Application of Intelligent Neural Networks to Prediction of Micro-electroforming for U-Type Micro-cavity

  • Authors:

  • Sheau-Wen Shiah;Pai-Yu Chang;Tzeng-Yuan Heh;Po-Hung Lin;Fu-Cheng Yang

  • Affiliations:

  • Department of Power Vehicle and System Engineering, Chung Cheng Institute of Technology, National Defense University,;Department of Product Design, Fortune Institute of Technology,;Department of Power Vehicle and System Engineering, Chung Cheng Institute of Technology, National Defense University,;Department of Power Vehicle and System Engineering, Chung Cheng Institute of Technology, National Defense University,;Department of Power Vehicle and System Engineering, Chung Cheng Institute of Technology, National Defense University,

  • Venue:
  • KES '07 Knowledge-Based Intelligent Information and Engineering Systems and the XVII Italian Workshop on Neural Networks on Proceedings of the 11th International Conference
  • Year:
  • 2007

Quantified Score

Hi-index 0.00

Visualization

Abstract

The physical transport mechanisms of micro-electroforming are innately complex and generally entail a great effort to comprehend the nature through either experimental measurements or theoretical simulations. In this study, the approach of using neural networks is implemented for demonstrating its effectiveness in the real-time determination of an electrolyte to achieve on U-type micro-cavities. Three intelligent back-propagation neural networks are established via the training process with the Kondo's database to predict the distributions for cathode various parametersPeand for various cavity widths are determined. Comparisons of the predictions with the test target vectors indicate that the averaged root-mean-squared errors from three back-propagation neural networks are well. This study also examines in detail the effects of various network parameters, including number of hidden nodes, transfer function type, number of training pairs, learning rate-increasing ratio, learning rate-decreasing ratio, and momentum value, on the performance of neural networks.