簡易檢索 / 詳目顯示

回結果列表
研究生： 王恩榮
En-jung Wang
論文名稱： 應用於質子交換膜燃料電池催化劑層之智慧型噴製系統
An Intelligent System of Catalyst Layer Fabrication for Proton Exchange Membrane Fuel Cell
指導教授： 李敏凡
Min-Fan Ricky Lee
口試委員: 蔡明忠
Ming-Jong Tsai
周賢鎧
Shyan-kay Jou
學位類別： 碩士
Master
系所名稱： 工程學院 - 自動化及控制研究所
Graduate Institute of Automation and Control
論文出版年： 2014
畢業學年度： 102
語文別： 英文
論文頁數： 68
中文關鍵詞： 類神經網絡 、模糊邏輯控制系統 、自適應網絡模糊推論系統 、幅狀基底函數類神經網絡 、質子交換膜燃料電池 、催化劑層製造
外文關鍵詞： Neural network, fuzzy logic control, ANFIS, RBFN, PEM Fuel Cell, fabrication of catalyst Layer
相關次數： 點閱：355  下載：4 
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
  •  上一筆

    • 效能，耐用度及成本是影響質子交換膜燃料電池(Proton Exchange Membrane or Polymer Electrolyte Membrane Fuel Cell, PEMFC)商業化的重要因素。催化劑層的結構影響其主要元件膜電極組件(Membrane Electrode Assembly, MEA)的效能。厚度變化，偏差及不平整會導致不合格產品和過早失效的問題。
    本文建議的系統採用了多層次控制策略由多層組合而成包含了最上層的監控器至最下層的感測器。此建議的控制系統及催化劑層的製作都將以計算機來模擬。本文探討了不同噴製軌跡，不同的上層控制器的建模方法（幅狀基底函數類神經網絡，RBFN和自適應網絡模糊推論系統，ANFIS），並檢查所噴製的催化劑層。
    我們找到最佳的噴製軌跡，並證明配備ANFIS的系統可以比較精準地預測兩個關鍵的噴製控制動作及噴製一個催化劑層具有所需厚度（具誤差0.0736％MAPE）和均勻度（具誤差0.8393％MAPE）。這些微小的誤差說明了這個建議的系統將可幫助製造者獲得更精準的控制於催化劑層的製造且可將不合格產品和過早失效的問題降至放最低以達本文的目的。

    Performance, durability and cost are the main factors influencing Proton Exchange Membrane (or Polymer Electrolyte Membrane, PEM) Fuel Cell commercialization. The key component of a PEM Fuel Cell is membrane electrode assembly (MEA) and the performance of MEA relies heavily on the structure of the catalyst layer. The thickness variations and roughness will cause problems leading to substandard products and premature failures.
    This thesis adopts the concept of multi-level control strategy and proposes a system consisting of several levels from the supervisory controller on top to the sensors at the bottom. The proposed system and the fabrication of the catalyst layers are implemented through computer simulation. This thesis explores different spray patterns, different modeling methods (Radial Basis Function Network, RBFN and Adaptive Network Based Fuzzy Interference System, ANFIS) for the high-level controller, and examines the sprayed catalyst layers.
    We find the best spray pattern and prove that the system equipped with ANFIS can predict the two key spraying control actions better and create a catalyst layer with the desired thickness (with a 0.0736% MAPE) and evenness (with a 0.8393% MAPE). These small errors between the resulting catalyst layers and the desired values show that this proposed system will help the manufacturers gain precise control over the catalyst layer fabrication and minimize the substandard products and premature failures, which achieves the goal of this thesis.

    Abstract I 中文摘要 II Acknowledgements III Content IV List of Figures VI List of Tables VIII Chapter 1 – Introduction - 1 - 1.1 Background - 1 - 1.2 Literature Review - 2 - 1.3 Research Objective - 7 - 1.4 Thesis Organization - 8 - Chapter 2 - Method - 9 - 2.1 Control System Design - 9 - 2.2 Catalyst Layer Spray Process Simulation - 12 - 2.3 Thickness Pre-processor - 18 - 2.4 Evenness Pre-processor - 18 - 2.5 Special Consideration for Evenness Setpoint - 18 - 2.6 RBFN-based high-level controller - 19 - 2.6.1 Hybrid Learning Rule for RBFN - 21 - 2.7 ANFIS-based high-level controller - 22 - 2.7.1 Basic Learning Rule for ANFIS - 27 - 2.7.2 Hybrid Learning Rule for ANFIS - 28 - 2.8 Supervisory Controller - 30 - Chapter 3 - Results - 32 - 3.1 Spray Patterns and Spray Surfaces - 32 - 3.2 RBFN-Based High-Level Controller Training and results - 37 - 3.2.1 Thickness control RBFN training-First time - 38 - 3.2.2 Thickness control RBFN training results – First time - 38 - 3.2.3 Thickness control RBFN training – Second time - 40 - 3.2.4 Thickness control RBFN training results – Second time - 40 - 3.2.5 Evenness control RBFN training – First time - 41 - 3.2.6 Evenness control RBFN training results – First time - 42 - 3.2.7 Evenness control RBFN training – Second time - 43 - 3.2.8 Evenness control training results – Second time - 43 - 3.3 ANFIS-based High-level Controller Training and results - 44 - 3.3.1 Thickness control ANFIS training - First time - 45 - 3.3.2 Thickness control ANFIS training results – First time - 46 - 3.3.3 Thickness control ANFIS training – Second time - 49 - 3.3.4 Thickness control ANFIS training results – Second time - 49 - 3.3.5 Evenness control ANFIS training – First time - 52 - 3.3.6 Evenness control ANFIS training results – First time - 53 - 3.3.7 Evenness control ANFIS training- Second time - 56 - 3.3.8 Evenness control ANFIS training results – Second time - 57 - 3.4 Summary - 61 - Chapter 4 – Conclusion and Future Work - 64 - 4.1 Conclusion - 64 - 4.2 Future Work - 65 - References - 66 -

    [1] S. Kundu, M. W. Fowler, L. C. Simon, and S. Grot, "Morphological features (defects) in fuel cell membrane electrode assemblies," Journal of Power Sources, vol. 157, pp. 650-656, 2006.
    [2] F. Barbir, PEM Fuel Cells: Theory and Practice, Elsevier Academic Press, CA, USA, 2005.
    [3] V. Mehta and J. S. Cooper, "Review and analysis of PEM fuel cell design and manufacturing," Journal of Power Sources, vol. 114, pp. 32-53, 2003.
    [4] A. D. Taylor, E. Y. Kim, V. P. Humes, J. Kizuka, and L. T. Thompson, "Inkjet printing of carbon supported platinum 3-D catalyst layers for use in fuel cells," Journal of Power Sources, vol. 171, pp. 101-106, 2007.
    [5] W.H Lee, "Fabrication of MEA Using Ink-jet Printing Technology, " Fuel Cell Seminar & Exposition, Session #4E - Manufacturing Advancements, CA, USA, 2007.
    [6] Y.-H. Cho, S. J. Yoo, Y.-H. Cho, H.-S. Park, I.-S. Park, J. K. Lee, and Y.-E. Sung, "Enhanced performance and improved interfacial properties of polymer electrolyte membrane fuel cells fabricated using sputter-deposited Pt thin layers," Electrochimica Acta, vol. 53, pp. 6111-6116, 2008.
    [7] H. Su, T.-C. Jao, O. Barron, B. G. Pollet, and S. Pasupathi, "Low platinum loading for high temperature proton exchange membrane fuel cell developed by ultrasonic spray coating technique," Journal of Power Sources, vol. 267, pp. 155-159, 2014.
    [8] R. Sitanggang, A. B. Mohamad, W. R. W. Daud, A. A. H. Kadhum, and S. E. Iyuke, "Fabrication of gas diffusion layer based on x-y robotic spraying technique for proton exchange membrane fuel cell application," Energy Conversion and Management, vol. 50, pp. 1419-1425, 2009.
    [9] A. Lindermeir, G. Rosenthal, U. Kunz, and U. Hoffmann, "On the question of MEA preparation for DMFCs," Journal of Power Sources, vol. 129, pp. 180-187, 2004.
    [10] I.-S. Park, W. Li, and A. Manthiram, "Fabrication of catalyst-coated membrane-electrode assemblies by doctor blade method and their performance in fuel cells," Journal of Power Sources, vol. 195, pp. 7078-7082, 2010.
    [11] N. V. Aieta, P. K. Das, A. Perdue, G. Bender, A. M. Herring, A. Z. Weber, and M. J. Ulsh, "Applying infrared thermography as a quality-control tool for the rapid detection of polymer-electrolyte-membrane-fuel-cell catalyst-layer-thickness variations," Journal of Power Sources, vol. 211, pp. 4-11, 2012.
    [12] J. J. Shen, Z.Dong, and Y. Liu, "Apparatus for Manufacturing Fuel Cell Membrane Electrode Assembly, " US Patent 0098101 A1, May 12,2005.
    [13] E.-J. J. Wang, M.-F. R. Lee, and C.-H. K. Ko, "An intelligent system of catalyst layer deposition for proton exchange membrane fuel cell," Journal of the Chinese Institute of Engineers, vol. 33, pp. 89-98, 2010.
    [14] J.-S. R. Jang, "ANFIS: adaptive-network-based fuzzy inference system," IEEE Transactions on Systems, Man and Cybernetics, vol. 23, pp. 665-685, 1993.
    [15] F. O. Karray and C. de Silva, Soft computing and intelligent systems design, Addison Wesley, 2004.
    [16] EFD Inc., Ultra TT Automation Series User's Guide, Nordson Company, East Providence, RI, USA, 2007.
    [17] M. A. S. Arikan and T. Balkan, "Modeling of paint flow rate flux for elliptical paint sprays by using experimental paint thickness distributions," Industrial Robot, vol. 33, pp. 60-66, 2006.
    [18] Y. A. Cengel, J. M. Cimbala, "Fluid Mechanics: Fundamentals and Applications, 2nd Edition, " McGraw-Hill, 2010.
    [19] Spray System Co., Optimizing your spray system, TM410B, 2013.

    QR CODE