本研究之主要目的為建立一套應用於電路板表面黏著技術的最前段鋼製網版印刷製程,判斷網版印刷機於印刷後錫膏厚度機台的製程參數與該錫膏印刷流程的錫膏厚度之類神經網路模型,再將此類神經網路之知識萃取成布林規則形式,以提供建立專家系統知識庫之用。在資料的收集方面,包括了該道製程中,使用印刷壓力、印刷速度、脫膜距離、脫模速度及清潔次數與模式等,配合著自動光學影像量測機對於該錫膏印刷製程成品的檢查結果,分別以此兩項數據為輸入與輸出,採用倒傳遞的學習法則來建立類神經網路系統模型,總共蒐集了400資料作為訓練的樣本,來進行500次的訓練,訓練完畢後的類神經網路模型並且取樣65筆測試樣本的檢測,其平均的誤差值為2.24%,結果顯示出此一系統理論可以能夠有效的預測此道網版印刷製程之錫膏厚度品質狀況。由此神經網路模型,再使用規則萃取演算法來擷取出其所學習到知識,因為這些知識能以布林邏輯的規則方式呈現,所以將這些定性規則予以整理、檢查和解釋,並彙整成專家系統知識庫形式,就可以作為SMT製程工程師在調整印刷製程時更改印刷機台參數時的參考依據和調整方向,並且可以提升生產效率和縮短調整印刷機台時間。
The principal object of this research is to develop a neural network model, which can simulate the solder paste thickness process in Surface Mounting Technology (SMT).Then the Boolean logic rules were extracted from the trained network model to establish a knowledge base of the expert system. Here the input data of neural network was collected from the process parameters of printer machine (DEK) in the SMT department, which included the pressure and speed of the printing, distance and speed of the separation, frequency and mold of the cleaning. The inspecting values from the Auto Optical Inspection (AOI) were adopted as the output. The neural network model was constructed with back-propagation network (BPN) algorithm and trained 500 times with 400 training data. The model was tested with 65 test data and the error rate was ±2.24% in average. The extraction rules were obtained from the trained neural network with the rule extraction algorithm to describe the knowledge in the network system with logic rules. After checking, explaining and integrating the rules into the knowledge base, the rules can be the basics of solder paste thickness prediction and alarm diagnosis in print process system. When the parameters are abnormal, to make the solder paste thickness out of control limit, an alert message can be offered and also the knowledge base can provide a reference to the engineers for the work of recipe adjustments.