复杂过程的可视化故障诊断方法

引言 随着自动化水平的提高,现代化的工程控制系统规模日趋大型化、复杂化,一旦系统发生故障会造成人员和财产的巨大损失,迫切需要提高自动化系统的可靠性、可维修性和安全性.人们一直在研究各种新的故障诊断技术和方法,来提高故障诊断的正确率,从而防患于未然.     

Using improved self-organizing map for fault diagnosis in chemical industry process

There are numerous fault diagnosis methods studied for complex chemical process, in which the effective methods for visualization of fault diagnosis are more challenging. In order to visualize the occurrence of the fault clearly, a novel fault diagnosis method which combines self-organizing map (SOM) with correlative component analysis (CCA) is proposed. Based on the sample data, CCA can extract fault classification information as much as possible, and then based on the identified correlative components, SOM can distinguish the various types of states on the output map. Further, the output map can be employed to monitor abnormal states by visualization method of SOM. A case study of the Tennessee Eastman (TE) process is employed to illustrate the fault diagnosis and monitoring performance of the proposed method. The results show that the SOM integrated with CCA method is efficient and capable for real-time monitoring and fault diagnosis in complex chemical process.     

High dimension feature extraction based visualized SOM fault diagnosis method and its application in p-xylene oxidation process☆

Purified terephthalic acid (PTA) is an important chemical raw material. P-xylene (PX) is transformed to terephthalic acid (TA) through oxidation process and TA is refined to produce PTA. The PX oxidation reaction is a complex process involving three-phase reaction of gas, liquid and solid. To monitor the process and to im-prove the product quality, as wel as to visualize the fault type clearly, a fault diagnosis method based on self-organizing map (SOM) and high dimensional feature extraction method, local tangent space alignment (LTSA), is proposed. In this method, LTSA can reduce the dimension and keep the topology information simultaneously, and SOM distinguishes various states on the output map. Monitoring results of PX oxidation reaction process in-dicate that the LTSA–SOM can wel detect and visualize the fault type.     

Fault detection, identification and diagnosis using CUSUM based PCA

In this paper, a cumulative sum based statistical monitoring scheme is used to monitor a particular set of the Tennessee Eastman Process (TEP) faults that could not be properly detected or diagnosed with other fault detection and diagnosis methodologies previously reported.T² and Q statistics based on the cumulative sums of all available measurements were successful in observing these three faults. For the purpose of fault isolation, contribution plots were found to be inadequate when similar variable responses are associated with different faults. Fault historical data is then used in combination with the proposed CUSUM based PCA model to unambiguously characterize the different fault signatures. The proposed CUSUM based PCA was successful in detecting, identifying and diagnosing both individual as well as simultaneous occurrences of these faults.     

Reconstruction in integrating fault spaces for fault identification with kernel independent component analysis

In the original fault identification methods, contribution plots are popular. However, it is not accurate because of the smearing effect. In addition, traditional contribution plots cannot be applied to nonlinear process because there seems no way to accurately calculate variable contributions. As a comparison, the reconstruction method is widely used in fault identification for finding the root causes of the fault. For fault detection and identification of actual industrial process with nonlinear and non-Gaussian features, a new reconstruction-based fault identification method with kernel independent component analysis (KICA) is developed in this article. The proposed method, reconstruction in integrating fault spaces (RIFSs), extends the classic reconstruction-based fault identification approach to KICA for the first time, and develops the reconstruction method from unidimensional faults to multidimensional ones for nonlinear cases. Furthermore, the number of reconstruction is effectively reduced on the basis of the integrating fault spaces (IFSs) which are composed of fault subspaces satisfying orthogonal to each other from the known fault set. In addition, fault magnitude, indicating the adjustment magnitude of a fault sample back to normal range, is used as index to identify faults, and it makes the fault identification problem become more straightforward than with the existing fault identification index, such as ratio (index I) or the reconstructed statistics (index II). Finally, the proposed method is applied to the fault detection and identification on cyanide leaching of gold, which shows its feasibility and efficiency for both sensor faults and complex process faults.     

Fault diagnosis of nonlinear processes using multiscale KPCA and multiscale KPLS

New approaches are proposed for nonlinear process monitoring and fault diagnosis based on kernel principal component analysis (KPCA) and kernel partial least analysis (KPLS) models at different scales, which are called multiscale KPCA (MSKPCA) and multiscale KPLS (MSKPLS). KPCA and KPLS are applied to these multiscale data to capture process variable correlations occurring at different scales. Main contribution of the paper is to propose nonlinear fault diagnosis methods based on multiscale contribution plots. In particular, the nonlinear scores of the variables at each scale are derived. These nonlinear scale contributions can be computed, which is very useful in diagnosing faults that occur mainly at a single scale. The proposed methods are applied to process monitoring of a continuous annealing process and fused magnesium furnace. Application results indicate that the proposed approach effectively captures the complex relations in the process and improves the diagnosis ability.     

Fault Diagnosis in Chemical Process Based on Self-organizing Map Integrated with Fisher Discriminant Analysis

Fault diagnosis and monitoring are very important for complex chemical process. There are numerous methods that have been studied in this field, in which the effective visualization method is still challenging. In order to get a better visualization effect, a novel fault diagnosis method which combines self-organizing map （SOM） with Fisher discriminant analysis （FDA） is proposed. FDA can reduce the dimension of the data in terms of maximizing the separability of the classes. After feature extraction by FDA, SOM can distinguish the different states on the output map clearly and it can also be employed to monitor abnormal states. Tennessee Eastman （TE） process is em- ployed to illustrate the fault diagnosis and monitoring performance of the proposed method. The result shows that the SOM integrated with FDA method is efficient and capable for real-time monitoring and fault diagnosis in complex chemical process.     

Integrated nonlinear model predictive fault tolerant control and multiple model based fault detection and diagnosis

In this paper, a new fault-tolerant control approach is presented for a class of nonlinear systems, which preserves system stability despite a time delay in fault detection. The faults are assumed to occur in the actuators and are modeled for the general form of affine nonlinear systems. A fault detection and diagnosis (FDD) block is designed based on the multiple model method. The bank of extended Kalman filters (EKF) is used to detect predefined actuator faults and to estimate the unknown parameters of actuator position. The estimated parameters are then used to correct the model of the faulty system and to reconfigure the controller. The reconfigurable controller is designed based on the stabilizing nonlinear model predictive control (NMPC) scheme. On the other hand, in the duration between fault occurrence and fault detection, because of the mismatch between the process and the model, the system states may go off the attraction region. The proposed method is based on designing multiple local controllers for individual predefined faults. Depending on the value of a system variable at the moment of fault detection, one of these controllers will operate. This leads to a stability region of a set of auxiliary equilibrium points (AEPs), which is larger than the attraction region. Moreover, a framework for preserving system stability is presented. Finally, a practical chemical process example is presented to illustrate the effectiveness of this method.     

Fault detection and diagnosis in a non-Gaussian process with modified kernel independent component regression

Quality-related fault detection and diagnosis are crucial in the data-driven process monitoring field. Most existing methods are based on principal component analysis (PCA) or partial least squares (PLS), which will miss high-order statistical information when the industrial process does not satisfy a Gaussian distribution. Meanwhile, the traditional contribution plot is difficult to directly apply to nonlinear processes in some cases due to its limitation of convergence. As such, a modified kernel independent component regression (MKICR) model, which considers high-order statistical information, is proposed for quality-related fault detection and faulty variable identification. First, the relationship between the independent components and quality variables is established by kernel independent component regression, and the correlation matrix is obtained. Then, the kernel independent components can be suitably divided into quality-related and quality-unrelated parts. Finally, an analysis of the contribution of each variable to the statistics based on Lagrange's mean value theorem is presented. In addition, a numerical case and the Tennessee Eastman process (TEP) demonstrate the efficacy and superiority of the proposed method.     

Fault identification for process monitoring using kernel principal component analysis

In this research, we develop a new fault identification method for kernel principal component analysis (kernel PCA). Although it has been proved that kernel PCA is superior to linear PCA for fault detection, the fault identification method theoretically derived from the kernel PCA has not been found anywhere. Using the gradient of kernel function, we define two new statistics which represent the contribution of each variable to the monitoring statistics, Hotelling's T²and squared prediction error (SPE) of kernel PCA, respectively. The proposed statistics which have similar concept to contributions in linear PCA are directly derived from the mathematical formulation of kernel PCA and thus they are straightforward to understand. The main contribution of this work is that we firstly suggest a fault identification method especially applicable to process monitoring using kernel PCA. To demonstrate the performance, the proposed method is applied to two simulated processes, one is a simple nonlinear process and the other is a non-isothermal CSTR process. The simulation results show that the proposed method effectively identifies the source of various types of faults.     

Multi-block gated recurrent unit network-based fault diagnosis method for multivariate nonlinear process

The increasing complexity of industrial processes brings new challenges to fault diagnosis tasks, and different types of faults have higher and higher requirements on the performance of fault classification models. This paper proposes a novel multivariate nonlinear temporal-related fault diagnosis method based on gated recurrent units (GRUs). First, to improve the performance of the model in local information extraction and global integration, high dimensional variables are divided into multiple sub-blocks according to the structure of chemical process units, and a new block normalization method is proposed to improve the performance of local feature extraction. Second, aiming at the slow drifting faults, the GRU network is adopted inside the sub-block to extract local sparse and nonlinear temporal features. By combining the variance features of variables after block normalization, the performance of the model on multiplicative faults will improve. Finally, aiming at the complex correlation between variables, a new recurrent matrix method is proposed to extract the time transform information inside each variable to improve the comprehensive performance of the model. Through a multi-level feature integration strategy, the model can be trained in parallel to improve the training speed. The proposed method shows good performance in the Tennessee Eastman process, and the extracted multi-class features allow the model to be trained end-to-end and simultaneously diagnose multiple types of faults.     

Cycle temporal algorithm-based multivariate statistical methods for fault diagnosis in chemical processes

Multivariate statistical process monitoring methods are often used in chemical process fault diagnosis. In this article,(I) the cycle temporal algorithm(CTA) combined with the dynamic kernel principal component analysis(DKPCA) and the multiway dynamic kernel principal component analysis(MDKPCA) fault detection algorithms are proposed, which are used for continuous and batch process fault detections,respectively. In addition,(II) a fault variable identification model based on reconstructed-based ...     

ISOMAP-LDA方法用于化工过程故障诊断

针对化工连续生产过程的时序性及非线性等特征,提出一种新的基于数据驱动的化工过程故障诊断方法：ISOMAP-LDA。首先实行流形学习算法ISOMAP,在保持量测数据几何结构特性下完成非线性降维,然后基于提取的嵌入变量张成的低维空间,选用线性判别分析（LDA）构造故障模式类的判别函数,负责各采样个体故障类型的判定。将该方法用于仿真化工Tennessee Eastman 过程的故障诊断,结果表明,ISOMAP-LDA方法不仅拥有较高的故障诊断能力,而且取得采样在低维空间的可视化表示。     

Decentralized fault diagnosis using multiblock kernel independent component analysis

In this paper, a multiblock kernel independent component analysis (MBKICA) algorithm is proposed. Then a new fault diagnosis approach based on MBKICA is proposed to monitor large-scale processes. MBKICA has superior fault diagnosis ability since variables are grouped and the non-Gaussianity is considered compared to standard kernel methods. The proposed method is applied to fault detection and diagnosis in the continuous annealing process. The proposed decentralized nonlinear approach effectively captures the nonlinear relationship and non-Gaussianity in the block process variables, and shows superior fault diagnosis ability compared to other methods.     

Enhanced plant fault diagnosis based on the characterization of transient stages

This paper introduces a data-based fault diagnosis system that includes an enhanced characterization of faults during transient stages. First, data under abnormal operating conditions (AOC) is projected onto a reference PCA model constructed with data under normal operating conditions (NOC). T² and Q-statistic measures of this first PCA model are both used to detect the fault and to estimate the duration and delay of its transient evolution. After a dimensionality reduction, a second NOC PCA model is used to process data before diagnosing the faults by standard classification methods such as Artificial Neural Networks (ANN) or Support Vector Machines (SVM). A quantitative validation of the procedure has been carried out using simulated on-line data sets of the Tennessee Eastman Process (TEP). Results indicate that the incorporation of transient data in models improves the overall diagnosis performance, regardless of the particular choice between the statistical methods or the classification methods.     

Fault Detection and Diagnosis in a Sour Gas Absorption Column Using Neural Networks

Process fault detection and diagnosis is an important problem in plant control at the supervisory level. It is the central component of abnormal event management which has attracted a lot of attention recently. In this study, the use of artificial neural networks (ANN) for fault detection is explored. An ANN can represent nonlinear and complex relations between its inputs (sensor measurements) and outputs (faults). As a test case, absorption of CO₂ gas in monoethanolamine (MEA) by a pilot plant called “automatic absorption and stripping pilot plant” is studied. For detecting and diagnosis of faults, variations in feed rate, feed composition, liquid absorber rate and composition are imposed onto the plant. The faults in this process influence variables such as the composition of absorbed gas (CO₂) and temperature and pressure drop of the column. The CO₂ concentration in the product should not exceed a certain limit. By selecting a proper architecture for the network (5‐9‐10), it is possible to detect the faults accurately. The network is trained using the back propagation method. The developed fault diagnosis algorithm is tested using data that has not been seen by the network.     

Fault detection and diagnosis with parametric uncertainty using generalized polynomial chaos

This paper presents a new methodology to identify and diagnose intermittent stochastic faults occurring in a process. A generalized polynomial chaos (gPC) expansion representing the stochastic inputs is employed in combination with the nonlinear mechanistic model of the process to calculate the resulting statistical distribution of measured variables that are used for fault detection and classification. A Galerkin projection based stochastic finite difference analysis is utilized to transform the stochastic mechanistic equation into a coupled deterministic system of equations which is solved numerically to obtain the gPC expansion coefficients. To detect and recognize faults, the probability density functions (PDFs) and joint confidence regions (JCRs) of the measured variables to be used for fault detection are obtained by substituting samples from a random space into the gPC expansions. The method is applied to a two dimensional heat transfer problem with faults consisting of stochastic changes combined with step change variations in the thermal diffusivity and in a boundary condition. The proposed methodology is compared with a Monte Carlo (MC) simulations based approach to illustrate its advantages in terms of computational efficiency as well as accuracy.     

A signed directed graph-based systematic framework for steady-state malfunction diagnosis inside control loops

Although signed directed graphs (SDG) have been widely used for modeling control loops, due to lack of adequate understanding of SDG-based steady-state process modeling, special and cumbersome methods are used to analyze control loops. In this paper, we discuss a unified SDG model for control loops, in which both disturbances (sensor bias, etc.) as well as structural faults (sensor failure, controller failure, etc.) can be easily modeled under steady-state conditions. Various fault scenarios such as external disturbances, sensor bias, controller failure, etc. have been thoroughly analyzed. A new algorithm for steady-state fault diagnosis using the SDG model for the steady-state system, that uses a combination of forward- and backward-reasoning, is proposed. Three case studies are presented to show the utility of the steady-state SDG model for fault diagnosis. A tank-level control system is used as the first case study. The second case study deals with fault diagnosis of a multi-stream-controlled CSTR. The third case study deals with fault/failure diagnosis in a process flowsheet containing a CSTR with one control loop and a flash vaporizer with three control loops.     

Enhanced statistical analysis of nonlinear processes using KPCA, KICA and SVM

In this paper, some drawbacks of original kernel independent component analysis (KICA) and support vector machine (SVM) algorithms are analyzed for the purpose of multivariate statistical process monitoring (MSPM). When the measured variables follow non-Gaussian distribution, KICA provides more meaningful knowledge by extracting higher-order statistics compared with PCA and kernel principal component analysis (KPCA). However, in real industrial processes, process variables are complex and are not absolutely Gaussian or non-Gaussian distributed. Any single technique is not sufficient to extract the hidden information. Hence, both KICA (non-Gaussion part) and KPCA (Gaussion part) are used for fault detection in this paper, which combine the advantages of KPCA and KICA to develop a nonlinear dynamic approach to detect fault online compared to other nonlinear approaches. Because SVM is available for classifying faults, it is used to diagnose fault in this paper.For above mentioned kernel methods, the calculation of eigenvectors and support vectors will be time consuming when the sample number becomes large. Hence, some dissimilar data are analyzed in the input and feature space.The proposed approach is applied to the fault detection and diagnosis in the Tennessee Eastman process. Application of the proposed approach indicates that proposed method effectively captures the nonlinear dynamics in the process variables.     

基于加权统计特征KICA的故障检测与诊断方法

针对核独立元分析（kernel independent component analysis, KICA）在非线性动态过程中对微小故障检测率低的问题,提出一种基于加权统计特征KICA（weighted statistical feature KICA, WSFKICA）的故障检测与诊断方法。首先,利用KICA从原始数据中捕获独立元数据和残差数据;然后,通过加权统计特征和滑动窗口获取改进统计特征数据集,并由此数据集构建统计量进行故障检测;最后,利用基于变量贡献图的方法进行过程故障诊断。与传统KICA统计量相比,所提方法的统计量对非线性动态过程中的微小故障具有更高的故障检测性能。应用该方法对一个数值例子和田纳西-伊斯曼（Tennessee-Eastman, TE）过程进行仿真测试,仿真结果显示出所提方法相对于独立元分析（ICA）、KICA、核主成分分析（kernel principal component analysis, KPCA）和统计局部核主成分分析（statistical local kernel principal component analysis, SLKPCA）检测的优势。