簡易檢索 / 詳目顯示

研究生: 沈昌毅
Shang, Chang-Yi
論文名稱: 利用結構矩陣和有向圖模型的改良模糊診斷方法
Improved Fuzzy Diagnosis Method with Structure Matrix and SDG Model
指導教授: 張玨庭
Chang, Chuei-Tin
學位類別: 碩士
Master
系所名稱: 工學院 - 化學工程學系
Department of Chemical Engineering
論文出版年: 2010
畢業學年度: 98
語文別: 中文
論文頁數: 129
中文關鍵詞: 模糊推論全域局部失誤傳遞路徑症候發生順序
外文關鍵詞: IF-THEN, Fuzzy, FPP, SOO
相關次數: 點閱:48下載:3
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
  • 根據有向圖模式在大型化工程序中進行模糊診斷的現有方法,往往會需要眾多線上測量訊號,不但計算量龐大且鑑別度不高。本研究利用結構矩陣(Structure matrix)先將系統分割成不同區塊後再分別進行模糊推論,不但減少了IF-THEN法則數目,而且可以提升解析能力。我們利用MATLAB模擬測試,對一系列案例深入分析,從模擬結果可以觀察到,若使用上述局部推論方式執行失誤診斷,針對較上游發生的失誤,可以達到較高的診斷效率及鑑別度。

    The signed directed graphic (SDG) modal presently applying to fuzzy diagnosis performs in the chemical engineering process with large digital calculation and low differentiating degree. In this study, structure matrix is used to dividing the entire system into small blocks. the problem is solved by the way of fuzzy inference. Local fuzzy inference analysis showed the decreasing tendency of IF-THEN rules and enhancement of analytical ability. For the practical purpose, “MATLAB” is used to simulate the cases. By the way of simulation, the error observation can be accomplished at the upstream procedure by local fuzzy inference. The method for error detection can get higher level of diagnosis efficiency and differentiating degree.

    目錄 第一章 緒論 ................................................1 1.1 研究動機 ........................................1 1.2 文獻回顧 ........................................1 1.2.1 與失誤診斷相關研究 ........................1 1.2.1 以有向圖為基礎的模糊診斷技術 ..............2 1.2.3 現有模糊診斷方法在大型系統上應用的困難 ....4 1.3 研究目的 ........................................4 1.4 組織和章節 ......................................4 第二章 模糊診斷分析方法 ....................................5 2.1失誤診斷步驟 .....................................5 2.2有向圖的建立......................................5 2.3失誤傳遞路徑與症候發生順序 .......................8 2.4 模糊推論法則 ...................................12 2.5 線上診斷架構....................................14 2.6 模糊推論機制 ...................................15 第三章 大型系統的診斷策略 .................................29 3.1 系統分割策略....................................29 3.2 診斷法則的訂定..................................36 3.2.1 全域模糊推論法則..........................48 3.2.2 局部模糊推論法則..........................53 3.3 診斷效能之表較..................................58 第四章 應用案例............................................60 4.1 案例一..........................................60 4.2 案例二..........................................70 4.3 案例三..........................................83 4.4 失誤幅度的影響 .................................93 第五章 結論與展望.........................................112 5.1 結論...........................................112 5.2 展望 ..........................................113 參考文獻 .................................................114 附錄A:單水槽液位控制系統模型.............................116 附錄B:雙水槽液位控制系統模型.............................118 附錄C:五個儲槽系統模型...................................123 附錄D:案例一失誤液位模擬數據.............................124 附錄E:案例二失誤液位模擬數據 ............................126 圖目錄 圖2.1 單一儲槽液位控制系統..................................6 圖2.2 液位系統有向圖模式....................................7 圖2.3 負回饋迴路............................................9 圖2.4 第九組失誤源之失誤傳播路徑...........................11 圖2.5 第九組失誤源之症候發生順序...........................11 圖2.6 第二組失誤源(a)FPP(b)SOO.............................11 圖2.7 模糊推論系統架構.....................................15 圖2.8 偏差值隸屬函數.......................................16 圖2.9 可能性指標隸屬函數...................................16 圖2.10 第二種情境之模擬數據................................18 圖2.11 第二種情境之可能性指標..............................18 圖2.12 第五種情境的模擬數據................................19 圖2.13 第五種情境的可能性指標..............................19 圖2.14 第九種情境的模擬數據................................20 圖2.15 第九種情境的可能性指標..............................20 圖2.16 兩個儲槽的液位控制系統..............................21 圖2.17 兩個儲槽的液位控制系統的有向圖模式..................22 圖2.18 四種失誤干擾的SOO ..................................23 圖2.19 失誤源(a)之模擬數據.................................24 圖2.20 失誤源(a)可能性指標 ................................25 圖2.21 失誤源(b)之模擬數據 ................................25 圖2.22 失誤源(b)可能性指標.................................26 圖2.23 失誤源(c)之模擬數據 ................................26 圖2.24 失誤源(c)可能性指標 ................................27 圖2.25 失誤源(d)之模擬數據 ................................27 圖2.26 失誤源(d)可能性指標 ................................28 圖3.1 兩個攪拌儲槽連接而成的系統 .........................29 圖3.2 圖3.1的有向圖模型 .................................30 圖3.3 虛擬有向圖模型 .....................................31 圖3.4 表3.2之順序圖 .....................................32 圖3.5 表3.3之順序圖......................................34 圖3.6 儲槽一額外擾流q3的干擾.............................35 圖3.7 儲槽二額外擾流q4的干擾.............................35 圖3.8 五個儲槽連接而成系統................................36 圖3.9 五個儲槽系統的有向圖模型............................38 圖3.10 失誤f1的SOO.......................................39 圖3.11 失誤f2的SOO.......................................39 圖3.12 失誤f3的SOO.......................................39 圖3.13 失誤f4的SOO.......................................39 圖3.14 失誤f5的SOO.......................................39圖3.15 失誤f6的SOO.......................................39 圖3.16 失誤f7的SOO.......................................39 圖3.18 失誤f8的SOO.......................................40 圖3.19 失誤f9的SOO.......................................40 圖3.20 f1失誤模擬數據.....................................43 圖3.22 針對失誤f1模擬數據全域的模糊診斷結果(五個偵測點)...43 圖3.22 f3失誤液位模擬數據..................................44 圖3.23 針對失誤f3模擬數據全域的模糊診斷結果(五個偵測點)...44 圖3.24 f9失誤模擬數據......................................45 圖3.25 針對失誤f9模擬數據全域的模糊診斷結果(五個偵測點)...45 圖3.26 表3.12之影響順序圖.................................47 圖3.27 失誤f1的SOO........................................48 圖3.28 失誤f2的SOO........................................48 圖3.29 失誤f3的SOO........................................48 圖3.30 失誤f4的SOO........................................48 圖3.31 失誤f5的SOO........................................48 圖3.32 失誤f6的SOO........................................49 圖3.33 失誤f7的SOO........................................49 圖3.34 失誤f8的SOO........................................49 圖3.35 失誤f9的SOO........................................49 圖3.36 失誤f10的SOO.......................................49 圖3.37 針對失誤f1模擬數據全域的模糊診斷結果(三個偵測點)...52 圖3.38 針對失誤f3模擬數據全域的模糊診斷結果(三個偵測點)...52 圖3.39 針對失誤f9模擬數據全域的模糊診斷結果(三個偵測點)...53 圖3.40 對失誤f1局部診斷結果(五個偵測點)...................55 圖3.41 對失誤f3局部診斷結果(五個偵測點)...................55 圖3.42 對失誤f9局部診斷結果(五個偵測點)....................56 圖3.43 對失誤f1局部診斷結果(三個偵測點)....................56 圖3.44 對失誤f3局部診斷結果(三個偵測點)....................57 圖3.45 對失誤f9局部診斷結果(三個偵測點)....................57 圖4.1 案例一中三個儲槽系統................................61 圖4.2 案例一的有向圖模式..................................62 圖4.3 圖4.2之影響順序圖..................................63 圖4.4 案例一f1全域推論所得的可能性........................65 圖4.5 案例一全f4域推論所得的可能性........................66 圖4.6 案例一f6全域推論所得的可能性........................66 圖4.7 案例一f1局部推論所得的可能性........................67 圖4.8 案例一f4局部推論所得的可能性........................67 圖4.9 案例一f6局部推論所得的可能性........................68 圖4.10 案例二中的五個儲槽系統..............................71 圖4.11 案例二的系統有向圖..................................72 圖4.12 圖4.11的影響順序圖.................................73 圖4.13 案例二f1全域模糊診斷結果............................75 圖4.14 案例二f2全域模糊診斷結果............................76 圖4.15 案例二f3全域模糊診斷結果............................76 圖4.16 案例二f6全域模糊診斷結果............................77 圖4.17 案例二f7全域模糊診斷結果............................77 圖4.18 案例二f1局部模糊診斷結果............................78 圖4.19 案例二f2局部模糊診斷結果............................78 圖4.20 案例二f3局部模糊診斷結果............................79 圖4.21 案例二f6局部模糊診斷結果............................79 圖4.22 案例二f7局部模糊診斷結果............................80 圖4.23 案例三中的儲槽系統..................................83 圖4.24 八個儲槽的有向圖模型................................84 圖4.25 圖4.24簡化順序圖....................................86 圖4.26 f2全域模糊診斷結果...................................88 圖4.27 f4全域模糊診斷結果...................................88 圖4.28 f6全域模糊診斷結果...................................89 圖4.29 f7全域模糊診斷結果...................................89 圖4.30 f2局部模糊診斷結果...................................90 圖4.31 f4局部模糊診斷結果...................................90 圖4.32 f6局部模糊診斷結果...................................91 圖4.33 f7局部模糊診斷結果...................................91 圖4.34 全域Dist_2=200cm3/s模糊診斷結果.....................94 圖4.35 局部Dist_2=200cm3/s模糊診斷結果.....................94 圖4.36 全域Dist_2=300cm3/s模糊診斷結果.....................95 圖4.37 局部Dist_2=300cm3/s模糊診斷結果.....................95 圖4.38 全域Dist_2=400cm3/s模糊診斷結果.....................96 圖4.39 局部Dist_2=400cm3/s模糊診斷結果.....................96 圖4.40 全域Dist_2=500cm3/s模糊診斷結果.....................97 圖4.41 局部Dist_2=500cm3/s模糊診斷結果.....................97 圖4.42 全域Dist_2=600cm3/s模糊診斷結果.....................98 圖4.43 局部Dist_2=600cm3/s模糊診斷結果.....................98 圖4.44 全域Dist_4=200cm3/s模糊診斷結果.....................99 圖4.45 局部Dist_4=200cm3/s模糊診斷結果.....................99 圖4.46 全域Dist_4=300cm3/s模糊診斷結果....................100 圖4.47 局部Dist_4=300cm3/s模糊診斷結果....................100 圖4.48 全域Dist_4=400cm3/s模糊診斷結果....................101 圖4.49 局部Dist_4=400cm3/s模糊診斷結果....................101 圖4.50 全域Dist_4=500cm3/s模糊診斷結果....................102 圖4.51 局部Dist_4=500cm3/s模糊診斷結果....................102 圖4.52 全域Dist_6=200cm3/s模糊診斷結果....................103 圖4.53 局部Dist_6=200cm3/s模糊診斷結果....................103 圖4.54 全域Dist_6=300cm3/s模糊診斷結果....................104 圖4.55 局部Dist_6=300cm3/s模糊診斷結果....................104 圖4.56 全域Dist_6=400cm3/s模糊診斷結果....................105 圖4.57 局部Dist_6=400cm3/s模糊診斷結果....................105 圖4.58 全域Dist_6=500cm3/s模糊診斷結果....................106 圖4.59 局部Dist_6=500cm3/s模糊診斷結果....................106 圖4.60 全域Dist_7=200cm3/s模糊診斷結果....................107 圖4.61 局部Dist_7=200cm3/s模糊診斷結果....................107 圖4.62 全域Dist_7=300cm3/s模糊診斷結果....................108 圖4.63 局部Dist_7=300cm3/s模糊診斷結果....................108 圖4.64 全域Dist_7=400cm3/s模糊診斷結果....................109 圖4.65 局部Dist_7=400cm3/s模糊診斷結果....................109 圖4.66 全域Dist_7=500cm3/s模糊診斷結果....................110 圖4.67 局部Dist_7=500cm3/s模糊診斷結果....................110 圖D.1 失誤f1液位模擬.....................................124 圖D.2 失誤f4液位模擬.....................................124 圖D.3 失誤f6液位模擬.....................................125 圖E.1 失誤f1液位模擬.....................................126 圖E.2 失誤f2液位模擬.....................................126 圖E.3 失誤f3液位模擬.....................................127 圖E.4 失誤f4液位模擬.....................................127 圖E.5 失誤f5液位模擬.....................................128 表目錄 表2.1 液位控制系統之可能失誤源..............................8 表2.2 標準負回饋回路控制的穩態值...........................10 表2.3 失誤源f9的候選型式...................................13 表2.4 第九組失誤源之IF-THEN推論法則.......................14 表2.5 兩個儲槽的失誤發生情形...............................22 表2.6 失誤源(a)的候選型式..................................23 表2.7 失誤源(a)之IF-THEN 推論法則..........................24 表3.1 圖3.3的結構矩陣.....................................31 表3.2 表3.1的區塊三角形型式之結構矩陣.....................32 表3.3 圖3.2之結構矩陣.....................................34 表3.4 五個儲槽的失誤.......................................37 表3.5 失誤f1的候選型式.....................................40 表3.6 f1失誤源之IF-THEN推論法則...........................41 表3.7 失誤f3的候選型式.....................................41 表3.8 f3失誤源之IF-THEN推論法則...........................42 表3.9 失誤f9的候選型式....................................42 表3.10 f9失誤源之IF-THEN推論法則...........................42 表3.11 圖3.9的結構矩陣....................................46 表3.12 由表3.11轉換的區塊三角形的型式.....................47 表3.13 失誤f1的候選型式....................................50 表3.14 f1失誤源之IF-THEN推論法則..........................50 表3.15 失誤f3的候選型式....................................50 表3.16 f3失誤源之IF-THEN推論法則...........................51 表3.17 失誤f9的候選型式....................................51 表3.18 f9失誤源之IF-THEN推論法則...........................51 表3.19 五個偵測點matlab的模糊推論運算(sec)................58 表3.20 對五個偵測點失誤可能性到達0.5的 模糊推論診斷時間(sec)...............................58 表3.21 對三個偵測點的matlab的模糊推論運算時間(sec)........58 表3.22 對五個偵測點失誤可能性到達0.5的 模糊推論診斷時間(sec)...............................58 表3.23 對三個偵測點的診斷鑑別度............................59 表4.1 案例一中的可能失誤源................................60 表4.2 案例一的結構矩陣....................................63 表4.3 案例一全域及局部SOO.................................64 表4.4 案例一的matlab計算時間(sec)........................68 表4.5 案例一到達可能性0.5的時間(sec).....................68 表4.6 案例一的診斷鑑別度..................................69 表4.7 案例二中的可能失誤源................................70 表4.8 案例二的結構矩陣....................................73 表4.9 案例二全域及局部SOO.................................74 表4.10 案例二的matlab計算時間(sec)........................80 表4.11 案例二到達可能性0.5的時間(sec).....................81 表4.12 案例二的診斷鑑別度..................................81 表4.13 案例三中之可能失誤源................................85 表4.14 八個儲槽系統的結構矩陣..............................85 表4.15 案例三全域及局部SOO.................................86 表4.16 案例三的matlab計算時間(sec)........................92 表4.17 案例三到達可能性0.5的時間(sec).....................92 表4.18 各擾流的不同流量大小................................93 表A.1 系統參數值與狀態初始值.............................118 表B.1 系統參數值與狀態初始值.............................120 表C.1 系統參數值與狀態初始值.............................123

    Chang, C. T., Mah, K. N., and Tsai, C. S., “A Simple Design Strategy for Fault Monitoring Systems,” AIChE Journal. 39, 1146-1163, 1993.

    Chang, S. Y., Lin, C. R., and Chang, C. T., “A Fuzzy Diagnosis Approach Using Dynamic Fault Tree,” Chemical Engineering Science 57, 2971-2985, 2002.

    Chang, S. Y., and Chang, C. T., “A-Fuzzy-Logic Based Fault Diagnosis Straetgy for Control Loops,” Chemical Engineering Science 58, 3305-3411, 2003.

    Chen, J. Y., and Chang, C. T. “Fuzzy Diagnosis Method for Control Systems with Coupled Feed Forward and Feedback Loops,” Chemical Engineering Science 61, 3105-3128, 2006.

    Chen, Y. J., and Chang, C. T., “Systematic Enumeration of Fuzzy Diagnosis Rules for Identifying Multiple Faults in Chemical Process,” Industrial and Engineering Chemistry Research . 46, 3635-3655, 2007

    Choi, H. H., “Output Feedback Stabilization of Uncertain Fuzzy Systems Using Variable Structure System Approach,” Fuzzy Sets And Systems 160, 2812-2823, 2009.

    Dash, S., Rengaswamy, R., and Venkatasubramanian, V., “Fuzzy-Logic Based Trend,” Computer and Chemical Engineering 27, 347-362, 2003.

    Elsayed, T., “Fuzzy Inference System for The Risk Assessment of Liquefied Natural Gas Carriers During Loading Offloading at Terminals,” Applied Ocean Research 31, 179-185, 2009.

    Ferdous, R., Khan, F., Veitch, B., and Amyotte, P. R., “Methodology for Computer Aided Fuzzy Fault Tree Analysis,” Process Safety and Environmental Protection 87, 217-226, 2009.

    Horiuchi, J. I., “Fuzzy Modeling and Control of Biological Process,” Journal of Bioscience and Bioengineering 94, 574-578, 2002.

    Kilic, E., “Diagnosability of Fuzzy Discrete Event Systems,” Information Sciences 178, 858-870, 2008.

    Mathworks, 2002a. Simulink-Dynamic System Simulation for Matlab-Using Simulink, The Mathworks Inc.

    Mathworks, 2006a. Fuzzy Logic Toolbox-User Guid 2, The Mathworks Inc.

    Mathworks, 2006b. Simulink 5 Using Simulink , The Mathworks Inc.

    Maura, M. R., Rengaswamy, R., and Venkatasubramanian, V., “A Signed Directed Graph-Based Systematic Framework for Steady-State Malfunction Diagnosis inside Control Loops,” Chemical Engineering Science 61, 1790-1810, 2006.

    Montalva, M., Aracena. J., and Gajardo, A., “On the Complexity of Feedback Set Problems in Signed Digraphs,” Electronic Notes in Discrete Mathematics 30, 249-254, 2008.

    Rico, A., “Sugeno Integral in A Finite Boolean Algebra,” Fuzzy Sets and Systems 159, 1709-1718, 2008.

    Saridaakis, K. M., Chasalevris, A. C., Papadopoulos, C. A., and Dentsoras, A. J., “Applying Neural Networks, Genetic Algorithms and Fuzzy Logic for The Identification of Cracks in Shafts by Using Coupled Response Measurements,” Computers And Structure 86, 1318-1338, 2007.

    Sugeno, M.. Industrial Applications of Fuzzy Control, Elsevier Science Pub. Co, 1985.

    Tarifa, E. E., and Scenna, N. J., “Fault Diagnosis for MSF Dynamic States Using a SDG and Fuzzy Logic,” Desalination 166, 93-101. 2004.

    Tarifa, E. E., and Scenna, N. J., “Fault Diagnosis for MSF Dynamic States Using a SDG and Fuzzy Logic,” Desalination 152, 207-214. 2002.

    Ulerich, N. H., Powers, G. J., “On- Line Hazard Aversion and Fault Diagnosis in Chemical Process: The Digraph + Fault-Tree Method,” IEEE Sensors Journal. 37, 171-177, 1988.

    Wang, Z., and Klir, G. J., Fuzzy Measure Theory, Springer, 1993.

    Wang, L. X., A Course in Fuzzy Systems and Control, Prentice Hal, 1996.

    Yang, H., Ye H., and Wang, G., “Dynamic Reconstruction-Based Fuzzy Neural Network Method for Fault Detection in Chaotic System. ” Tsinghua Science And Technology 13, 65-70, 2008.

    Yuang, X. H., Li, H. X., Lee, E. S., “On The Definition of The Intuitionistic Fuzzy Subgroups,” Computers and Mathematics with Applications 59, 3117-3129, 2010.

    Zhang, Z. Q., Wu, C. Q., Zhang, B. K., Xia, Tao., and Li, A.F., “SDG Multiple Fault Diagnosis by Real-Time Inverse Inference,” Reliability Engineering and System Safe

    下載圖示 校內:立即公開
    校外:立即公開
    QR CODE