稀疏核局部保持投影及其在非线性过程故障检测中的应用(英文)

Locality preserving projection (LPP) is a newly emerging fault detection method which can discover local manifold structure of a data set to be analyzed, but its linear assumption may lead to monitoring performance degradation for complicated nonlinear industrial processes. In this paper, an improved LPP method, referred to as sparse kernel locality preserving projection (SKLPP) is proposed for nonlinear process fault detection. Based on the LPP model, kernel trick is applied to construct nonlinear kernel model. Furthermore, for reducing the computational complexity of kernel model, feature samples selection technique is adopted to make the kernel LPP model sparse. Lastly, two monitoring statistics of SKLPP model are built to detect process faults. Simulations on a continuous stirred tank reactor (CSTR) system show that SKLPP is more effective than LPP in terms of fault detection performance.     

基于LPP-GNMF算法的化工过程故障监测方法

提出了基于LPP-GNMF算法的化工过程故障监测方法。非负矩阵分解（NMF）是一种新兴的降维算法,由于它在机理上具有潜变量的正向纯加性的特点,所以在对数据进行压缩时,可以基于数据内部的局部特征有效描述数据信息,相比于传统的多元统计过程监控方法如主元分析（PCA）等有更好的解释能力。然而NMF要求原始数据满足非负性的要求,实际的化工过程有时并不能保证,为放宽对原始数据的非负要求,引入了广义非负矩阵分解（GNMF）算法。其次,GNMF在分解的过程中没有考虑到样本间的局部结构和几何性质,可能存在不能准确处理数据的问题。针对这一问题,提出了将GNMF与LPP（局部投影保留）相结合的算法。将提出的LPP-GNMF算法应用于TE过程来评估其监测性能,并与PCA算法、NMF算法、SNMF算法进行比较,仿真模拟结果表明所提算法的可行性。     

多模态化工过程的全局监控策略

引言基于数据驱动的过程监控方法从20世纪80年代建立以来得到了蓬勃的发展,理论体系逐渐完善,功能模块不断丰富。特别是最近几年,来自人工智能,机器学习及信号处理领域的各种方法的引入为该领域注入了新的活力。目前,多数基于数据     

基于动态稀疏保局投影的故障检测方法

针对保局投影(locality preserving projections,LPP)没有考虑过程数据的全局信息和动态性的问题,提出一种新的基于动态稀疏保局投影(dynamic sparse locality preserving projections,DSLPP)的故障检测方法。该方法首先将原始数据矩阵扩展为考虑时序相关的增广矩阵,然后通过求解最优稀疏表示(sparse representation,SR)问题,得到能够表示数据全局稀疏重构关系的稀疏系数矩阵,并将其与LPP算法结合,构建综合考虑数据局部和全局关系的目标函数进行数据降维,最后分别在特征空间和残差空间构造T²统计量和Q统计量进行故障检测。TEP的仿真结果表明,与LPP方法相比,新方法能更迅速检测故障发生并降低过程监控漏报率。     

Adaptive local kernel-based learning for soft sensor modeling of nonlinear processes

Soft sensor techniques have been widely used to estimate product quality or other key indices which cannot be measured online by hardware sensors. Unfortunately, their estimation performance would deteriorate under certain circumstances, e.g., the change of the process characteristics, especially for global learning approaches. Meanwhile, local learning methods always only utilize input information to select relevant instances, which may lead to a waste of output information and inaccurate sample selection. To overcome these disadvantages, a new local modeling algorithm, adaptive local kernel-based learning scheme (ALKL) is proposed. First, a new similarity measurement using both input and output information is proposed and utilized in a supervised locality preserving projection technique to select relevant samples. Second, an adaptive weighted least squares support vector regression (AW-LSSVR) is employed to establish a local model and predict output indices for each query data. In AW-LSSVR, instead of using traditional cross-validation methods, the trade-off parameters are adjusted iteratively and the local model is updated recursively, which reduces the computational complexity a lot. The proposed ALKL is applied to an online crude oil endpoint prediction in an industrial fluidized catalytic cracking unit (FCCU) process. The experimental results demonstrate the high precision of our ALKL approach.     

Hierarchical monitoring of industrial processes for fault detection,fault grade evaluation,and fault diagnosis

Traditional process monitoring methods cannot evaluate and grade the degree of harm that faults can cause to an industrial process. Consequently, a process could be shut down inadvertently when harmless faults occur. To overcome such problems, we propose a hierarchical process monitoring method for fault detection, fault grade evaluation, and fault diagnosis. First, we propose fault grade classification principles for subdividing faults into three grades: harmless, mild, and severe, according to the harm the fault can cause to the process. Second, two‐level indices are constructed for fault detection and evaluation, with the first‐level indices used to detect the occurrence of faults while the second‐level indices are used to determine the fault grade. Finally, to identify the root cause of the fault, we propose a new online fault diagnosis method based on the square deviation magnitude. The effectiveness and advantages of the proposed methods are illustrated with an industrial case study. © 2017 American Institute of Chemical Engineers AIChE J, 63: 2781–2795, 2017     

一种基于改进KICA的非高斯过程故障检测方法

针对基于核独立元分析(kernel independent component analysis,KICA)的故障检测方法只考虑非高斯信息提取而忽略局部近邻结构保持的问题,提出基于改进KICA的过程故障检测方法。将KICA法中只考虑非高斯信息提取的负熵最大化准则转换为熵最小化准则,结合局部保持投影的相似局部近邻结构准则,提出了同时考虑非高斯信息提取和局部近邻结构保持的目标函数,通过粒子群优化算法进行全局寻优,然后建立监控统计量对过程进行监控。在Tennessee Eastman过程上的仿真结果说明,与基于KICA的故障检测方法相比,所提方法能够在保持数据集局部近邻结构的同时,提取非高斯信息,能够有效缩短故障检测的延迟时间,提高故障检测率。     

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。     

基于核Fisher判别分析方法的非线性统计过程监控与故障诊断

化工过程中大量的生产数据反应了生产过程的内在变化和系统的运行状况,基于数据驱动的统计方法可以有效地对生产过程进行监控。对于复杂的化工和生化过程,其过程变量之间的相关关系往往具有很强的非线性特性,传统的线性统计过程监控方法显得无能为力。本文提出了基于核Fisher判别分析的非线性统计过程监控方法,首先利用非线性核函数将数据从原始空间映射到高维空间,在高维空间中利用线性的Fisher判别分析方法提取数据最优的Fisher特征矢量和判别矢量来实现过程监控与故障诊断,能有效地捕获过程变量之间的非线性关系,通过对流化催化裂化（FCCU）过程的仿真表明该方法的有效性。     

基于非线性动态全局局部保留投影算法的化工过程故障检测

传统基于核映射的非线性故障检测方法的性能受核函数类型和核参数的调优影响较大,且实际工业环境中对过程变量的非线性阶数存在很多物理限制。针对这一问题,提出一种非线性动态全局局部保留投影(nonlinear dynamic global-local preserving projections,NDGLPP)的故障检测算法。该方法首先使用动态全局局部保留投影算法对数据矩阵进行降维;然后对降维后的矩阵建立二阶多项式映射提取非线性空间的相关特性;接着通过迭代这两个步骤以获得高阶非线性映射;最后,将所提方法应用于乙烯精馏过程和Tennessee Eastman(TE)过程仿真中,验证了检测方法的有效性和可行性。     

基于KSFDA-SVDD的非线性过程故障检测方法

慢特征分析(SFA)是一种无监督的线性学习算法,没有考虑过程数据的类别信息和非线性特征。针对此问题,提出一种基于核慢特征判别分析(KSFDA)和支持向量数据描述(SVDD)的非线性过程故障检测方法KSFDA-SVDD。该方法首先利用核技巧将数据从原始空间映射到高维空间,然后通过最大化正常工况数据和故障模式数据之间伪时间序列的时间变化同时最小化正常工况数据内部伪时间序列的时间变化计算判别矩阵,最后利用SVDD描述采用判别矩阵降维后的正常工况数据的分布域,构建监控统计量检测过程故障。在连续搅拌反应器(CSTR)过程上的仿真结果表明所提出方法的故障检测性能优于传统的KPCA方法。     

基于改进多模型FDA的间歇生产过程的故障诊断

1 INTRODUCTION In recent decades, batch processes have been a wide concern in the chemical fields because of their low-volume, high-value products and capabilities of easily tracking changing market situations. Therefore, it is necessary to monitor them in order to ensure safe, decrease the production costs and enhance the quality of products. Batch processes are characterized by the precise sequencing and automation of all stages in the sequence. They convert raw materials into products …     

基于两步矩阵投影的数据分类算法

提出了一种数据分类的两步矩阵投影算法.指出Crowe提出的矩阵投影算法在数据分类中存在由于投影矩阵不惟一,导致已测可校正数据分类不彻底的缺点.采用已测数据预分类的方法,对其进行了修正.在此基础上,将矩阵投影算法引入到了未测数据分类中,提出了基于矩阵投影算法的未测数据分类算法.新算法只需求解两个投影矩阵就可以实现所有数据分类.从而避免了常规方法在未测数据分类时,求解未测数据关联矩阵绝对线性无关列的计算,提高了计算效率.数学推导和算例验证了新算法的有效性.     

Localized Fisher discriminant analysis based complex chemical process monitoring

Complex chemical process is often corrupted with various types of faults and the fault‐free training data may not be available to build the normal operation model. Therefore, the supervised monitoring methods such as principal component analysis (PCA), partial least squares (PLS), and independent component analysis (ICA) are not applicable in such situations. On the other hand, the traditional unsupervised algorithms like Fisher discriminant analysis (FDA) may not take into account the multimodality within the abnormal data and thus their capability of fault detection and classification can be significantly degraded. In this study, a novel localized Fisher discriminant analysis (LFDA) based process monitoring approach is proposed to monitor the processes containing multiple types of steady‐state or dynamic faults. The stationary testing and Gaussian mixture model are integrated with LFDA to remove any nonstationarity and isolate the normal and multiple faulty clusters during the preprocessing steps. Then the localized between‐class and within‐class scatter mattress are computed for the generalized eigenvalue decomposition to extract the localized Fisher discriminant directions that can not only separate the normal and faulty data with maximized margin but also preserve the multimodality within the multiple faulty clusters. In this way, different types of process faults can be well classified using the discriminant function index. The proposed LFDA monitoring approach is applied to the Tennessee Eastman process and compared with the traditional FDA method. The monitoring results in three different test scenarios demonstrate the superiority of the LFDA approach in detecting and classifying multiple types of faults with high accuracy and sensitivity. © 2010 American Institute of Chemical Engineers AIChE J, 2011