An Artificial Neural Network (ANN) model for the pull-down voltage analysis of Radio Frequency (RF) NEMS(Nano Electro-Mechanical) switch is presented in this paper. The fixed-fixed beam and cantilever beam structure have been chosen and analyzed in terms of pull-down voltage with the variation of its physical parameter values. An ANN model was trained with several training algorithms to achieve higher performance and computational efficiency.The pull-down voltage predicted from theLevenberg-Marquardt backpropagationANN modelhas been compared with previous results presented in experimental and theoretical studies available in the literature.The proposed ANN model was also used to investigate the geometric and physical influence of the nano-switch on the pull-in voltage, and results were reported.The proposed ANN model performs better, with less Mean Square Error (MSE) of 0.000045 and 0.000092 for cantilever and fixed-fixed beam structures respectively.

An Artificial Neural Network (ANN) model for the pull-down voltage analysis of Radio Frequency (RF) NEMS (Nano Electro-Mechanical) switch is presented in this paper. The fixed-fixed beam and cantilever beam structure have been chosen and analyzed in terms of pull-down voltage with the variation of its physical parameter values. An ANN model was trained with several training algorithms to achieve higher performance and computational efficiency. The pull-down voltage predicted from the Levenberg-Marquardt backpropagation ANN model has been compared with previous results presented in experimental and theoretical studies available in the literature. The proposed ANN model was also used to investigate the geometric and physical influence of the nano-switch on the pull-in voltage, and results were reported. The proposed ANN model performs better, with less Mean Square Error (MSE) of 0.000045 and 0.000092 for cantilever and fixed-fixed beam structures respectively.