臺灣博碩士論文加值系統

為了降低生產成本並提高產能，高精密的光收發模組自動化封裝設備不斷推陳出新，然而讓這些設備發揮功能之關鍵為：必須具備高速且穩定的尋光演算控制軟體，以控制這些設備。 本文主要針對光收發模組封裝製程中之耦光技術進行研究，主要著重於細尋光搜尋技術。 研究中提出一新的耦光演算法與技術，經實驗證實，此一耦光技術大幅提升光收發模組自動化封裝之良率與生產速度。
目前眾多尋光演算法中，由於KMSM法具備快速收斂及容易執行等特性，已證實為目前最適合應用於自動化之尋光演算法。 但此方法僅限於使用於平面耦光，若直接應用於多維度耦光，將容易受到多維空間中局部峰值之影響而發生誤判情形，而導致尋光失敗。 本文所提出耦光技術之主要構想為，將多維度耦光分為平面對準及光軸對準兩個階段，然後整合KMSM演算法與模糊邏輯法則，發展出一套新的演算法，以有效提高尋光演算法之效率與良率，此技術稱為2S-FSM。
所發展之2S-FSM，根據光學物理特性，定義最佳化初始參數，然後將參數最佳化之結果，整理為簡單可利用之方程式，以增加實際應用之便利性。 此外，本文利用實際雷射二極體之光功率分佈建構一個完整數學模式，應用此一模式，除了可以藉由改變參數而研究不同型式之雷射二極體外，更可藉由此模組測試不同尋光演算法之效能，以縮短演算法開發時程。
為比較所提演算法及傳統演算法之差異，我們以相同的起始位置，在相同的條件之下，分別利用此二種演算法，利用十組不同的光收發模組與實際雷射封裝機台(MORITEX FZ-82A)，進行一百次的耦光實驗。 實驗結果顯示，所提出的尋光演算法，耦光成功率可高達到100% ，而其生產速度與傳統演算法相較，提升了2-4倍。

Of various issues in the transceiver packaging, automating the fiber-optic alignment operation is one of the most challenges and also a critical manufacturing technique, because it dominates the cost and the yield rate in manufacture of the transceiver. The fiber-laser alignment functions to identify a position at which the fiber-laser coupling efficiency is a maximum. The alignment operation involves coarse and fine alignment processes. This thesis focuses on the fine alignment process, especially for the multi-degree-of-freedom (DOF) alignment.
Among the alignment algorithms used, the King’s modified simplex method (KMSM) has been proven to be very suitable for the fine alignment process because of its fast convergence and ease of implementation. However, direct implementation of this method on the multi-DOF alignment would lead to local peak sticking problem due to the power distribution nature of the laser diode (LD).
In this thesis, a two-stage fuzzy simplex algorithm (hereafter called the 2S-FSM) is proposed, in which the search of the maximum coupling efficiency position (i.e., the global peak) is divided into two stages, namely, the 2D planar alignment and the 3D optic axis alignment. In the searching operations, the fuzzy logic is integrated into the conventional KMSM to prevent the simplex from sticking on local peaks and meanwhile accelerates the search process. This novel strategy is proven to be capable of dealing with the problem of local peak sticking in the multi-DOF alignment.
Experiments have demonstrated that the proposed idea is more effective than the conventional search methods in terms of the alignment success rate and the searching speed in all of the 100 alignment trials for 10 different LD-fiber pairs, regardless of the initial positions.

摘要 I
Abstract III
誌謝 V
Table of Contents VI
List of Figures VII
List of Tables IX
Chapter 1 Introduction 1
Chapter 2 The Simplex Method 5
2.1 The King’s Modified Simplex Method (KMSM) 5
2.2 Fuzzy Simplex method (FSM) 8
2.2.2 Outputs of the fuzzy modules 16
2.2.3 The fuzzy modules 17
Chapter 3 The Proposed Strategy 22
Chapter 4 Numerical Simulations 28
4.1 Optic power distribution model of a laser diode 28
4.2 Simulations for 2D planar alignment 33
4.3 Simulations for the 3D optic axis alignment 39
Chapter 5 Experiments 44
5.1 Experimental apparatus 44
5.2 Search implementation 47
Chapter 6 Conclusions 54
Reference 55
Appendix - Experimental data 55
Biosketch of Author 63

Reference
1.Sterling D. J., “Technician’s Guide to Fiber Optic 3rd edition,” Thomson Learning, 2003.
2.Cohen M. S., Cina M. F., Opyrsko M. M., Speidell J. L., Canora F. J., DeFranza M. J., and Bassous E., “Packaging of High-Density Fiber/Laser Modules Using Passive Alignment Techniques,” IEEE Transactions on Components, Hybrids, and Manufacturing Technology, Vol. 15, Issue 6, pp. 944 –955,1992.
3.Dautartas M. F., Blonder G. E., Wong Y. H., and Chen Y. C., “A Self-Aligned Optical Subassembly for Multi-Mode Devices,” IEEE Transactions on Components, Packaging, and Manufacturing Technology, Part B, Vol. 18, Issue 3, pp. 552 -557, 1995.
4.Hauffe R., Siebel U., Petermann K., Moosburger R., Kropp J. R., and Arndt F., “Methods for Passive Chip Coupling of integrated Optical Devices,” Electronic Component and Technology Conference, pp. 238-243, 2000.
5.曾國宗，劉松河，蔡武融，光收發元件製程技術，機械工業雜誌258期，137-158頁，2004。
6.Russell S., and Norvig P., “Artificial Intelligence - A Modern Approach,” Prentice-Hall, 1995.
7.Tang Z., Zhang R., and Shi F. G., “Effects of angular misalignment on fiber-optic alignment automation,” Optics Communications, Vol.196, pp. 173-180, 2001.

8.Mondal S. K., and Shi F. G., “Optimization of Fiber-Laser Coupling with Respect to Both the Coupling Efficiency and Carrier-to-Noise Ratio：A Novel Design,” IEEE Electronic Components and Technology Conference, pp. 1-4, 2002.
9.Pham D. T. and Castellani M., “Inelligent control of fiber optic components assembly,” Proc. ImechE, part B, Vol. 215, pp. 1177-1189, 2001.
10.Melles Griot, “Application Note – NanoTrak Operation,” http://hardware.mellesgriot.com/, pp. 1-9, 2003.
11.Dantzig G. B., Linear Programming and Extensions, Princeton University Press, Princeton, NJ, 1963.
12.Spendley W., Hext G. R., and Himsworth F. R., “Sequential Application of Simplex Designs in Optimization and Evolutionary Operation,” Techno metrics Vol. 4, pp. 441-461, 1962.
13.Nelder J. A., and Mead R., “A Simplex Method for Function Minimization,” Computer J., Vol. 7, pp. 308-313, 1965.
14.King P. G., PhD Thesis, UNIPLEX: Single-Factor Optimization of Response in the, Emory University, Atlanta, GA, 1974.
15.Zhang R., and Shi F. G., “A Novel Algorithm for Fiber-Optic Alignment Automation,” IEEE Electronic Components and Technology Conference, pp. 256-260, 2003.
16.Chen W. S., The Study of Automation Method for Multi-DOF Fiber-Optic Alignment, National Pingtung University of Science and Technology, Taiwan, 2004.
17.Tseng C. Y., and Wang J. P., “Automation of Multi-Degree-of-Freedom Fiber-Optic Alignment Using a Modified Simplex Method,” International Journal of Machine Tools and Manufacture, Vol. 45, Issue 10, pp. 1109-1119, 2005.
18.Zhang R., Guo J., and Shi F., “Fast Fiber-Laser Alignment: Beam Spot-Size Method,” journal of Lightwave Technology, Vol. 23, No. 3, pp. 1083- 1087, 2005.
19.Tseng C. Y., and Wang J. P., “Study of Automatic Alignment Technology for Fiber-Optic Devices Packaging,” IEEE International Conference on Mechatronics, pp. 7-12, 2005.
20.Tseng C. Y., Chang C. L., Lin C. F., and Wang J. P., “A Fuzzy Simplex Algorithm for Active Fiber-Laser Alignment,” Journal of the Chinese Society of Mechanical Engineers, Vol. 26, No. 6, pp. 759-764, 2005.
21.Ryan P. B., Barr R. L., and Todd H. D., “Simplex techniques for nonlinear optimization,” Analytical Chemistry, pp. 1460-1467, 1980.
22.Lewis A., Abramson D., and Peachey T., “RSCS: A Parallel Simplex Algorithm for the Nimrod/O Optimization Toolset”, International Symposium on Parallel and Distributed Computing, HeteroPar’04, 2004.
23.Passino KM, and Yurkovich S., “Fuzzy Control”, Addison Wesley, 1998.
24. Zhang R., and Shi F. G., “A Novel Algorithm for Fiber-Optic Alignment Automation”, IEEE Transactions on Advanced Packaging, Vol. 27, Issue 1, pp. 173-178, 2004.