling network are demonstrated by simulation experiments to model an oscillating pendulum and a nonlinear drying plant.

KW - fuzzy-neural systems

KW - Gradient descent

KW - State-space

KW - Takagi-Sugeno

UR - http://www.scopus.com/inward/record.url?scp=84906689506&partnerID=8YFLogxK

U2 - 10.1109/INISTA.2014.6873620

DO - 10.1109/INISTA.2014.6873620

M3 - Conference article in proceedings

SN - 978-1-4799-3019-7

SP - 212

EP - 217

BT - 2014 IEEE International Symposium on Innovations in Intelligent Systems and Applications (INISTA)

PB - Institute of Electrical and Electronic Engineers IEEE

ER -