Extracting the dielectric constant of materials using ABC-based ANNs and NRW algorithms


ÖZTÜRK T. , Elhawil A., DÜĞENCİ M. , Unal I., ULUER İ.

JOURNAL OF ELECTROMAGNETIC WAVES AND APPLICATIONS, cilt.30, ss.1785-1799, 2016 (SCI İndekslerine Giren Dergi) identifier identifier

  • Cilt numarası: 30 Konu: 13
  • Basım Tarihi: 2016
  • Doi Numarası: 10.1080/09205071.2016.1215266
  • Dergi Adı: JOURNAL OF ELECTROMAGNETIC WAVES AND APPLICATIONS
  • Sayfa Sayıları: ss.1785-1799

Özet

Five different Nicolson-Ross-Weir (NRW) extracting techniques are used to extract the dielectric constants of Teflon, Rexolite, Glass (borosilicate and soda-lime), Paper, and Ultralam 3850HT from S-parameters. The results of these extraction techniques are used to train the Artificial Neural Networks. In order to improve the accuracy of the results, the weights of ANNs are calculated using artificial bee colony estimation method. The results are compared with that obtained using NRW, Newton-Raphson, and genetic algorithm. The obtained results indicate that the proposed model gives good extracted parameters as compared with the previously published results.