應用多重品質特性最佳化方法於隱形眼鏡光學設計之研究

The Optimal Parameter Design of Multiple Quality Characteristics for Optical Contact Lens Design

葉喆文(Che-Wen Yeh)

指導教授 ： 顏志達

國立虎尾科技大學/電機資訊學院/電機工程研究所/碩士(2013年)

摘要

眼鏡與隱形眼鏡歷史發展久遠，其設計與人息息相關，尤其是隱形眼鏡的光學設計部分，除了人眼光學參數之外，對於生理上相關參數的設計與決定更為重要。其主要功能仍與傳統眼鏡相同為視力矯正，包括：近視、遠視、散光、老花、像差等。
在複雜的隱形眼鏡光學設計流程中有許多不確定的影響變數，如何降低驗配後各種非理想變因以增加配戴者舒適度成為一大課題。在本研究中，我們將重點放在光學像差對於隱形眼鏡影響上面，而一般隱形眼鏡在球面像差(Spherical Aberration, SA)的值似乎較大。如果我們採用光學軟體CODE V優化設計結合田口方法(Taguchi Methods)，以降低球面像差(Spherical Aberration, SA)為目標，進行田口實驗以找出最佳製成的光學系統參數，球面像差值固然降低，但調製傳遞函數(Modulation transfer function, MTF)和正切彗星像差(Tangential Coma, TCO)的效能會變得相對較差。
因此在本研究中主要目標為使用集成演算法技術，以達到多重品質特性(SA，TCO 和MTF) 的最佳化。論文中提出兩種最佳化技術來達成多重品質最佳化：第一種方式為主成分分析法(principal component analysis, PCA) 搭配田口實驗來實現多重品質特性最佳化演算法，結果顯示與單純使用CODE V軟體優化後相比，球差、慧差與調製傳遞函數分別改善25.6%、91.88%與2.4%；另一種方法則為類神經網路(Neural Network, NN)結合基因演算法(Genetic Algorithms, GA)，來找出最佳化製成的參數，與單純使用CODE V優化後相比，球差、慧差與調製傳遞函數分別改善50.03%、45.78%與24.7%。從以上結果所示，所提出的兩種集成演算法技術的確能達到多重品質特性最佳化目的。

關鍵字

多重品質特性 ； 田口法 ； 主成分分析法 ； 類神經網路 ； 基因演算法

並列摘要

The optical design of contact lenses and glasses have been development for a long time. Besides the optical parameters of human eye, the physiological relevant parameters in the design issue become more important. The function of contact lens is still the same as traditional glasses such as eyesight correction, myopia, hyperopia, astigmatism, presbyopia and aberration.
There are many uncertain variables influence in complex contact lenses design process. How to reduce uncomfortable situation when wearing a pair of glasses is more important issue for designer. In this study, we focus on the impact of aberrations for contact lenses. The general contact lenses spherical aberration (SA) value is bigger than other aberrations. If we reduce spherical aberration as the goal, using the Taguchi method combining with the optical design software CODE V as our optimal method to perform experiments to find the best optical system parameters for contact lens. Although the SA is reduced but the performance of the modulation transfer function (MTF) and Tangential coma (TCO) will become more seriously.
In this study, the main objective is to use multi-objective optimal algorithm to achieve multiple quality characteristics (SA, TCO and MTF) optimization. There are two optimization algorithms to achieve multiple quality optimizations are proposed in our thesis. First, the principal component analysis (PCA) with Taguchi method to achieve multiple quality characteristics optimization algorithm is present. The simulation results show that comparing the CODE V built-in optimal scheme with our proposed method, the SA, TCO and MTF are improved 25.6%, 91.88% and 2.4%, respectively. The other method is neural networks (NN) combined with genetic algorithm (GA) to find the optimal optical design parameters. The proposed method compared with the CODE V built-in optimal scheme, the SA, TCO and MTF were improved 50.03%, 45.78% and 24.7%, respectively. The simulation results show that the better performances are presented in our two proposed multiple quality optimization methods.

並列關鍵字

Multi-response ； Taguchi method ； Principal Component Analysis ； Neural Network ； Genetic Algorithms

參考文獻

[1] Cho P., Cheung S.W., Edwards M.H. and Fung J., 2003, “An Assessment of Consecutively Presenting Orthokeratology Patients in a Hong Kong Based Private Practice”, Clin Exp Optom, 86, pp. 331–338.

[2] Cho P., Cheung S.W. and Edwards M.H., 2002, “Practice of Orthokeratology by a Group of Contact Lens Practitioners in Hong Kong Part 1. General Overview”, Clin Exp Optom, 85, pp. 365–371.

[4] Nowakowski, M., 2011, “Study of Ocular Aberrations Within a 10 Deg Central Visual Field”, Applied Optics Group School of Physics, National University of Ireland, Galway, pp. 9-53, March.

[5] Lin, C., Wu, C., Yang, P., and Kuo, T., 2009, “Application of Taguchi Method in Light-Emitting Diode Backlight Design for Wide Color Gamut Displays”, Journal of Display Technology, 5(8), pp. 323-330, August.

[7] Atchison, D. A., 1995, “Aberrations Associated with Rigid Contact Lenses”, JOSA A, 12(10), pp. 2267-2273.