為了在今日全球化的市場上競爭,製造業者不斷的嘗試應用許多的品質管制方法(如管制圖)來監測產品的生產流程以降低成本。但由於監測的產品品質特性,基本上是一種具有高度自我相關的觀測資料,因此常造成製程干擾現象被誤判的結果。為了解決這樣的問題,在實際執行監測的過程中,品管人員常會採用一般化時間序列模式下的殘差控制圖來進行資料異動形態之辨別。然而,使用殘差控制圖的辨識成功率卻會隨著觀測資料的自我相關程度(當期資料與前一期資料的相關係數高低)而變動;而且在多數情況下,資料相關程度的先前資訊往往無法事先獲得。由於相關性製程資料可被視為一個混合了雜訊及製程干擾訊號的混合資料(mixture data),因此本研究導入近年來被成功應用在訊號分離(signal separation)之獨立成份分析(Independent Component Analysis, ICA)技術於製程管制中,利用ICA能將混合訊號分離出潛在來源訊號(latent source signals)之能力,提出一個結合獨立成份分析與分類迴歸樹(Classification and Regression Tree, CART)的製程監控架構。在所提的架構中,本研究先利用EPC與ICA將製程觀察值(即混合訊號)分離出相互獨立之獨立成份(independent component),之後再使用分類迴歸樹針對獨立成份進行監控,期望能解決傳統管制圖需要對相關性製程的資料進行繁複處理的問題。換言之,本研究的研究重點有二:(一)提出獨立成份分析(Independent component analysis, ICA)的轉換機制,使得原始資料可被分離出更容易被辨識的資訊。(二)嘗試應用分類迴歸樹進行干擾型態的辨識以解決應用傳統管制圖無法清楚解釋製程管制規則的缺點。本研究針對在IMA(1, 1)下兩種不同類型的製程干擾資料進行製程管制的模擬研究測試,以驗證所提出整合方法之有效性。根據研究結果顯示,藉由SPC/EPC/ICA技術的應用,雜訊(noise)確實可以從相關性製程資料中被成功的分離出來。
It is well known for many years that statistical process control (SPC) and engineering process control (EPC) are both effective in the monitoring and adjusting of the manufacturing process. However, both schemes assume that process data are identically and independently distributed. Moreover, the real process data are actually serially correlated. The presence of autocorrelation has an adverse effect on the performance of traditional SPC or EPC approach. Because the correlated process data could be a mixture of noise and process characteristics such as process disturbances and/or autocorrelation, a disturbance identification scheme based on independent component analysis (ICA) is proposed in this paper. Basically, ICA is a novel statistical signal processing technique that was originally applied to find the latent source signals from observed mixture signal. In this paper, the ICA technique is first integrated with SPC and EPC approach to extract the independent components that contain different characteristics of the process. Then, the CART methodology is used to monitor the independent components and identify the process disturbance.
為了持續優化網站功能與使用者體驗,本網站將Cookies分析技術用於網站營運、分析和個人化服務之目的。
若您繼續瀏覽本網站,即表示您同意本網站使用Cookies。