Research of Optical Microlens Array and Optical Waveguide Mold Fabrication in Photoresist using UV Proximity Printing
| Autor: | Tsung-Hung Lin, 林宗鴻 |
|---|---|
| Rok vydání: | 2008 |
| Druh dokumentu: | 學位論文 ; thesis |
| Popis: | 96 This thesis aims to develop the micro optical component production using UV-proximity printing, such as microlens array molds and optical waveguides are illustrated. In addition, artificial neural network and genetic algorithm technologies are combined with the Taguchi method to create a robust design for the high fill factor microlens array modeling. The finding of the robust parameters combination can result in the uniform microlens array fabrication. The optical component profiles were produced by utilizing a printing gap between the mask and photoresist substrate. The microlens array can use PDMS optical material of to replicate in mass production. A horizontal frustum optical waveguide with a both lateral and vertical taper structure was produced. The orthogonal and inclined masks with the diffraction effect were employed in the lithography process. A horizontal frustum optical waveguide provides a coupling efficiency higher than 52% from laser diode to the single-mode fiber when using 1550 nm laser diode. The artificial neural network and genetic algorithm technologies are combined with the robust design to reduce the variations in the focal length of the high fill factor microlens array. The orthogonal array was used for these experiments and calculated the S/N ratio of quality characteristic. The orthogonal array was used as the learning data for the artificial neural network to construct system model that can predict the focal length for arbitrary parameters setting. Then, the genetic algorithm was applied to obtain the parameters setting. The ANN/GA model is compared with the experimental result, the predicted error is about 2%. The experimental results prove that 34% and 56% reductions in sensitivity variation can be achieved using the Taguchi method and the ANN/GA model, respectively. |
| Databáze: | Networked Digital Library of Theses & Dissertations |
| Externí odkaz: |
|