臺灣博碩士論文加值系統

　　現今化學工業程序中，利用電腦來量測及收集數據是相當的容易及頻繁，由於數據筆數龐大，往往使得我們無法立刻的分辨出程序操作是否為正常狀態。如果不能立刻偵測出不正常情形的發生，則此失誤現象可能導致財物的損失、工廠設備的停工甚至危害操作人員的安全。
　　在本研究中，利用Shewhart Chart、CUSUM及EWMA等三種單變數管制分析方法對單變數系統做分析；多變數管制分析PCA方法對多變數系統做分析。另外，利用數學模式及實驗數據兩種方式對每種分析方法加以驗證其優點及缺點。在蒸餾塔操作實驗中，偵測產品品質的方式將利用溫度的量測代替濃度的量測。由分析的結果看出，雖然單變數分析的方式較為平常及容易，但是對一多變數系統而言，由於變數間存在著共線性關係，以單變數管制分析的方式則會對系統發生誤判的情形，這在一般化工程序中常會存在這種現象。本研究最後以測試的例子驗證PCA方式的優點。

The amount of data that we collect and store in modern chemical industrial plants with well-equipped computerized measurement devices has become quite large, and the volume of acquired data often exceeds our capability to detect whether the process is normal. If the abnormal event cannot be identified in time, the malfunction would cause monetary loss, plant shutdown, or even significant impacts on personnel safety.
In this research, statistical process control, including single variable and multivariable methods, is discussed. Shewhart Chart, CUSUM and EWMA for Single variable analysis are used here; PCA (Principal Component Analysis) for multivariable is applied here. Extensive testing and comparison on mathematical models and real experimental data from the distillation column are made to demonstrate the pros and cons. Temperature measurement instead of product concentration is used to detect the product quality in the distillation column. In our analysis, although the single variable statistical analysis is simple and easy, it is not appropriate for the multivariable system when collinearities occur, because it may lead to the false detection. This is particularly true in the most of chemical industrial processes. The effectiveness and excellence of the PCA technique are demonstrated from the testing examples.

摘要 ………………………………………………………….Ⅰ
Abstract ………………………………………………………Ⅱ
目錄 ………………………………………………………….Ⅲ
圖目錄 ……………………………………………………….Ⅴ
表目錄 ……………………………………………………….Ⅷ
第一章　前言 ……………………………………………….1
1-1　變數管制分析 ……………………………….……...1
1-2　文獻回顧 ……………………………………….…...2
1-3　研究動機 ……………………………………….…...4
第二章　多變數主要成份因素分析 ……………………….5
2-1　單變數統計程序控制 ………………………….…...6
2-1-1　Shewhart Chart ……………………………………6
2-1-2　累積和管制方式 ………………………………….9
2-1-3　指數加權移動平均管制方式 …………………….13
2-2　多變數主要成份分析 ………………………….…...15
第三章　系統介紹 ………………………………………….29
3-1　實驗裝置 ……………………………………….…...29
3-2　實驗裝置名稱及用途 ………………………….…...37
第四章　蒸餾系統的監控 ………………………………….39
4-1　操作步驟及數據收集 ………………………….…...39
4-2　異常品質的監控 ……………………………….…...44
4-3　討論 …………………………………………….…...57
第五章　結論 ……………………………………………….58
附錄：蒸餾塔實驗操作 …………………………………….59
1.目的 ……………………………………………………..59
2.實驗操作流程 …………………………………………..59
3.實驗操作設定條件 ……………………………………..60
4.實驗步驟 ………………………………………………..61
5.注意事項 ………………………………………………..65
6.實驗測試比較 …………………………………………..65
符號說明 …………………………………………………….76
參考文獻 …………………………….………………….…..78

Bozenhardt, H. F., “Modern Control Tricks Solve Distillation Problems,” Hydrocarbon Process, Jun, 47-50(1988).
Ewan, W. D., “When and How to Use Cu-Sum Charts,” Technometrics, Vol. 5(1963)
Hotelling, H., “Multivariate Quality Control” in Techniques of Statistical Analysis edited by Eisenhart, Hastay, and Wallis., McGraw Hill, New York, NY. (1947)
Jackson, J. E., “Principal Components and Factor Analysis: Part I-Principal Components,” Journal of Quality Technology, 12, 201-213(1980)
Jackson, J. E., “A User’s Guide to Principal Components,” John Wiley & Sons, New York, NY. (1991)
Johnson, N. L., “A Simple Theoretical Approach to Cumulatives Sum Control Charts,” Journal of the American Statistical Association, Vol. 54(1961)
Johnson, N. L. and F. C. Leone, “Cumulative Sum Control Charts-Mathematical Principals Applied to Their Construction and Use,” Part I, Industrial Quality Control, Vol. 18(1962a)
Johnson, N. L. and F. C. Leone, “Cumulative Sum Control Charts-Mathematical Principals Applied to Their Construction and Use,” Part II, Industrial Quality Control, Vol. 18(1962b)
Johnson, N. L. and F. C. Leone, “Cumulative Sum Control Charts-Mathematical Principals Applied to Their Construction and Use,” Part III, Industrial Quality Control, Vol. 18(1962c)
Keller, H. R. and D. L. Massart, “Chemometrics and Intelligent Laboratory Systems,” B12, 209(1992)
Kister, H. Z., Distillation Operation; McGraw-Hill: New York, 545 (1990)
Kourti, Theodora and John F. MacGregor, ”Process analysis, monitoring and diagnosis, using multivariate projection methods,” Chemometrics and Intelligent Laboratory Systems, 28, 3-21(1995)
Kourti, Theodora and John F. MacGregor, ”Multivariate SPC Methods for Process and Product Monitoring,” Journal of Quality Technology, 28(4), Oct. 1996
Kresta, James V., John F. MacGregor and Thomas E. Marlin, ”Multivariate Statistical Monitoring of Process Operating Performance,” The Canadian Journal of Chemical Engineering,” Vol. 69, Feb. (1991)
Locus, J. M., “The Design and Use of Cumulative Sum Quality Control Schemes,” Journal of Quality Technology, Vol. 8, (1976).
Luyben, W. L., “Control of Distillation Columns with Sharp Temperatures Profiles,” AIChE J., 17, 713-718(1971)
Mejdell, Thor and Sigurd Skogestad, “Estimation of Distillation Compositions from Multiple Temperature Measurements Using Partial-Least-Squares Regression.” Ind. Eng. Chem. Res., 30, 2543-2555(1991)
Page, E. S., “Continuous Inspection Schemes,” Biometrics, Vol. 41(1954).
Rademaker, O., J. E. Rijnsdorp and A. Maarleveld, “Dynamics and Control of Continuous Distillation Units,” Elsevier: Amsterdam (1975)
Ray, W. H., “Chemometric Methods for Process Monitoring and High-Performance Controller-Design,” AIChE J., 38(10), 1593-1608(1992)
Roberts, S. W., “Control Chart Tests Based on Geometric Moving Averages,” Technometrics, Vol. 1(1959)
Shewhart, W. A., “Economic Control of Manufactured Product,” Van Nostrand Princeton, NJ (1931).
Wahl, Edward F. and Peter Harriott, “Understanding and Prediction of the Dynamic Behavior of Distillation Columns,” Ind. Eng. Chem. Proc. Des. Develop., 9(3) (1970)
Wise, Barry M. and Neal B. Gallagher, “The Process Chemometrics Approach to Process Monitoring and Fault Detection,” J. Proc. Cont., 6(6), 329-348(1996)
Whitehead, D. B. and Parnis M., “Computer Control Improves Ethylene Plant Operation,” Hydrocarbon Process, 105-108(1987)