变间距组合预测模型在电子注采样中的应用  

Application of variable-space combinatorial predictionmodel in electron beam sampling

作　　者：施炜 曹林林[1,2] 黄明光 Shi Wei;Cao Linlin;Huang Mingguang(Research and Development Center of Space Traveling Wave Tubes,Institute of Electronics,Chinese Academy of Sciences,Beijing 101407,China;College of Electronics,Electrics and Communication Engineering,University of Chinese Academy of Sciences,Beijing 101408,China)

机构地区：[1]中国科学院电子学研究所空间行波管研究发展中心,北京101407 [2]中国科学院大学电子电气与通信工程学院,北京101408

出　　处：《电子测量与仪器学报》2018年第11期50-57,共8页Journal of Electronic Measurement and Instrumentation

基　　金：国家自然科学基金(61401430)资助项目

摘　　要：为使电子注电流密度预测精度更高,基于现有等间距和变间距预测模型,提出了改进的变间距灰色模糊马尔可夫组合预测模型在电子注采样中的应用。将变间距灰色模型用于预测各个采样位置的采样结果,将模糊数学方法用于分割论域、模糊化区间,将变间距马尔可夫模型用于修正灰色预测的结果。根据实验得到的原始数据,代入组合模型计算可知,组合模型能很好地预测电流密度数据,可应用于某些采样位置数据缺失的情形,预测精度达98%以上,且跟踪效果良好。模型依靠有限位置的原始电流密度值预测后续采样点的电流密度分布,对测量探头的冷却时间有个大致的估计,不至于让探头持续过热影响检测性能,从而提高测量精度、延长使用寿命。In order to improve the prediction accuracy of electron beam current density,the application of the improved variable-distance grey-fuzzy-Markov forecasting model in electron beam sampling is discussed based on the existing equal and variable interval prediction models.The variable-space grey model is used to predict the sampling results of each sampling point. The fuzzy mathematics method is used to segment the discourse domain and to blur the interval. The variable-distance Markov model is used to modify the grey prediction results. According to the experimental raw data, and substitution calculations the model can be used to predict the current density. When the data is used in the case of missing some sampling positions, the prediction accuracy can be higher than 98% with satisfactory tracking effect. The proposed model relies on the original current density values of the finite positions to predict the current density distribution of the subsequent sampling points, and has a rough estimate of the cooling time of the measuring probe, so that the probe does not overheat continuously and affect the detection performance, thereby improving the measurement accuracy and prolonging service life.

关 键 词：电子注 变间距模型 灰色模糊马尔可夫预测 

分 类 号：TP29[自动化与计算机技术—检测技术与自动化装置] TN98[自动化与计算机技术—控制科学与工程]