期刊名称：International Journal of Computer Science and Network
印刷版ISSN：2277-5420
出版年度：2018
卷号：7
期号：2
页码：115-125
出版社：IJCSN publisher
摘要：Artificial Neural Networks is considered as one of the successful machine learning technique to model expert behavioral systems and it is competent enough to minimize human efforts in manual prediction. This research anticipates the applicability of the Adaptive Neuro-Fuzzy Modeling approach in the Dielectric Loss prediction. Prediction of Dielectric loss parameter has wide range of applicability in electric circuits of radio, and television systems. The difficulty lies in Dielectric loss parameter is its high dependency on the nature of dielectric material and at different frequency. Doped and undoped Terbium Manganite is chosen as dielectric material to perform comparative analysis. The Adaptive Neuro Fuzzy Inference System, with its inherent knowledge representation mechanism, nonlinear behavior and adaptive control property can replace almost all basic predictor neural networks. This hybrid model establishes the superior capability in prediction over Feed forward back-propagation networks. Finally an analysis is made statistically between Hybrid learning and back-propagation learning mechanism in Adaptive Neuro-Fuzzy inference system to achieve best suited learning algorithm in dielectric loss prediction.
关键词：Adaptive Neuro-Fuzzy Systems; Dielectric Loss; Doping; Tolerance; Transfer Functions; Semiconductor
