基于流形学习的局部保持PCA算法在故障检测中的应用

提出一种新的基于流形学习的数据降维及特征提取方法:局部保持 PCA 算法(LPPCA).通过在 PCA 的优化目标中融入流形学习的思想,不仅使投影得到的低维空间和原始样本空间具有相似的全局结构,并且保持了相似的局部近邻结构,克服了传统 PCA 方法只关注全局结构特征而忽略局部流形特征的缺陷,同时给出了 LPPCA 在故障检测中的应用方法. S-Curve 和 Swiss-roll 曲面数值仿真和 TE 过程仿真验证了算法的有效性和优越性.

Locally preserving PCA method based on manifold learning and its application in fault detection

A novel dimensionality reduction and feature extraction method based on manifold learning, locally preserving
principal component analysis(LPPCA) is proposed. In order to overcome the defects that the traditional PCA can only keep
the structure in global and can not maintain the manifold structure in local, the idea of locality preserving is incorporated
into the optimization goals of the PCA. The fault detection based on LPPCA is researched. The validity and superiority
of the LPPCA are verified by the S-Curve numerical simulation, Swiss-roll surface numerical simulation and TE process
simulation.