研究生: |
魏俊安 Chun-An Wei |
---|---|
論文名稱: |
卷積神經網路應用於含菲涅爾微結構隱形眼鏡之射出成形殼模形狀誤差及微結構複製率分析研究 Convolution Neural Network for Form Error Analysis and Groove Filling Ratio of Structured Contact Lens with Micro Fresnels by Injection Molding |
指導教授: |
陳炤彰
Chao-Chang Chen |
口試委員: |
陳炤彰
Chao-Chang Chen 張榮語 Rong-Yeu Chang 莊水旺 Shuei-Wan Juang 黃忠偉 Jong-Woei Whang 莊程媐 Cheng-Hsi Chuang |
學位類別: |
碩士 Master |
系所名稱: |
工程學院 - 機械工程系 Department of Mechanical Engineering |
論文出版年: | 2021 |
畢業學年度: | 109 |
語文別: | 中文 |
論文頁數: | 173 |
中文關鍵詞: | 含菲涅爾微結構隱形眼鏡殼模 、射出成形 、卷積神經網路 、基因演算法 、人類眼球模擬 |
外文關鍵詞: | Soft Contact Lens Shell Mold, Fresnel Microstructure, Injection Molding (IM), Convolutional Neural Network (CNN), Genetic Algorithm (GA) |
相關次數: | 點閱:270 下載:0 |
分享至: |
查詢本校圖書館目錄 查詢臺灣博碩士論文知識加值系統 勘誤回報 |
本研究主要分析隱形眼鏡製程中,射出成形製造含菲涅爾微結構隱形眼鏡殼模及製備隱形眼鏡乾片及濕片造成之誤差對於隱形眼鏡的光學性能影響。分析形狀誤差(Form Error) Rt、平均Z軸位移量及微結構複製率(Groove Filling Ratio, GFR),並透過卷積神經網路(Convolutional Neural Networks, CNN)及基因演算法(Genetic Algorithm, GA)進行預測及最佳化。 透過Moldex3D R16.0及實驗設計(Design of Experiment, DoE)對微結構隱形眼鏡殼模進行射出成形參數分析,並使用彩色共軛焦對光學區前弧進行量測。透過CNN預測隱形眼鏡殼模是否達到GFR 70 %以上,GA進行射出成形參數最佳化並輸出目標值。製備隱形眼鏡乾片及濕片進行量測,分析對於屈光度之光學性能的影響,探討模仁入子至射出成形殼模誤差與殼模二次模造至乾片的誤差。結果顯示最具影響形狀誤差為射出速度,最具影響GFR因子為模具溫度。最佳化殼模形狀誤差Rt為38.42 μm、平均Z軸位移量為15.10 μm及微結構複製率GFR為70.03 %,優於實驗設計及先前研究成果有效降低形狀誤差。乾片形狀誤差Rt為152.52 μm,因HEMA向厚度較厚區域收縮造成形狀誤差提升。濕片的屈光度約為+1.25 D至+2.75 D,最具影響濕片屈光度的誤差為Rt影響程度達43.1%,其次為GFR影響程度為37.83 %,平均Z軸位移量影響程度為18.08 %。此外透過ABAQUS進行不同淚液層厚度下眨眼模擬並量化舒適度,眼球受力結果低於平均眼壓不會造成舒適度影響,未來可應用於含微結構隱形眼鏡舒適度設計。
This research focuses on the errors which are caused by the injection molding manufacturing of the contact lens shell mold with Fresnel microstructure and caused by the manufacturing of dry and wet contact lenses of the contact lens manufacturing process. Analyze the form error (Rt), the average Z-axis displacement, and the groove filling ratio(GFR). Convolutional Neural Networks (CNN) and Genetic Algorithm (GA) are used for prediction and optimization. Moldex3D R16.0 and Design of Experiment (DoE) are used to analyze the injection molding parameters of the microstructure contact lens shell mold, and the Confocal is used to measure the anterior curve of the optical zone. CNN is used to predict whether the GFR of contact lens shell mold reaches 70%. GA can optimize the injection molding parameters and output the target value. To manufacture and measure the dry and wet contact lenses, then analyze the impact on the optical power. Results show that the most influential form error is the injection velocity, and the most influential GFR factor is the mold temperature. The optimized shell mold form error Rt is 38.42 μm, the average Z-axis displacement is 15.10 μm, and the GFR is 70.03%. The optimized result is better than the group of DoE and previous research results. The Rt of the dry lens is 152.52 μm, which is caused by the shrinkage of HEMA to the thickness area. The diopter of the wet lens is about +1.25 D to +2.75 D. The most influential error of wet lens diopter is Rt that the influence ratio is 43.1 %, followed by GFR influence ratio of 37.83 %, and average Z-axis displacement influence ratio of 18.08 %. In addition, ABAQUS is used to simulate blinking under different tear layer thicknesses and quantify the value of comfort by pressure. Result shows that the eyeball pressure is lower than the average intraocular pressure, which may not affect comfort. Future work can be focused on the comfortability of contact lens with microstructure.
[1] 王滿堂, 隱形眼鏡學(下冊). 藝軒圖書出版社, 2005.
[2] 陳柏佑, "運用機器學習於單晶矽晶圓鑽石線鋸加工之進給最佳化研究," 碩士, 機械工程系, 國立臺灣科技大學, 台北市, 2020.
[3] J. S. Wolffsohn, T. Drew, S. Dhallu, A. Sheppard, and G. J. Hofmann, "Impact of soft contact lens edge design and midperipheral lens shape on the epithelium and its indentation with lens mobility," Invest Ophthalmol Vis Sci, vol. 54, no. 9, pp. 6190-7, Sep 11 2013.
[4] D. Madrid-Costa, J. Ruiz-Alcocer, S. García-Lázaro, T. Ferrer-Blasco, and R. Montés-Micó, "Optical power distribution of refractive and aspheric multifocal contact lenses: effect of pupil size," vol. 38, no. 5, pp. 317-321, 2015.
[5] A. Davis and F. Kühnlenz, "Optical design using Fresnel lenses: Basic principles and some practical examples," vol. 2, no. 4, pp. 52-55, 2007.
[6] T. Fujii, A. Goulet, K. Hattori, K. Konno, and A. Tanaka, "Fresnel lens sidewall design for imaging optics," vol. 10, 2015.
[7] A. R. A. Manaf, T. Sugiyama, and J. Yan, "Design and fabrication of Si-HDPE hybrid Fresnel lenses for infrared imaging systems," vol. 25, no. 2, pp. 1202-1220, 2017.
[8] K. V. Balasaheb, "菲涅爾隱形眼鏡設計與殼模射出成形之 體積收縮分析研究," 碩士, 機械工程系, 國立臺灣科技大學, 台北市, 2020.
[9] C-C. A Chen and S.-W. Chang, "Shrinkage Analysis on Convex Shellby Injection Molding," vol. 23, no. 1, pp. 65-71, 2008.
[10] 邱鈺婷, "非球面軟式隱形眼鏡之殼模射出成形研究," 碩士, 機械工程系, 國立臺灣科技大學, 台北市, 2016.
[11] R. J. Bensingh, S. R. Boopathy, and C. Jebaraj, "Minimization of variation in volumetric shrinkage and deflection on injection molding of Bi-aspheric lens using numerical simulation," vol. 30, no. 11, pp. 5143-5152, 2016.
[12] A.-K. Holthusen, O. Riemer, J. Schmütz, and A. Meier, "Mold machining and injection molding of diffractive microstructures," vol. 26, pp. 290-294, 2017.
[13] C.-M. Lin and H.-K. Hsieh, "Processing optimization of Fresnel lenses manufacturing in the injection molding considering birefringence effect," vol. 23, no. 12, pp. 5689-5695, 2017.
[14] P. Dai, H. Han, Y. Zhao, and M. Fan, "Finite element analysis of the mechanical characteristics of glaucoma," vol. 16, no. 04, p. 1650060, 2016.
[15] A. Abass, L. White, A. Elsheikh, and J. Clamp, "Finite Element Modelling of Soft Contact Lenses on Eye," 2017.
[16] A. Eliasy, K.-J. Chen, R. Vinciguerra, B. T. Lopes, and A. Abass, "Determination of corneal biomechanical behavior in-vivo for healthy eyes using CorVis ST tonometry: stress-strain index," vol. 7, p. 105, 2019.
[17] R. Mukhopadhyay, P. S. Panigrahy, G. Misra, and P. Chattopadhyay, "Quasi 1D CNN-based fault diagnosis of induction motor drives," in 2018 5th International Conference on Electric Power and Energy Conversion Systems (EPECS), 2018, pp. 1-5: IEEE.
[18] 廖紘毅, "銅膜晶圓化學機械拋光之終點偵測研究," 碩士, 機械工程系, 國立臺灣科技大學, 台北市, 2020.
[19] D. Mathivanan and N. Parthasarathy, "Sink-mark minimization in injection molding through response surface regression modeling and genetic algorithm," vol. 45, no. 9-10, p. 867, 2009.
[20] L. Eren, T. Ince, and S. Kiranyaz, "A generic intelligent bearing fault diagnosis system using compact adaptive 1D CNN classifier," vol. 91, no. 2, pp. 179-189, 2019.
[21] E. Kobler, C. Kastner, and G. Steinbichler, "Application of a convolutional neural network in polymer injection foam molding," in AIP Conference Proceedings, 2020, vol. 2289, no. 1, p. 020033: AIP Publishing LLC.
[22] H. Prautzsch, W. Boehm, and M. Paluszny, Bézier and B-spline techniques. Springer, 2002.
[23] V. T. Lien, "NURBS多焦隱形眼鏡設計及殼模射出成形之收縮分析研究," 博士, 機械工程系, 國立臺灣科技大學, 台北市, 2018.
[24] V. T. Lien, "整合模擬退火法之NURBS近似於非對稱曲面之最佳化應用," 碩士, 機械工程系, 國立臺灣科技大學, 台北市, 2010.
[25] I. Tranoudis and N. Efron, "Tensile properties of soft contact lens materials," vol. 27, no. 4, pp. 177-191, 2004.
[26] M. D. Willcox, P. Argüeso, G. A. Georgiev, J. M. Holopainen, and G. W. Laurie, "TFOS DEWS II tear film report," vol. 15, no. 3, pp. 366-403, 2017.
[27] D. C. Turner et al., "Transient intraocular pressure fluctuations: source, magnitude, frequency, and associated mechanical energy," vol. 60, no. 7, pp. 2572-2582, 2019.
[28] 佩喬斯,凱大衛及韓力希, "用於隱形眼鏡製造之模具," 中華民國專利 I449614, 2014.
[29] 陳正修, "隱形眼鏡之成形方法及模具," 中華民國專利 I330129, 2010.
[30] 黃澄儀, "混合式隱形眼鏡及其製作模具組與製造方法," 中華民國專利 I495921, 2015.
[31] 武氏蓮,陳炤彰及邱鈺婷, "漸進變焦隱形眼鏡及其製造方法," 中華民國專利 I584022, 2017.
[32] Thi-Lien Vu, Chao-Chang Chen, Yu-Ting Qiu, " Progressive multifocal contact lens and producing method thereof. " United States, US20180024380A1, 2019.