Abstract
In the case of the classical ANNs such as the MLP, the problem reduces to the selection of the number of layers and the neurons in a particular layer. If the obtained network does not satisfy prespecified requirements, then a new network structure is selected and the parameters estimation is repeated once again. The determination of the appropriate structure and parameters of the model in the presented way is a complex task. Furthermore, an arbitrary selection of the ANN structure can be a source of the model uncertainty.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
Copyright information
© 2014 Springer International Publishing Switzerland
About this chapter
Cite this chapter
Mrugalski, M. (2014). GMDH Networks in Robust Fault Detection of Dynamic Non-linear Systems. In: Advanced Neural Network-Based Computational Schemes for Robust Fault Diagnosis. Studies in Computational Intelligence, vol 510. Springer, Cham. https://doi.org/10.1007/978-3-319-01547-7_5
Download citation
DOI: https://doi.org/10.1007/978-3-319-01547-7_5
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-01546-0
Online ISBN: 978-3-319-01547-7
eBook Packages: EngineeringEngineering (R0)