基于并发改进偏最小二乘的质量相关和过程相关的故障诊断

并发潜结构投影(CPLS)与传统贡献图法是多元统计过程监控中常用的故障检测与诊断方法.过程监控通常要求监测的时效性与诊断的准确性,然而,由于CPLS计算复杂以及传统贡献图诊断结果易受初始贡献较大的变量影响,因此它们反馈的监控结果可能并不准确.针对上述问题分别提出一种并发改进偏最小二乘(CMPLS)方法和新的相对贡献图法(NRC).首先,CMPLS将输入和输出数据同时投影到与过程相关或质量相关的多个子空间,在相应子空间分别构造适用于各种故障报警的监测指标进行过程监测;然后,结合所提出的NRC进行故障识别.所提方法对过程故障实现全面监测的同时避免了过多的迭代过程,并消除了过程变量中对检测指标初始贡献较大变量的影响.最后利用数值仿真和田纳西伊士曼过程验证了所提方法的有效性.     

基于鲁棒规范变量分析的故障诊断方法

针对工业过程的建模数据中含有离群点的情况,提出一种基于鲁棒规范变量分析(CVA)的故障诊断方法.该方法使用相关系数的鲁棒估计代替传统的相关系数,通过基于粒子群算法的投影寻踪技术计算最大化鲁棒相关系数的规范变量,从而建立统计模型并监控统计量检测过程的变化.连续搅拌反应器(CSTR)系统的仿真结果说明,鲁棒规范变量分析方法能在含离群点数据的基础上建立准确的统计模型,比规范变量分析更有效地监控过程变化.     

基于动态全潜结构投影的热连轧厚度监控

本文利用带钢热连轧生产过程的数据,采用动态全潜结构投影算法(T-PLS),建立了带钢厚度的动态模型.该模型对于厚度有良好的预测精度.利用动态T-PLS的优点,把过程变量空间分解为4个正交子空间.在不同的子空间中,可以对带钢厚度有关的故障进行监测.通过热连轧机3个典型故障的检测,充分验证了动态T-PLS在过程质量监控中的优良性能,加强了带钢热连轧过程的监控.     

基于多尺度核主元分析的非线性过程监控方法研究

针对化工过程数据的多尺度性和非线性特性,提出了一种多尺度核主元分析方法(MSKPCA)监控过程的运行状态。使用小波变换在不同尺度下分解测量信号．然后借助于核函数对分解后的数据进行非线性变换,在变换后的线性空间中用主元分析(PCA)提取过程数据的主要特征,构造监控统计量T2和Q来检测故障。在此基础上,提出了一种贡献图方法．计算过程变量对故障的贡献量,用于故障变量的分离。在TE过程上的监控结果表明,MSKPCA可以比PCA和动态PCA更迅速地检测到过程故障,贡献图方法能够正确地分离故障变量。     

基于NPE算法的环网柜故障检测方法研究

为保障供电系统的安全、可靠运行,对电网环网柜在线故障检测问题进行研究,提出了一种新的基于数据局部特征的环网柜数据建模和在线监控方法。利用邻域保持嵌入(NPE)算法局部特征提取的策略,基于环网柜的多个测量变量信息以及环境变量信息,获取实时数据特征,构建了基于数据特征的环网柜故障检测模型。将构建的NPE模型应用于实际环网柜在线检测,并将原始数据空间划分为不相关的特征空间和数据残差空间。针对这两个空间,分别构造Hotelling T~2和预测误差平方和(SPE)的监控统计量,并基于这两个监控统计量,实现了环网柜的在线实时监控和故障报警。将该故障检测方法应用于实际环网柜的的监控案例研究中,试验结果证明了该方法在环网柜故障检测方面的有效性。通过该数据监控模型,改善了环网柜故障检测的效果,为降低风险、提高环网柜的安全稳定和运行品质提供技术保障。     

基于统计差分LPP的多模态间歇过程故障检测

针对工业过程数据存在的非高斯和多模态特性,提出一种基于统计差分LPP的多模态间歇过程故障检测方法。首先将统计模量分析的方法应用到间歇过程训练数据集中,计算统计过程变量的均值和方差,将不等长的批次变成等长的统计量,保证统计模量近似服从高斯分布;然后运用差分算法使多模态变为单模态,最后运用LPP算法进行降维和特征提取,计算样本的T2统计量,并利用核密度估计确定控制限。对于新来的测试样本数据统计差分处理后,向LPP模型上进行投影,计算新数据的T2统计量并与控制限比较进行故障检测。最后通过半导体过程数据的仿真结果表明,该算法的故障检测效果最好,验证了所提方法的有效性。     

基于动态核独立分量分析的高含硫天然气净化过程异常检测与诊断

目前高含硫天然气净化过程存在多参数动态相关的特性,导致基于静态多元统计过程监控方法对于异常状态检测效果较差。提出一种考虑参数时序自相关性的动态核独立分量分析(DKICA)异常检测与诊断方法。首先,引入自回归(AR)模型,通过参数辨识确定模型阶次,描述监控过程的时序自相关性;然后,将原始变量投影到核独立元空间,通过监控独立元对应的T²和SPE统计量是否超出正常状态设定的控制限,实现异常检测;最后计算所述T²统计量对原始变量的一阶偏导数,绘制贡献图实现异常诊断。以某高含硫天然气净化厂采集的数据进行分析,结果表明基于DKICA高含硫天然气净化过程异常检测精度要优于静态独立分量分析所得的检测精度。     

动态内全潜结构投影的空间扩展故障检测方法

动态内偏最小二乘(DiPLS)方法是基于数据驱动的潜结构投影的动态扩展算法, 用于动态特征提取和关键 性能指标预测. 在大型装备系统中, 传感器采集的当前时刻样本受历史样本的影响且可能包含较大噪声. 在动态特 征提取中, 因DiPLS算法未按降序提取主成分, 导致残差空间仍存在较大变异, 动态和静态信息难以有效分离, 影响 故障检测性能. 为此, 本文提出了一种基于动态内全潜结构投影的故障检测方法(DiTPLS). 首先, 使用动态内偏最小 二乘方法和向量自回归模型建立动态模型并检测故障, 用于捕捉质量相关动态信息; 基于结构化动态主成分分析 算法建立一种改进的动态潜在变量模型, 用于残差分解, 提取质量无关的动态信息和静态信息, 并构造合适的统计 量进行故障检测. 数值仿真和田纳西–伊斯曼过程实验验证了DiTPLS算法的有效性.     

基于规范变量残差的化工过程微小故障检测与诊断

微小故障因其幅值低而易被噪声和过程扰动所掩盖,并且会随时间慢慢演变成过程中的严重故障.因此,微小故障的检测和诊断变得越来越重要.为了更有效地监测和诊断微小故障,提出了基于规范变量残差的化工过程微小故障检测和诊断方法.首先,对Hankel矩阵执行奇异值分解来获得主元和残差空间并根据过去和未来数据的差异,求得两个不同的规范变量残差d_1, d_2.其次,考虑数据的时间序列特性,提出了基于规范变量残差的两个加权平均统计量W_(D1), W_(D2)及其控制限,进行故障检测;然后,计算出各个统计量的归一化贡献并绘制二维贡献图,进行故障诊断.最后,在连续搅拌釜式反应器(CSTR)过程中进行两种微小故障的应用研究.结果表明,与传统的统计量T~2,Q以及规范变量差异分析(CVDA)中统计量D相比,基于规范变量残差的加权平均统计量W_(D1), W_(D2)不仅能够及时检测到微小故障,而且在故障检测率和诊断率方面,均有不同程度的提高.     

基于局部保持投影–加权k近邻规则的 多模态间歇过程故障检测策略

针对多模态间歇过程故障检测问题,本文提出一种基于局部保持投影–加权k近邻规则(LPP--Wk NN)的故障检测策略.首先,应用局部保持投影(LPP)方法将原始数据投影到低维主元子空间;接下来,在主元子空间中,应用样本第k近邻的局部近邻集确定每个样本的权重并计算权重统计量Dw;最后,应用核密度估计方法确定Dw控制限并进行故障检测.本文方法应用LPP对过程数据进行维数约减,既能够降低训练过程中离群点对模型的影响,又能够降低在线故障检测的计算复杂度.同时,加权k近邻规则(Wk NN)方法通过引入权重规则能够使得过程故障检测统计量分布具有单模态结构.相比传统的k NN统计量,本文引入的权重统计量具有更高的故障检测性能.通过数值例子和半导体蚀刻过程的仿真实验,并与主元分析(PCA), k NN, Wk NN, LPP--k NN等方法进行比较,实验结果验证了本文方法的有效性.     

Performance monitoring of processes with multiple operating modes through multiple PLS models

Many industrial processes possess multiple operating modes in virtue of different manufacturing strategies or varying feedstock. Direct application of many of the current multivariate statistical process monitoring (MSPM) techniques such as PCA (principal component analysis) and PLS (projection to latent structures) to such a process tends to produce inferior performance. This can most be attributed to the adopted assumption by most MSPM methodologies of only one nominal operating region for the underlying process. It is therefore reasonable to develop separate models for different operating modes. In this paper, based on metrics in the form of principal angles to measure the similarities of any two models, a multiple PLS model based process monitoring scheme is proposed. Popular multivariate statistics such as SPE (squared prediction error) and T² can be incorporated in this framework straightforwardly. The proposed technique is assessed through application to the monitoring of an industrial pyrolysis furnace.     

A probabilistic principal component analysis-based approach in process monitoring and fault diagnosis with application in wastewater treatment plant

Probabilistic principal component analysis (PPCA) based approaches have been widely used in the field of process monitoring. However, the traditional PPCA approach is still limited to linear dimensionality reduction. Although the nonlinear projection model of PPCA can be obtained by Gaussian process mapping, the model still lacks robustness and is susceptible to process noise. Therefore, this paper proposes a new nonlinear process monitoring and fault diagnosis approach based on the Bayesian Gaussian latent variable model (Bay-GPLVM). Bay-GPLVM can obtain the posterior distribution rather than point estimation for latent variables, so the model is more robust. Two monitoring statistics corresponding to latent space and residual space are constructed for PM-FD purpose. Further, the cause of fault is analyzed by calculating the gradient value of the variable at the fault point. Compared with several PPCA-based monitoring approaches in theory and practical application, the Bay-GPLVM-based process monitoring approach can better deal with nonlinear processes and show high efficiency in process monitoring.     

基于精简并发潜结构映射的竖炉焙烧过程综合故障诊断

竖炉焙烧过程因运行条件异常变化或操作不当会造成上火、冒火、过还原和欠还原等运行故障.这些故障直接影响过程运行安全和产品质量（比如,磁选管回收率）,但难以采用基于模型和基于知识的方法建模故障与产品质量的关系,以及诊断故障变量.针对上述问题,本文提出数据驱动的基于并发潜结构映射（Concurrent projection to latent structures,CPLS）的竖炉焙烧过程综合故障诊断方法.首先,将并发潜结构映射分解的过程变量共有子空间与残差空间精简合并来建立磁选管回收率相关的过程变化空间,提出基于精简并发潜结构映射模型的竖炉焙烧过程综合监控方法;接下来,定义相应的重构贡献图并与竖炉焙烧过程相结合,提出CPLS精简重构贡献方法用于竖炉焙烧过程故障变量诊断;最后,利用竖炉焙烧过程半实物仿真平台采集的数据进行实验研究,结果表明所提方法不仅可以诊断出质量相关的故障,而且可诊断出回路设定值之外的故障变量.     

Multiblock PLS-based localized process diagnosis

In this paper, we discuss a new fault detection and identification approach based on a multiblock partial least squares (MBPLS) method to monitor a complex chemical process and to model a key process quality variable simultaneously. In multivariate statistical process monitoring using MBPLS, four kinds of monitoring statistics are discussed. In particular, new definitions of the block and variable contributions to T² and Q statistics are proposed and derived in order to identify faults. Also, the relative contribution, which is the ratio of the contribution to the corresponding upper control limit, is considered to find process variables or blocks responsible for faults. As an application study, a large wastewater treatment process in a steel mill plant is monitored and the effluent chemical oxygen demand, which indicates the current process performance, is modeled based on the proposed MBPLS-based fault detection and diagnosis method.     

Online process operating performance assessment and nonoptimal cause identification for industrial processes

Although industrial processes are usually operated at the optimal point in the early stage of the production, the operating performance may deteriorate with time due to process disturbances. In order to pursue optimal comprehensive economic benefit (CEB), online process operating performance assessment on optimality has become a key issue. However, a little work has been published in this research area. In this paper, a new online operating performance assessment and nonoptimal cause identification method for industrial process are proposed. The contributions of this paper can be summarized as follows: a novel performance-similarity-based online operating performance assessment method is proposed; total projection to latent structures (T-PLS) is applied to the area of process performance assessment for the first time; the online assessment results include not only the deterministic performance grades, but also the performance grade conversions which were not covered in the existing assessment method; when the assessment result is nonoptimal, a novel automatic nonoptimal cause identification strategy is developed based on variable contributions, which is meaningful for guiding the further production adjustment. Finally, the feasibility and efficiency of the proposed method are illustrated with a case of gold hydrometallurgical process.     

Reconstruction-based multivariate contribution analysis for fault isolation: A branch and bound approach

Identification of faulty variables is an important component of multivariate statistical process monitoring (MSPM); it provides crucial information for further analysis of the root cause of the detected fault. The main challenge is the large number of combinations of process variables under consideration, usually resulting in a combinatorial optimization problem. This paper develops a generic reconstruction based multivariate contribution analysis (RBMCA) framework to identify the variables that are the most responsible for the fault. A branch and bound (BAB) algorithm is proposed to efficiently solve the combinatorial optimization problem. The formulation of the RBMCA does not depend on a specific model, which allows it to be applicable to any MSPM model. We demonstrate the application of the RBMCA to a specific model: the mixture of probabilistic principal component analysis (PPCA mixture) model. Finally, we illustrate the effectiveness and computational efficiency of the proposed methodology through a numerical example and the benchmark simulation of the Tennessee Eastman process.     

Latent Variable Regression for Supervised Modeling and Monitoring

A latent variable regression algorithm with a regularization term(r LVR) is proposed in this paper to extract latent relations between process data X and quality data Y. In rLVR,the prediction error between X and Y is minimized, which is proved to be equivalent to maximizing the projection of quality variables in the latent space. The geometric properties and model relations of rLVR are analyzed, and the geometric and theoretical relations among r LVR, partial least squares, and canonical correlation analysis are also presented. The rLVR-based monitoring framework is developed to monitor process-relevant and quality-relevant variations simultaneously. The prediction and monitoring effectiveness of rLVR algorithm is demonstrated through both numerical simulations and the Tennessee Eastman(TE) process.     

Contribution rate plot for nonlinear quality-related fault diagnosis with application to the hot strip mill process

In this paper, a nonlinear fault diagnosis scheme is established for the hot strip mill process (HSMP). In HSMP, the faults affecting quality index are denoted as quality-related faults, which should be taken care as soon as possible. Projection to latent structures (PLS) is a basic model for quality-related fault detection in linear processes. In the presented work, a total kernel PLS (T-KPLS) model is utilized for modeling and monitoring HSMP, which is a typical nonlinear process. However, diagnosis tools have not been developed aiming at the nonlinear case based on T-KPLS model. Motivated by the successful use of contribution plot for the linear case, a contribution rate plot is proposed to extend contribution plots to the nonlinear case. In the end of this paper, the proposed method is applied to the hot strip mill process effectively.     

Informative PLS score-loading plots for process understanding and monitoring

Principal component regression (PCR) based on principal component analysis (PCA) and partial least squares regression (PLSR) are well known projection methods for analysis of multivariate data. They result in scores and loadings that may be visualized in a score-loading plot (biplot) and used for process monitoring. The difficulty with this is that often more than two principal or PLS components have to be used, resulting in a need to monitor more than one such plot. However, it has recently been shown that for a scalar response variable all PLSR/PCR models can be compressed into equivalent PLSR models with two components only. After a summary of the underlying theory, the present paper shows how such two-component PLS (2PLS) models can be utilized in informative score-loading biplots for process understanding and monitoring. The possible utilization of known projection model monitoring statistics and variable contribution plots is also discussed, and a new method for visualization of contributions directly in the biplot is presented. An industrial data example is included.