首页 | 本学科首页   官方微博 | 高级检索  
     

神经网络用于估算透镜尺寸误差对束斑的影响
引用本文:屠彦 张雄. 神经网络用于估算透镜尺寸误差对束斑的影响[J]. 东南大学学报(自然科学版), 1998, 28(6): 54-58
作者姓名:屠彦 张雄
作者单位:东南大学电子学研究所
基金项目:国家教委留学回国人员科研启动基金
摘    要:将有限差分法与神经网络技术结合起来用于估算电子枪主透镜尺寸误差的存在对屏上电子束束斑的影响。与以往的计算方法相比,该方法可避免许多复杂计算,节省了计算时间,一旦网络训练完毕,在很短的时间内即可估算出电子枪主透镜尺寸误差对屏上电子束束斑的影响,对提高CRT的成品率很有帮助。

关 键 词:估算 透镜 束斑 神经网络 有限差分 电子枪 尺寸误差 电子束

Application of Neural Networks Technique to Estimate the Effects of Main Lens Dimension Error on the Spots
Tu Yan Zhang Xiong. Application of Neural Networks Technique to Estimate the Effects of Main Lens Dimension Error on the Spots[J]. Journal of Southeast University(Natural Science Edition), 1998, 28(6): 54-58
Authors:Tu Yan Zhang Xiong
Abstract:The infinite difference method is combined with the neural network technique. It is used to estimate the effects of the electron gun main lens dimension error to the electron beam spots on the screen. Compared with the normal calculation method, this method can save computation time, avoid much complex calculation and save many computer sources. Once the network is well trained, it can predict the effects of the electron gun main lens dimension error to the electron beam spots on the screen in a short time. This method is very useful to the CRT production.
Keywords:neural networks  finite difference method  electron gun  dimension error  
本文献已被 CNKI 维普 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号