應用SPC與適應參數估計方法於系統監控

Abstract

隨著系統安全性與可靠度的要求日益提高，一個良好的系統必須同時 兼顧品質與安全考慮，所以監控系統與錯誤診斷系統已成為精密製程中及 關鍵的系統。 在本篇論文可分為兩個部份，一、利用統計程序控 制 ( statistic process control ; SPC ) 的方法建立離子植入均勻監 控系統，由於離子植入為 IC 製程中極關鍵的步驟，若發生植入不均勻的 情形，將造成經濟上的損失。我們藉由對於離子植入機的瞭解，擷取離子 植入均勻之監控訊號以及監控參數，並且應用機率與統計的方法對於監控 參數進行監控。二、利用系統參數估測的方法建立監控系統或是錯誤診斷 系統，這部份的重點是，使用適應 IIR 濾波器的方法來做線上參數估測 ，而我們採用了變數調整量 ( variable step size ; VSS ) 的方式於二 次限制方程式誤差 ( quadratic constraint eqation error ; QCEE ) 調整法，且對於參數的估測過程具有收斂速度快的特性。對於參數收斂的 結果具有準確性的特性，對於突然的系統改變，我們設計的適應 IIR 濾 波器，其估測參數具有較快的反應力。 As the demands on safety and reliability of modern systems become more critical,it is desired that systems not only achieve the performance requirements but also guarantee the safety concern. Therefore, the monitoring and diagnostic system has become an important subsystem in the precise processing. There are two parts in this thesis. In first part, the statistic process control ( SPC ) method is employed for the ion implanter to monitoring the uniformity of the ion implantation. Ion implantation process plays an important roll in the IC manufacturing process, nonuniform implantation over wafer will lead to serious economic loss. By processing themonitoring signals, we achieve the suitable monitoring indices and they are monitored during the ion implantation process. In the second part, we integrate the variable step size ( VSS ) method and the quadratic constraint equation error ( QCEE ) adaptive algorithm to achievea new adaptive infinite impulse filter (IIR) for parameter estimation. The proposed adaptive IIR can be applied to on-line monitoring for general dynamicsystems. Results indicate that this new approach achieves faster converge and more accurate estimation.