ANFIS MODELS FOR THE QUASISTATIC ANALYSIS OF COPLANAR STRIP LINE STRUCTURES

In this work, computer-aided design models based on adaptive-network-based fuzzy inference system (ANFIS) for the quasistatic analysis of three different coplanar strip line structures are presented. These strip line structures are conventional coplanar strip lines, asymmetrical coplanar strip lines with infinitely wide strip, and asymmetrical coplanar strip lines with infinitely thick dielectric substrate. The design parameters of the proposed ANFIS models are optimally determined by using four different optimization algorithms, hybrid learning, simulated annealing, least-squares, and genetic algorithm. When the performances of ANFIS models are compared with each other, the best results for training and test are obtained from the models trained with hybrid learning algorithm. There is a good agreement among the results of ANFIS models, quasistatic analysis, a full-wave. electromagnetic simulator IE3D, and experimental works realized in this study. (C) 2010 Wiley Periodicals, Inc. Microwave Opt Technol Lett 52: 1990-1996,2010; Published online in Wiley InterScience (www.interscience.wiley.com). DOI 10.1002/mop.25372