應用類神經網路於製程品質參數之即時監控-以印刷電路板之DES製程為例

Application of the Neural Network in developing an on-line real-time monitoring and diagnostic system for DES process of PCB

李龍麟(Long-Lin Lee)

指導教授 ： 張百棧

元智大學/工程學院/工業工程與管理學系/碩士(2001年)

摘要

PCB產業產品的製造過程中有一段是DES之製程，有關顯影(Develop)、蝕刻(Etch)、去墨烘乾(Dry Film Strip)部分，其產出後印刷電路板上面之品質為數項參數所影響；本研究以倒傳遞類神經網路為主，迴歸分析為輔，案例式推導為比較，以建立印刷電路板DES製程即時品質監控為目標，分析其製程參數對於品質的影響，經由線上即時(online real time)製程參數的取得，利用迴歸分析找出與品質有顯著關係的製程參數，透過類神經網路模組便可得知在此製程狀態下，其品質的是否合格，同時本研究也探討那些製程參數對於PCB品質有主要之影響。比較案例式推導與類神經網路對於本研究案例的適用性，結果發現倒傳遞類神經網路對於品質預測之結果較為準確。

關鍵字

倒傳遞類神經網路 ； DES製程 ； 印刷電路板 ； 蝕刻因子

並列摘要

In this research, we developed an on-line real-time monitoring and diagnostic system for the Develop, Etch and dry film Strip (i.e., DES process) operations in the Printed Circuit Board (PCB) factory. From the empirical data of PCB factory, quality is affected by several key factors. Consequently, key factors that influenced the process should be identified first and through statistical regression analysis. The neural network expert control system is used to implement the real-time diagnostics and decisions will be made according to the input on-line captured from the system. Moreover, the criterion, root mean square error is used to evaluate the performances of case-base reasoning and backpropagation neural network. The comparisons result in that backpropagation neural network approach is superior to the case-base reasoning in this research.

並列關鍵字

Backpropagation neural network ； DES process ； Printed Circuit Board ； Etch factor

參考文獻

[1] Butler, Karen L. and A. James Momoh, "Robust features selection scheme for fault diagnosis in an electric power distribution system," Canadian Conference Electrical and Computer Engineering (Cat. No. 93TH0590-0), Vol.1, pp. 209-213, 1993.

[2] Carolyn M. et al., "An Investigation of Machine Learning Based Prediction Systems," The Journal of Systems and Software pp. 23-29, 2000.

[3] Dominguez, S. et al., "A Neural network Based Quality Control System for Steel Strip Manufacturing," IFAC Artificial Intelligence in Real Time Control, Valencia, Spain, pp. 185-190, 1994.

[4] He, Zhenyz et al., "Neural network and its application on machinery fault diagnosis," IEEE International Conference on Systems Engineering (Cat. No. 92CH3179-9), pp. 576-579, 1992.

[5] Hebb, D. O., Organization of behavior, Wiley, New York, 1949.