Quasi-static models based on artificial neural networks for calculating the characteristic parameters of multilayer cylindrical coplanar waveguide and strip line
| Autor: | Türkmen, Mustafa, Yıldız, Celal |
|---|---|
| Jazyk: | angličtina |
| Rok vydání: | 2008 |
| Předmět: | |
| Popis: | In this paper, two different neural models are proposedfor calculating the quasi-static parameters of multilayer cylindricalcoplanar waveguides and strip lines. These models were basicallydeveloped by training the artificial neural networks with the numericalresults of quasi-static analysis. Neural models were trained withfour different learning algorithms to obtain better performance andfaster convergence with simpler structure. When the performances ofneural models are compared with each other, the best test results areobtained from the multilayered perceptrons trained by the LevenbergMarquardt algorithm. The results obtained from the neural modelsare in very good agreements with the theoretical results available inthe literature. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|