應用約略集合理論建立電磁干擾診斷系統

Application of Rough Set Theory to EMI Diagnosis

黃承龍(Cheng-Lung Huang) ； 李得盛(Te-Sheng Li) ； 彭定國(Ting-Kuo Peng) ； 施政男(Cheng-Nan Shih)

《管理與系統》 11卷3期 (2004/09) Pp. 367-385

摘要

「電磁干擾」(EMI, Electromagnetic Interference)可能造成工廠機器故障、錯誤動作而發生不可預料的情況。如果產品有EMI品質不良，必須針對不符合測試標準的產品進行修改，直到產品通過標準測試值之後，才能將產品出貨。EMI診斷由有經驗的工程師找到EMI問題的根源，此工作耗時且困難，建立一套EMI診斷決策系統有助於縮短診斷時間。本研究應用資料探勘領域中的知識發掘理論－約略集合理論(RST, Rough Set Theory)找到診斷知識。RST用來處理含糊與不精確資訊，此法不需要對資料集有先驗或是額外的資訊。以RST探勘輸入變數與不良位置之間存在的法則知識，其步驟如下：資料蒐集、資料前處理、屬性值離散化、屬性化簡、最小屬性集過濾、產生法則、過濾法則、做分類、計算分類正確率以及建立診斷系統等步驟。以國內知名主機板廠商的EMI歷史資料做資料探勘，透過上述步驟所建立的RST診斷法則平均有八成的正確率，已達到實務應用的要求。

關鍵字

資料探勘 ； 電磁干擾 ； 約略集合理論 ； 屬性化簡

並列摘要

Electromagnetic emissions are radiated from every part of motherboards of personal computers, and thus electromagnetic interference (EMI) occurs. EMI has a bad effect on the surrounding environment because EMI may cause malfunctions or fatal problems of other digital devices. EMI engineers diagnose EMI problems of motherboard from the electromagnetic noise data measured by the Spectrum Analyzer. It is time consuming to find out the sources (PS2, USB, VGA, etc.) of electromagnetic noise. Rough set theory (RST) is a new mathematical approach to data analysis. This paper constructs an EMI diagnostic system based on RST. There are the following steps: Data Collection, Data Preprocessing, Descretization, Attribute Reduction, Reduction Filtering, Rule Generation, Rule Filtering, Classification, and Accuracy Calculation. Historical EMI noise data, colleted from a famous motherboard company in Taiwan, are used to generate diagnostic rules. The result of our research (average diagnostic accuracy of 80%) shows that RST model is a promising approach to EMI diagnostic support system.

並列關鍵字

Data Mining ； Rough Set Theory ； Attributes Reduction ； Electromagnetic Interference

參考文獻

Ahna, B. S.,Cho, S. S.,Kim, C. Y.(2000).The Integrated Methodology of Rough Set Theory and Artificial Neural Network for Business Failure Prediction.Expert Systems with Applications.18,65-74.

Archambeault, B.,Connor, S.(2002).IEEE International Symposium on EMC.

Bayardo, R. J.,Rakesh, A.,Dimitrios G.(1999).Constraint-Based Rule Mining in Large, Dense Databases.Proceeding of the 15th International Conference on Data Engineering.(Proceeding of the 15th International Conference on Data Engineering).

Ben-Arieh, D.,Chopra, M.,Bleyberg, M. Z.(1998).Data Mining Application for Real-time Distributed Shop Floor Control.IEEE International Conference on Systems, Man, and Cybernetics.(IEEE International Conference on Systems, Man, and Cybernetics).

Chen, F.-L.,Liu, S.-F.(2000).A Neural-Network Approach To Recognize Defect Spatial Pattern In Semiconductor Fabrication.IEEE transactions on semiconductor manufacturing.13(3),366-373.