Title :
A pole estimation approach to the synthesis of the dynamic GMDH neural networks
Author :
Mrugalski, Marcin ; Witczak, Marcin ; Korbicz, Józef
Author_Institution :
Inst. of Control & Comput. Eng., Univ. of Zielona Gora, Zielona Góra, Poland
Abstract :
This paper presents an identification method of Multi-Input Multi-Output (MIMO) dynamic systems based on the Group Method of Data Handing. In particular, a new structure of the dynamic neuron is proposed. The synthesis of the neural model is performed with the application of the pole estimation approach. The final part of this work contains an illustrative example regarding the application of the proposed approach to a fault detection of the tunnel furnace.
Keywords :
MIMO systems; control system synthesis; estimation theory; fault diagnosis; identification; neurocontrollers; pole assignment; MIMO dynamic systems; dynamic GMDH neural networks; dynamic neuron; fault detection; group method of data handling; identification method; multi-input multi-output dynamic systems; neural model synthesis; pole estimation approach; tunnel furnace; Approximation algorithms; Artificial neural networks; Furnaces; MIMO; Mathematical model; Neurons; Stability analysis;
Conference_Titel :
Control and Fault-Tolerant Systems (SysTol), 2010 Conference on
Conference_Location :
Nice
Print_ISBN :
978-1-4244-8153-8
DOI :
10.1109/SYSTOL.2010.5675987
