臺灣博碩士論文加值系統

面對網路經濟時代製造環境的變化，傳統的製造業需要建立一種順應市場需求且具有快速回應機制的網路化製造模式，這將是當前乃至若干年後製造業所面臨的最緊迫的任務之一，也是製造業擺脫困境，贏得市場，掌握競爭主動權的關鍵。
本研究利用網路技術以取得現場設備之生產資訊，建立以類神經網路與Data Mining之Rough Set Theory於設備製程監控診斷系統。現場資料之擷取透過Ethernet網路，將製造現場機台狀況及生產數據送至監控端，以進行資料之收集、判斷及分析，管理者可隨時經由網路掌握現場即時生產狀況及品質變異情形；而對於生產參數與生產品質間之關係則透過類神經網路來訓練學習以應用於生產監控資料之及時辨識；另外本研究將運用資料挖礦(Data-mining)之粗集理論Rough Set Theory挖掘出生產參數與生產品質資料間的關連性或規則，提供決策者設定生產條件之依據。

Under the impact of quickly environmental changes of network economy ages, the manufacturing industry has to build up a better manufacturing system based on computer network, in order to meet the requirement of markets and respond quickly to markets. To build up the manufacturing system under the computer network environment will become one of the most urgent tasks that manufacturing industry will face in the near future, and will be the key factor that the company becomes competitive in market.
This research makes use of the network technique to obtain the production information of the monitored equipment, uses the back propagation neural network to monitor and predict product quality situations, and uses the rough set theory to extract the relationship between manufacturing parameters and product quality situations. Several sensors were mounted on a monitored machine to acquire the manufacturing parameters. Product quality information was keyed-in by operators through a human-computer interface. The manufacturing parameters and product quality situations were sent to a monitor computer through Ethernet network for further analysis. Therefore, managers can have a full control and real time control of the shop floor situations and quality variations. This research finds that properly selected sensors with Ethernet network facilitate remote manufacturing process monitoring, the back propagation neural network has very good accuracy in predicting manufacturing output situations, and the rough set theory is useful to extract the rules and relationships between manufacturing parameters and output situations.

目錄
書名頁………………………………………………………………………….i
授權書………………………………………………………………………….ii
中文摘要……………………………………………………………………….v
英文摘要……………………………………………………………………….vi
誌謝…………………………………………………………………………….vii
目錄……………………………………………………………………………..viii
表目錄…………………………………………………………………………..x
圖目錄…………………………………………………………………………..xi
第一章、緒論………………………………………………………………….1
1﹒1前言………………………………………………………………………1
1﹒2研究背景與動機…………………………………………………………2
1﹒3研究目的…………………………………………………………………4
1﹒4研究方法…………………………………………………………………5
1﹒5研究進行步驟……………………………………………………………5
第二章、文獻探討……………………………………………………….……7
2﹒1製造業遠距服務系統架構………………………………………………7
2.1.1現場設備資訊收集、控制及故障診斷系統的建立………….….…7
2.1.2 監控資料庫系統的建立…………..…………………….….…….....8
2.1.3現場監控系統效益………………………..………………..…..……9
2﹒2自動化與網路應用技術…………………………………………………9
2.2.1網路監控工具…………………………………………………..….…12
2﹒3智慧型監控技術…………………………………………………………13
2﹒4應用類神經網路技術……………………………………………………15
2.4.1類神經網路架構與符號表示法……………….…………….…..…17
2.4.2 倒傳遞演算法則…………………………..…..……………….…19
2.4.3神經網路的收斂速度…………………………………………..…24
2﹒5應用資料挖礦技術……………………………………………………25
2.5.1資料挖掘的方法和技術……………………..………….……...…28
2.5.2資料挖掘步驟………………………………………………….…..29
2.5.3粗集理論簡介………………………………………………….…..31
第三章、系統架構與設計模組………………………………………..….35
3﹒1系統架構…………………………………………………………..…35
3﹒2訊號擷取模組………………………………………………….….…37
3﹒3資料分析儲存架構………………………………………….…….…41
3﹒4系統設計技術項目……………………………………………..……42
3﹒5資料擷取監視軟體…………………………………………….……43
3﹒6現場資料庫儲存格式…………………………………………….…47
3﹒7知識推論演算法……………………………………………………49
第四章、監控資料分析結果……………………………………………52
4.1資料篩選分析…………………………………….…………………52
4.2倒傳遞網路演算法……………………………………………….…55
4.3粗集理論演算法………………………………………………….…59
4.4資料關聯分析…………………………………………………….…64
第五章、結論與未來研究方向………………..…………….…………66
5.1結論……………..…………………………………………..…….…66
5.2未來研究方向………………………………………………..…...…67
參考文獻………………………………………………………….….…68

[1] Bell, D. A. and Guan, J. W., “Computational Methods for Rough Classification and Discovery”, Journal of the American Society for Information Science, Vol. 49, No. 5, pp. 403-414, 1998.
[2] Kandil, N.; Sood, V.K.; Khorasani, K.; Patel, R.V. “Fault identification in an AC-DC transmission system using neural networks”，Proceeding of the IEEE International Conference on Power Industry Computer Application, 1991 .Page(s): 285 —292.
[3] Kusiak, A., ” Computational Intelligence in Design and Manufacturing” , John Wiley & Sons, Inc., 2000.
[4] Lai, L.L. ,“Application of neural networks to fault classification and protection” Proceeding of the 4th International Conference on Advances in Power System Control, Operation and Management, APSCOM-97., Page(s): 72 —76.
[5] Leonard, J.A. and Kramer,M.A ,”Radial basis function networks for classifying process faults”, IEEE Control Systems Magazine, 1991,April,pp.31-38.
[6] Lee, J., 2000, Industry/University Cooperative Research Center on Intelligent Maintenance System (IMS), Unpublished Report of University of Wisconsin at Milwaukee and University of Michigan at Ann Arbor.
[7] Liberatore, A.; Manetti, S.; Piccirilli, M.C. ,“Network symbolic analysis for automated fault diagnosis “, Proceeding of the 1992 IEEE International Symposium on Circuits and Systems, ( ISCAS ''92), Page(s): 1169 -1172 .
[8] Lingras, P. J. and Yao, Y. Y., “Data Mining Using Extensions of the Rough Set Model”, Journal of the American Society for Information Science, Vol. 49, No. 5, pp. 415-422, 1998.
[9] Marko, K.A.; Feldkamp, L.A.; Puskorius, G.V. ,“Automotive diagnostics using trainable classifiers: statistical testing and paradigm selection”, The IEEE International Conference on 1990 IJCNN International Joint Neural Networks, Page(s): 33 -38 vol.1
[10] Naidu, S.R.; Zafiriou, E.; McAvoy, T.J. ,“Use of neural networks for sensor failure detection in a control system” , IEEE Control Systems Magazine Volume: 10 3 , April 1990 , Page(s): 49 —55.
[11] Polkowski, S. L. and Skowron, A.,”Rough Sets in Knowledge Discovery 1: Methodology and Applications”, Physical-Verlag, 1998.
[12] S.C. Hui , G. Jha “Data mining for customer service support.”，Information & Management 38 (2000) 1-13.
[13] Shen, L., Tay, F. E. H., Qu, L., and Shen, Y., “Fault Diagnosis Using Rough Sets Theory”, Computers in Industry, Vol. 43, No. 1, pp. 61-72.
[14] Wang, D.D.; Debing Yang; Jinwu Xu; Ke Xu ,“Computational intelligence based machine fault diagnosis”, Proceedings of The IEEE International Conference on industrial Technology, Page(s): 465 —469.
[15] 王姿婷，”約略集合論於數量形資料之知識擷取”，義守大學資訊工程研究所.碩士論文，1999。
[16] 李嘉慧，"柔性解算應用於管制圖：設計策略之研究"，元智工學院碩士論文，1998。
[17] 李家祥，”遠距診斷維修系統平台建置”，國立台北科技大學，機電整合研究所碩士論文，1999。
[18] 林文修，”「Data Mining探索”，叡揚資訊，經營決策論壇13 期。
[19] 林淇淵，”在POS資料倉儲中挖掘關聯規則”，國立中國大學資訊管理碩士論文，2000。
[20] 宮亭汝，"應用類神經網路偵測製程平均值及變異數變化性之研究"，元智工學院博士論文，1997。
[21] 梁丁文，”以粗略集合方法從臨床資料庫中學習相依規則”，國立清華大學資訊科學研究所碩士論文，1992。
[22] 凌俊青，”在包裹資料庫中挖掘數量關聯規則”，國立中國大學資訊管理碩士論文，1999。
[23] 陳智宗，”以資料挖礦法挖掘多屬性序列式資料規則之研究”，國立中國大學資訊管理碩士論文，2000。
[24] 劉柏青，”專家系統在設備遠程維修與監控之應用”，元智大學工業工程研究所碩士論文，1998。
[25] ITIS2000 ”科技產業現況與市場趨勢研討會”,電機產業、機械產業。