Multi-way principal component analysis（MPCA）has received considerable attention and been widely used in process monitoring.A traditional MPCA algorithm unfolds multiple batches of historical data into a two-dimensio...Multi-way principal component analysis（MPCA）has received considerable attention and been widely used in process monitoring.A traditional MPCA algorithm unfolds multiple batches of historical data into a two-dimensional matrix and cut the matrix along the time axis to form subspaces.However,low efficiency of subspaces and difficult fault isolation are the common disadvantages for the principal component model.This paper presents a new subspace construction method based on kernel density estimation function that can effectively reduce the storage amount of the subspace information.The MPCA model and the knowledge base are built based on the new subspace.Then,fault detection and isolation with the squared prediction error（SPE）statistic and the Hotelling（T2）statistic are also realized in process monitoring.When a fault occurs,fault isolation based on the SPE statistic is achieved by residual contribution analysis of different variables.For fault isolation of subspace based on the T2 statistic,the relationship between the statistic indicator and state variables is constructed,and the constraint conditions are presented to check the validity of fault isolation.Then,to improve the robustness of fault isolation to unexpected disturbances,the statistic method is adopted to set the relation between single subspace and multiple subspaces to increase the corrective rate of fault isolation.Finally fault detection and isolation based on the improved MPCA is used to monitor the automatic shift control system（ASCS）to prove the correctness and effectiveness of the algorithm.The research proposes a new subspace construction method to reduce the required storage capacity and to prove the robustness of the principal component model,and sets the relationship between the state variables and fault detection indicators for fault isolation.展开更多
为了处理张量数据,传统的学习算法常常把张量展成向量,但会造成破坏原始数据固有的高阶结构和内在相关性,导致信息丢失,或产生高维向量,使得后期学习过程中容易出现过拟合、维度灾难和小样本问题.近年提出了许多基于张量模式的分类算法...为了处理张量数据,传统的学习算法常常把张量展成向量,但会造成破坏原始数据固有的高阶结构和内在相关性,导致信息丢失,或产生高维向量,使得后期学习过程中容易出现过拟合、维度灾难和小样本问题.近年提出了许多基于张量模式的分类算法,而支持高阶张量机算法是张量分类算法中最有效的方法之一.考虑到张量的高维性和高冗余性,本文提出基于多线性主成分分析的支持高阶张量机分类算法(Multilinear Principle Component Analysis Based Support High-Order Tensor Machine,MPCA+SHTM).该算法首先利用多线性主成分分析对张量进行降维,然后利用支持高阶张量机对降维后的张量进行学习.在12个张量数据集上的实验表明:MPCA+SHTM在保持测试精度的情况下有效地降低了SHTM的计算时间.展开更多
以掺假山茶油样为气相离子迁移谱(gas chromatography-ion mobility spectrometry,GC-IMS)检测对象,利用多维主成分分析(multi-way principal component analysis,MPCA)法和偏最小二乘(partial least squares,PLS)回归分析处理二维谱图...以掺假山茶油样为气相离子迁移谱(gas chromatography-ion mobility spectrometry,GC-IMS)检测对象,利用多维主成分分析(multi-way principal component analysis,MPCA)法和偏最小二乘(partial least squares,PLS)回归分析处理二维谱图数据,探索并建立一种山茶油纯度检测方法。对配制的不同比例3种食用植物油的掺假油样进行GC-IMS检测,采用MPCA压缩并提取矩阵中的得分矩阵进行主成分分析,将提取的得分矩阵进行PLS分析,建立掺假量的定量预测模型。结果表明,MPCA处理后的主成分图可以明显区分山茶油样和掺入不同种类食用油的掺假山茶油样,且不同掺入比例组有其明显的归属区域;采用PLS对MPCA的得分矩阵进行回归分析,可实现对山茶油掺假比例的准确定量测定。该方法具有快速、准确、无损的特点,可应用推广到其他联用仪器的数据分析处理中,在食用油品质控制与评价方法中具有很大的应用前景。展开更多
文章在移动主成分分析(moving principal component analysis,MPCA)基础上,提出一种优化的MPCA特征——特征向量差方向角(directional angle of eigenvector variation,DAEV),并将其作为机器学习的输入建立损伤识别模型。利用双跨连续...文章在移动主成分分析(moving principal component analysis,MPCA)基础上,提出一种优化的MPCA特征——特征向量差方向角(directional angle of eigenvector variation,DAEV),并将其作为机器学习的输入建立损伤识别模型。利用双跨连续梁的仿真应变监测数据验证了以DAEV建立机器学习模型诊断结构损伤的有效性。结果表明,与MPCA特征向量相比,DAEV能更好地表征桥梁状态的变化,以DAEV为输入的机器学习模型损伤识别能力更强;对于早期损伤,以DAEV特征为输入的模型识别准确率比以MPCA特征向量为输入的模型高38%~79%。展开更多
针对传统的多向主元分析(Multiway Principal component Analysis,MPCA)常会导致误诊断,且对批生产过程难以保证在线状态监测和故障诊断的实时性,提出了一种改进的MPCA与动态时间错位(Dynamic Time Warping,DTW)方法,该方法采用多模型...针对传统的多向主元分析(Multiway Principal component Analysis,MPCA)常会导致误诊断,且对批生产过程难以保证在线状态监测和故障诊断的实时性,提出了一种改进的MPCA与动态时间错位(Dynamic Time Warping,DTW)方法,该方法采用多模型非线性结构代替传统的MPCA单模型线性化结构,并利用对称式DTW算法解决了多元轨迹同步化的问题。将该方法应用到青霉素发酵批过程的在线故障监测中,结果表明它克服了MPCA不能处理非线性过程和实时性问题,并避免了MPCA 在线应用时预报未来测量值带来的误差,提高了批过程性能监测和故障诊断的准确性。展开更多
基金Supported by National Hi-tech Research and Development Program of China(863 Program,Grant No.2011AA11A223)
文摘Multi-way principal component analysis（MPCA）has received considerable attention and been widely used in process monitoring.A traditional MPCA algorithm unfolds multiple batches of historical data into a two-dimensional matrix and cut the matrix along the time axis to form subspaces.However,low efficiency of subspaces and difficult fault isolation are the common disadvantages for the principal component model.This paper presents a new subspace construction method based on kernel density estimation function that can effectively reduce the storage amount of the subspace information.The MPCA model and the knowledge base are built based on the new subspace.Then,fault detection and isolation with the squared prediction error（SPE）statistic and the Hotelling（T2）statistic are also realized in process monitoring.When a fault occurs,fault isolation based on the SPE statistic is achieved by residual contribution analysis of different variables.For fault isolation of subspace based on the T2 statistic,the relationship between the statistic indicator and state variables is constructed,and the constraint conditions are presented to check the validity of fault isolation.Then,to improve the robustness of fault isolation to unexpected disturbances,the statistic method is adopted to set the relation between single subspace and multiple subspaces to increase the corrective rate of fault isolation.Finally fault detection and isolation based on the improved MPCA is used to monitor the automatic shift control system（ASCS）to prove the correctness and effectiveness of the algorithm.The research proposes a new subspace construction method to reduce the required storage capacity and to prove the robustness of the principal component model,and sets the relationship between the state variables and fault detection indicators for fault isolation.
文摘为了处理张量数据,传统的学习算法常常把张量展成向量,但会造成破坏原始数据固有的高阶结构和内在相关性,导致信息丢失,或产生高维向量,使得后期学习过程中容易出现过拟合、维度灾难和小样本问题.近年提出了许多基于张量模式的分类算法,而支持高阶张量机算法是张量分类算法中最有效的方法之一.考虑到张量的高维性和高冗余性,本文提出基于多线性主成分分析的支持高阶张量机分类算法(Multilinear Principle Component Analysis Based Support High-Order Tensor Machine,MPCA+SHTM).该算法首先利用多线性主成分分析对张量进行降维,然后利用支持高阶张量机对降维后的张量进行学习.在12个张量数据集上的实验表明:MPCA+SHTM在保持测试精度的情况下有效地降低了SHTM的计算时间.
文摘以掺假山茶油样为气相离子迁移谱(gas chromatography-ion mobility spectrometry,GC-IMS)检测对象,利用多维主成分分析(multi-way principal component analysis,MPCA)法和偏最小二乘(partial least squares,PLS)回归分析处理二维谱图数据,探索并建立一种山茶油纯度检测方法。对配制的不同比例3种食用植物油的掺假油样进行GC-IMS检测,采用MPCA压缩并提取矩阵中的得分矩阵进行主成分分析,将提取的得分矩阵进行PLS分析,建立掺假量的定量预测模型。结果表明,MPCA处理后的主成分图可以明显区分山茶油样和掺入不同种类食用油的掺假山茶油样,且不同掺入比例组有其明显的归属区域;采用PLS对MPCA的得分矩阵进行回归分析,可实现对山茶油掺假比例的准确定量测定。该方法具有快速、准确、无损的特点,可应用推广到其他联用仪器的数据分析处理中,在食用油品质控制与评价方法中具有很大的应用前景。
文摘文章在移动主成分分析(moving principal component analysis,MPCA)基础上,提出一种优化的MPCA特征——特征向量差方向角(directional angle of eigenvector variation,DAEV),并将其作为机器学习的输入建立损伤识别模型。利用双跨连续梁的仿真应变监测数据验证了以DAEV建立机器学习模型诊断结构损伤的有效性。结果表明,与MPCA特征向量相比,DAEV能更好地表征桥梁状态的变化,以DAEV为输入的机器学习模型损伤识别能力更强;对于早期损伤,以DAEV特征为输入的模型识别准确率比以MPCA特征向量为输入的模型高38%~79%。
文摘针对传统的多向主元分析(Multiway Principal component Analysis,MPCA)常会导致误诊断,且对批生产过程难以保证在线状态监测和故障诊断的实时性,提出了一种改进的MPCA与动态时间错位(Dynamic Time Warping,DTW)方法,该方法采用多模型非线性结构代替传统的MPCA单模型线性化结构,并利用对称式DTW算法解决了多元轨迹同步化的问题。将该方法应用到青霉素发酵批过程的在线故障监测中,结果表明它克服了MPCA不能处理非线性过程和实时性问题,并避免了MPCA 在线应用时预报未来测量值带来的误差,提高了批过程性能监测和故障诊断的准确性。
文摘2012年4～8月,在太行山猕猴国家级自然保护区济源管理局天坛山管护区(北纬35°05'～35°15',东经112°12'～112°22'),对太行山猕猴王屋1群(WW-1)内的3个母系单元(matrilineal unit)中大于(等于)3岁龄的26只个体进行面部拍照,获取其面部特写照片,进而利用分块主成分分析(modular principal component analysis,MPCA)法,对个体进行面部识别分析,旨在探讨个体间面部相似度与亲缘关系的相关性。结果表明:(1)太行山猕猴个体间的面部相似度与亲缘类型有关,母亲与大于3岁龄子代间的面部相似度为0.93±0.00,显著高于单元内(0.89±0.00)和单元间(0.84±0.01)的面部相似度;(2)太行山猕猴个体的面部特征随年龄增长而变化,4岁(含4岁)龄以上个体与母亲间的面部相似度较高(0.88～0.95),依此值可准确地识别母子关系。本研究采用量化方法对非人灵长类个体间面部相似度进行分析,发现太行山猕猴个体间的面部相似度与亲缘关系密切相关;研究结果可为非人灵长类的个体识别提供较为客观的手段和方法。
