研究生: |
鄭又慈 |
---|---|
論文名稱: |
變動 Quasi MMSE 控制器在動態批次回饋系統之績效分析 Variable Quasi MMSE Run-to-Run Controller for Dynamic Systems |
指導教授: | 曾勝滄 |
口試委員: |
洪志真
鄭順林 林建華 曾勝滄 |
學位類別: |
碩士 Master |
系所名稱: |
理學院 - 統計學研究所 Institute of Statistics |
論文出版年: | 2011 |
畢業學年度: | 99 |
語文別: | 中文 |
論文頁數: | 55 |
中文關鍵詞: | 批次控制 、動態系統 、Quasi MMSE 控制器 、最適變動折扣因子 |
外文關鍵詞: | Run-to-run Control, Dynamic System, Quasi MMSE Controller, Optimal Variable Discount Factor |
相關次數: | 點閱:2 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 |
傳統的批次回饋控制方法大多假設製程之投入-產出模型為線性靜態模型,並以 Single EWMA 控制器來調整製程產出。然而,實際製程之投入-產出模型大多為動態系統,此時 Single EWMA 控制器若採用適當的折扣因子來調整生產製程,雖可確保製程收斂,但將導致製程變異放大,不良率增加。針對此缺點,在一階動態系統下,雖有Quasi MMSE (minimum mean square error) 控制器可用以克服此困難,唯當起始偏差過大時將造成出初期程變異控制不佳。因此,本研究擬提出變動折扣因子之 Quasi MMSE 控制器,以期使製程產出的不良率降低。
首先,本文以有限批量的總均方誤差 (Total Mean Square Error,簡稱TMSE) 極小化為準則,建構出具有最適變動折扣因子之Quasi MMSE控制器;並以重製率 (Rework Rate,簡稱RR) 為準則比較此控制器與固定折扣因子的Quasi MMSE控制器之績效表現;文中發現當預測模型中的斜率參數高估且估計誤差小時,本文所提出之控制器重製率的績效表現優於固定折扣因子Quasi MMSE控制器。此外,本文亦提出估計未知動態參數之方法。為確保變動折扣因子Quasi MMSE控制器重製率的績效表現優於固定折扣因子Quasi MMSE控制器,當動態參數越大時,所需樣本數越大,可使估計誤差降低。且在相同的動態參數設定下,當預測模型中的斜率參數高估時,所需樣本數會較斜率參數低估時來的少。
[1]Box, G. E.,G. M. Jenkins, and G. C. Reinsel. (1994). Time Series Analysis-Forecasting and Control. 3nd ed. Upper Saddle River, NJ:Prentice-Hall.
[2]Capilla, C., Ferrer, A., Romero, R. and Hualda, A. (1999). “Integration of statistical and engineering process control in a continuous polymerization process,” Technometrics, 41 (1), 14-28.
[3]Del Castillo, E. (2002). Statistical Process Adjustment for Quality Control, John Wiley & Sons Inc., New York.
[4]Ingolfsson, A. and Sachs, E. (1993). “Stability and sensitivity of an EWMA controller,” Journal of Quality Technology, 25, 271-287.
[5]Jen, C. H. and Jiang, B. C. (2008). “Combining on-line experiment and process control methods for changes in a dynamic model,” International Journal of Production Research, 46 (13), 3665-3682.
[6]Pan, R. and Del Castillo, E. (2001). “Identification and fine tuning of closed-loop process under discrete EWMA and PI adjustments,” Quality of Reliability Engineering International, 17, 419-427.
[7]Sachs, E., Guo, R. S., Ha, S. and Hu, A. (1991). “Process control system for VLSI fabrication,” IEEE Transactions on Semiconductor Manufacturing, 4, 134-144.
[8]Tseng, S. T. and Tsung, F. and Liu, P-Y. (2007). “Variable EWMA run-to-run controller for a drift process,” IIE Transactions, 39, 291-301
[9]Tseng, S. T. and Lin, C. H. (2009). “Stability analysis of single EWMA controller under dynamic models,” IIE Transactions, 41, 654-663.
[10]Tseng, S. T.*, Jou, B. Y. and Liao, C. H. (2010). “Adaptive variable EWMA controller for drifted processes,” IIE Transactions (Focus on Quality and Reliability Engineering), (SCI).
[11]Tseng, S. T.*, Yeh, A. B., Tsung, F., Chan, I. Y. (2003). “A study of variable EWMA controller,” IEEE Transactions on Semiconductor Manufacturing, 16(4), 633-643.
[12]李淑伶 (2010). “動態批次回饋系統的Quasi MMSE 控制器之績效分析,” 國立清華大學統計學研究所碩士論文.