时序约束NPE算法在化工过程故障检测中的应用

针对动态过程的故障检测问题,在邻域保持嵌入算法中改进邻域挑选,提出一种新的维度约简方法：时序约束邻域保持嵌入（time constrained neighborhood preserving embedding,TCNPE）算法。与邻域保持嵌入（neighborhood preserving embedding,NPE）算法只通过欧氏距离挑选邻域不同的是,TCNPE考虑到数据之间的时序相关性,在一定长度的时间窗之内采用k-近邻方法挑选邻域,并对时间窗内近邻与非近邻构造局部约束关系。首先,利用TCNPE提取数据特征,进行线性降维,然后构造T²和SPE统计量并利用密度估计（kernel density estimation,KDE）确定其控制限。最后,通过数值例子和TE过程（Tennessee-Eastman process）仿真来说明本文方法的有效性。     

时序约束NPE算法在化工过程故障检测中的应用

针对动态过程的故障检测问题,在邻域保持嵌入算法中改进邻域挑选,提出一种新的维度约简方法:时序约束邻域保持嵌入(time constrained neighborhood preserving embedding,TCNPE)算法。与邻域保持嵌入(neighborhood preserving embedding,NPE)算法只通过欧氏距离挑选邻域不同的是,TCNPE考虑到数据之间的时序相关性,在一定长度的时间窗之内采用k-近邻方法挑选邻域,并对时间窗内近邻与非近邻构造局部约束关系。首先,利用TCNPE提取数据特征,进行线性降维,然后构造T~2和SPE统计量并利用密度估计(kernel density estimation,KDE)确定其控制限。最后,通过数值例子和TE过程(Tennessee-Eastman process)仿真来说明本文方法的有效性。     

局部时差约束邻域保持嵌入算法在故障检测中的应用

传统的邻域保持嵌入(neighborhood preserving embedding,NPE)算法通过k近邻(k-nearest neighbors,k-NN)方法选择邻域进行重构来实现降维。但在实际工业过程中采集的样本具有时序相关性,仅仅通过欧氏距离选择近邻样本不能充分反映数据中包含的信息,从而影响检测效果。因此,提出一种局部时差约束邻域保持嵌入(local time difference constrained neighborhood preserving embedding,LTDCNPE)算法,充分考虑样本间的时间和空间关系,从而建立准确的故障检测模型。首先,该算法在固定尺度的时间窗内,根据样本的时序关系和空间特征挑选出邻域。其次,利用样本间的时间差异为邻域样本进行加权,使数据特征保留了高维空间的局部结构。然后,对降维后得到的主元空间和残差空间构建T^{\mathrm{2}}和SPE统计量并确定控制限。最后,通过数值例子和Tennessee-Eastman(TE)过程仿真验证LTDCNPE算法的有效性。     

基于局部密度估计的多模态过程故障检测

以市场需求为导向的现代工业过程的生产条件要根据市场的需求不断做出调整,因此实际工业过程中存在多种工况的复杂情况,而过程的数据将不再完全服从高斯分布,其均值与协方差结构往往随着工况的切换而发生较大变化,为了能及时检测此类生产过程中的故障,提出一种新的基于带宽可变的局部密度估计的过程在线监控策略。首先利用局部投影保留（locality preserving projection, LPP）将高维数据投影到低维子空间中,充分地保留数据的局部结构;然后通过带宽可变的非参数密度核函数来进行局部密度估计,并采用局部密度因子（local density factor, LDF）的思想构造监控统计量,进而对工业过程故障进行在线检测;最后通过仿真研究,结果表明所提方法能够有效地应用于多模态过程的故障检测。     

基于局部密度估计的多模态过程故障检测

以市场需求为导向的现代工业过程的生产条件要根据市场的需求不断做出调整,因此实际工业过程中存在多种工况的复杂情况,而过程的数据将不再完全服从高斯分布,其均值与协方差结构往往随着工况的切换而发生较大变化,为了能及时检测此类生产过程中的故障,提出一种新的基于带宽可变的局部密度估计的过程在线监控策略。首先利用局部投影保留(locality preserving projection,LPP)将高维数据投影到低维子空间中,充分地保留数据的局部结构;然后通过带宽可变的非参数密度核函数来进行局部密度估计,并采用局部密度因子(local density factor,LDF)的思想构造监控统计量,进而对工业过程故障进行在线检测;最后通过仿真研究,结果表明所提方法能够有效地应用于多模态过程的故障检测。     

基于双子空间并行回归的化工过程质量相关故障检测方法

邻域保持嵌入(neighborhood preserving embedding, NPE)是一种常用的无监督学习方法,在故障检测领域得到了广泛应用。由于NPE提取的数据特征无法解释输入数据和输出数据之间的关系,因此在化工过程质量相关故障检测方面存在局限性。另外,NPE在提取数据流形结构时忽略了动态信息的表征。为了解决上述问题,基于NPE和慢特征分析(slow feature analysis,SFA)算法提出了一种名为双子空间并行回归(twin-space parallel regression,TSPR)的质量相关故障检测方法,该方法能够同时提取数据的流形特征和变化速度信息。首先,通过基于互信息的策略将原始过程空间分为序列相关子空间和序列无关子空间,以应对变量在时间序列相关性的差异。其次,在两个子空间中分别应用提出的邻域保持-慢特征嵌入算法(neighborhood preserving-slow feature embedding regression,NP-SFE)和NPE算法提取数据的有效结构特征,并同时用最小二乘回归在两个特征子空间中构建过程变量与质量变量的回归关系。随后,...     

基于稀疏局部嵌入深度卷积网络的冷水机组故障诊断方法

针对于冷水机组提出一种基于稀疏局部嵌入深度卷积网络（sparsely local embedding network,SLENet）的故障诊断方法。采用稀疏局部嵌入方法代替卷积核,对输入数据进行特征选择,避免了复杂的训练和调参过程。另外采用空间金字塔最大池化作为网络的输出层,减少了网络的输出维数和分类器的计算量。针对美国采暖、制冷与空调工程师学会提供的冷水机组的典型故障数据进行分类,结果表明,该方法相比深度卷积网络（CNN）和支持向量机（SVM）方法具有更高的故障诊断精度。     

基于局部相对概率密度kNN的多模态过程故障检测

针对多模态过程数据方差差异明显的空间分布特点,提出一种基于局部相对概率密度k近邻(LRPD-kNN)的多模态过程故障检测方法。首先对训练数据进行标准化,计算训练数据的局部相对概率密度估计值,消除多模态数据的方差差异。然后,对预处理后的数据建立kNN模型,计算统计量和控制限。对于测试数据,计算与训练数据局部相对概率密度的欧式距离平方和,通过比较统计量与控制限进行多模态故障检测。将该方法应用到数值例子和半导体生产过程,仿真结果表明,提出的算法效果要优于PCA、kNN和局部离群因子(LOF)方法,说明算法在方差差异较大的多模态过程故障检测方面具有很高的准确性。     

基于聚类选择k近邻的LLE算法及故障检测

针对化工过程在多种运行模式下多种流形结构具有不同最优近邻数问题,提出了基于聚类选择k近邻的局部线性嵌入(LLE)过程监控方法。使用LLE算法提取高维数据的低维子流形,通过局部线性回归得到高维数据空间到特征空间的映射矩阵;选择Silhouette指标作为聚类有效性指标评估嵌入空间样本信息的相似性,进而确定最优近邻数,根据映射矩阵构建故障监控统计量及其控制限,进行故障检测。最后将所提算法与其他经典算法应用于TE化工过程对比分析,验证了算法的有效性。     

基于TGNPE算法的间歇过程故障诊断

间歇过程数据是由批次、变量和时间构成的三维数据,数据内包含了丰富的对过程监控有用的全局和局部结构信息,如何充分提取间歇过程的特征信息是故障诊断的关键。传统方法处理三维数据都是将其展开成二维数据,展开过程必然会导致数据内在结构破坏,并且通常只考虑了数据的全局信息或者只考虑了数据的局部信息,这就不能充分提取过程的有用信息导致诊断效果欠佳。针对以上问题,提出了张量全局-局部邻域保持嵌入(TGNPE)算法,首先用张量分解的方法直接对三维数据进行建模,而不对数据进行展开,这就有效地保存了数据的内部结构,再用邻域保持嵌入算法充分提取数据局部结构信息的同时兼顾数据的全局信息,这就实现了对数据特征信息更加充分地提取,用TGNPE算法检测到故障后用贡献图法诊断出故障变量。通过青霉素发酵过程验证了本文提出的算法对间歇过程数据信息提取更加充分,更利于故障诊断。     

The chemical process monitoring method based on temporal extended orthogonal neighbourhood preserving embedding (TONPE)

Due to the high dimensionality, non-linearity and dynamic characteristics of chemical process data, a fault monitoring method based on temporal extension orthogonal neighbourhood preserving embedding (TONPE) is proposed. In order to make up for the shortcomings of the orthogonal neighbourhood preserving embedding (ONPE) algorithm, an information extraction strategy based on spatio-temporal structure is developed. First, a local neighbourhood set with spatio-temporal characteristics is established, and a weight matrix with spatio-temporal is reconstructed for each time point through the nearest neighbour in space and time. Then, a projection matrix with orthogonal constraints is obtained to establish a monitoring model. The TONPE algorithm can fully capture the local dynamic changes of high-dimensional data by extracting two different manifold features, so that the low-dimensional space has better performance capabilities. The simulation results of the continuous stirred tank reactor process and the Tennessee Eastman process verify the effectiveness of the TONPE algorithm in chemical process monitoring.     

Multimode Process Fault Detection Using Local Neighborhood Similarity Analysis

Traditional data driven fault detection methods assume unimodal distribution of process data so that they often perform not well in chemical process with multiple operating modes. In order to monitor the multimode chemical process effectively, this paper presents a novel fault detection method based on local neighborhood similarity analysis (LNSA). In the proposed method, prior process knowledge is not required and only the multimode normal operation data are used to construct a reference dataset. For online monitoring of process state, LNSA applies moving window technique to obtain a current snapshot data window. Then neighborhood searching technique is used to acquire the corresponding local neighborhood data window from the reference dataset. Similarity analysis between snapshot and neighborhood data windows is performed, which includes the calculation of principal component analysis (PCA) similarity factor and distance similarity factor. The PCA similarity factor is to capture the change of data direction while the distance similarity factor is used for monitoring the shift of data center position. Based on these similarity factors, two monitoring statistics are built for multimode process fault detection. Finally a simulated continuous stirred tank system is used to demonstrate the effectiveness of the proposed method. The simulation results show that LNSA can detect multimode process changes effectively and performs better than traditional fault detection methods.     

基于马氏距离局部离群因子方法的复杂化工过程故障检测

为了满足实际的生产需要,复杂化工过程往往包含多个运行模态。同时过程的复杂性使得同一模态下的数据分布是一种高斯分布和非高斯分布混合存在的不确定情况。数据的多模态分布特性以及同一模态下数据分布的不确定性使得传统多元统计监控(MSPM)方法很难给出令人满意的结果。针对这一问题,本文提出一种新的马氏距离局部离群因子(MDLOF)方法进行故障检测。通过利用马氏距离挖掘变量局部结构中包含的有用信息,并对样本的邻域密度加以考虑,形成对数据分布具有鲁棒性的基于密度的监控指标。最后通过数值仿真例子及Tennessee Eastman过程验证其有效性。     

Quality-relevant and process-relevant fault monitoring based on GNPER and the fault quantification index for industrial processes

Traditional quality-relevant fault monitoring methods focus on extracting the relationship between the global structural features of the process and quality variables but ignore the local features. At the same time, they lack the quantification of quality-relevant faults. To solve these problems, a quality-relevant and process-relevant fault monitoring method and its fault quantification index based on global neighbourhood preserving embedding regression (GNPER) are proposed. First, by seeking the direction of maximum global variance, the global objective function is applied to neighbourhood preserving embedding algorithm, and the global neighbourhood preserving embedding (GNPE) model is established to fully extract the global and local information of process data. Second, on the basis of GNPE, through the idea of projection regression, the GNPER model is established to obtain mapping relationships among process variables and quality variables, and quality-relevant subspace and process-relevant subspace are extracted, the corresponding subspace statistics are established for fault monitoring. Finally, the fault quantification index is established for the faults in the two subspaces, which can provide more meaningful fault monitoring results. A numerical example, the hot rolling mill and the Tennessee Eastman (TE) process, verify the superiority and accuracy of the proposed method.