《计算机应用研究》|Application Research of Computers

改进的加权t-SNE算法及在故障诊断中的应用

Improved weighted t-SNE algorithm and application in fault diagnosis

免费全文下载 (已被下载 次)  
获取PDF全文
作者 夏丽莎,方华京
机构 1.上海理工大学 管理学院,上海 200093;2.华中科技大学 自动化学院,武汉 430074
统计 摘要被查看 次,已被下载
文章编号 1001-3695(2020)07-032-2078-04
DOI 10.19734/j.issn.1001-3695.2018.12.0952
摘要 对随机邻域嵌入算法(SNE)中的距离进行改进,提出一种基于Manhattan距离的加权t-SNE(Mwt-SNE)算法。使用受空间维数影响较小的Manhattan距离作为度量方式,使用K-均值聚类算法将高维空间数据样本点距离分为三类,基于表格法进行权重参数寻优与加权,以加权相对Manhattan距离代替欧氏绝对距离计算相似度条件概率,从而增大数据对象之间的区分度,提升降维效果,增强分类显著性。提出基于Mwt-SNE算法的在线故障诊断模型,使用核密度估计(KDE)确定控制限并进行在线监控。TE化工过程实验表明,Mwt-SNE算法能有效降低误报率和漏报率,从而提高故障诊断稳定性和准确性。
关键词 故障诊断; 加权t-SNE; Manhattan距离; 核密度估计
基金项目 国家自然科学基金资助项目(61473127,71572113)
本文URL http://www.arocmag.com/article/01-2020-07-032.html
英文标题 Improved weighted t-SNE algorithm and application in fault diagnosis
作者英文名 Xia Lisha, Fang Huajing
机构英文名 1.School of Business,University of Shanghai for Science & Technology,Shanghai 200093,China;2.School of Automation,Huazhong University of Science & Technology,Wuhan 430074,China
英文摘要 This paper proposed a novel Manhattan distance based weighted t-SNE(Mwt-SNE) algorithm on the basis of improved distance in stochastic neighbor embedding(SNE). Firstly, it calculated samples Manhattan distances rather than Euclidean distance from high dimensional space for their less affections of dimension. Next, it divided these Manhattan distances into three groups with K-means clustering algorithm and implemented weighting processing separately with tabular parameter optimization method. Then it calculated similarity conditional probabilities with weighted Manhattan distances according to values category distribution. The aim of weighted Manhattan distances was to enlarge the data distinction, promote dimension reduction and enhance classification significance. Finally, it established an Mwt-SNE algorithm based on-line fault diagnosis model and the corresponding control limit with KDE. The experimental results on TE chemical process show that the proposed Mwt-SNE algorithm reduces FAR(false alarm rate) and MAR(missing alarm rate) as well as improves stability and accuracy.
英文关键词 fault diagnosis; weighted t-SNE; Manhattan distance; kernel density estimation(KDE)
参考文献 查看稿件参考文献
 
收稿日期 2018/12/13
修回日期 2019/1/25
页码 2078-2081
中图分类号 TP277
文献标志码 A