欢迎您！东篱公司
退出
-
申报数据库
-
立项数据库
-
成果数据库
-
项目获奖数据库
中文摘要：
针对批次过程数据具有高维、非线性及多模态等特性,提出一种自适应LPP-k NN的过程监视方法.利用局部保持映射算法（LPP）提取高维多模态批次数据的自适应变换矩阵构成新的建模数据.采用局部近邻标准化方法（LNS）进行标准化,并利用k NN算法构造统计监测指标.最后,通过在半导体工业实例中的应用验证了所提算法的有效性.
英文摘要：
In order to address the high dimensionality and multiple conditions o f batch process data, amethod of LPP--NN is proposed in this aticle. Firstly,this method is based on locality preserving projection （LPP） which can extact adaptive tansformation matrix of the Vidor High modal batch modeling data. Then,standardization o local neighborhood （LNS ） is processed to overcome the data charac-ter of multiple conditions. Meanwhile, fc-nearest neighbor （ NN ） is applied for fault detection with constrnc- ting statistical indicators. Finally, va ie ty of improved kNN /gorielms ae applied in semiconductor indus-try examples and te effectiveness of the proposed metod has been verified by comparing.
同期刊论文项目
同项目期刊论文
Palmprint Recognition Based on Multilinear Principal Component Analysis and Fisher Linear Discrimina
期刊信息
位置：