查詢結果分析
來源資料
相關文獻
- 類神經網路於半導體平坦化製程預測監控功能之研究
- Applying Artificial Neural Network to Semiconductor Planarization Process Forecasting and Monitoring Functions
- Neural Network Procedures for Taguchi's Dynamic Problems
- A Fast and Efficient Competitive Learning Design Algorithm Based on Weight Vector Training in Transform Domain
- 專家系統振動訊號圖型判別之研究
- 反傳遞模糊類神經網路於流量推估之應用
- 類神經網路(Neural Networks)的種類及其在影像處理上的應用
- C++Fuzzy類神經網路物件導向發展系統之建立
- 臺灣汽保費率之估計--對數線性費率模式與類神經網路之比較
- 運用類神經網路於股價指數之套利--以日經225指數為例
頁籤選單縮合
題 名 | Applying Artificial Neural Network to Semiconductor Planarization Process Forecasting and Monitoring Functions=類神經網路於半導體平坦化製程預測監控功能之研究 |
---|---|
作 者 | 田效文; 謝家祥; 鍾宜展; 蔡志弘; 王志宏; | 書刊名 | 品質學報 |
卷 期 | 12:4 民94.12 |
頁 次 | 頁273-284 |
分類號 | 494.566 |
關鍵詞 | 半導體製程平坦化; 類神經網路; Planarization process; Artificial neural network; |
語 文 | 英文(English) |
中文摘要 | 半導體產業存在許多多樣化的製程及複雜的製程參數,其中在平坦化製程中是硼磷矽玻璃(Boron Phosphate Silicon Glass, BPSG) 之製程,其製程品質將影響微影技術達到微小圖形的高解析度;由於平坦化製程是需要BPSG化學氣相沈積 (Chemical Vapor Deposition)、熱回流圓滑法 (Glass Thermal Flow)、化學機械研磨 (Chemical-Mechanical Polishing) 三段不同的製程站點所完成,本研究以倒傳遞類神經網路為主,以建立預測平坦化製程之品質為目標,減少工程人員在尋求製程最佳化時,重複實驗的次數與時間,並分析其製程參數對於品質的影響,經由線上即時製程與產品參數的取得,找出與品質相關的可能製程參數,透過三組不同類神經網路模式的組合可得知在此製程狀態下,其品質的是否合格,同時本研究也探討平坦化製程中對於品質存在影響之參數。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 (Back Propagation Artificial Neural Network) 對於品質預測之結果較為準確。 |
英文摘要 | There are a great variety of processes and complex process parameters in semiconductor industry. In planarization process, the quality of BPSG (Boron Phosphate Silicon Glass) process determines the clarity of mini pictures by lithography. Therefore, this research focuses on back propagation artificial neural network, aims to establish the quality of planarization process forecast, reduce the number and time of repeated experiments when engineers seek process optimization, analyze the impact of process parameters on quality, identify the potential process parameters that are related to quality by acquiring online real-time process parameters and product parameters, determine whether the quality meets the requirements under the specified process through combinations of three different artificial neural network models, and analyze the parameters of the impact that planarization process has on quality. In studying the applicability of artificial neural network to the cases in this research, the back propagation artificial neural network shows more accurate quality forecast. |
本系統中英文摘要資訊取自各篇刊載內容。