A fuzzy-logic based fault diagnosis strategy for process control loops

By considering the fault propagation behaviors in process systems with control loops, a fuzzy-logic based fault diagnosis strategy has been developed in the present work. The proposed fault diagnosis methods can be implemented in two stages. In the off-line preparation stage, the fault origins of a system hazard are identified by determining the minimal cut sets of the corresponding fault tree. The fault propagation patterns in a feedback loop are obtained on the basis of system digraph. The occurrence order of observable symptoms caused by each fault origin is derived accordingly and then encoded into a set of IF-THEN diagnosis rules. In the next on-line diagnosis stage, the occurrence indices of the top event and also the fault origins are computed in a fuzzy inference system based on real-time measurement data. Simulation studies have been carried out to demonstrate the feasibility of the proposed approach.     

Fuzzy diagnosis method for control systems with coupled feed forward and feedback loops

By considering the fault propagation behaviors in control systems with coupled feed forward and feedback loops, a fuzzy-logic-based fault diagnosis strategy has been developed in the present work. The proposed methods can be implemented in two stages. In the off-line preparation stage, the root causes of a system hazard are identified by determining the minimal cut sets of the corresponding fault tree. The occurrence order of observable disturbances caused by each fault origin is derived from the system digraph. All possible patterns of the on-line symptoms and their evolution sequences can then be deduced accordingly. These sequences are used as the basis for constructing a two-layer fuzzy inference system. In the next on-line implementation stage, the occurrence indices of the root causes are computed with the IF-THEN rules embedded in the inference engine using the real-time measurement data. Numerical simulation studies have been carried out to demonstrate the feasibility of the proposed approach.     

Fault diagnosis with automata generated languages

A SDG-based simulation procedure is proposed in this study to qualitatively predict the effects of one or more fault propagating in a given process system. These predicted state evolution behaviors are characterized with an automaton model. By selecting a set of on-line sensors, the corresponding diagnoser can be constructed and the diagnosability of every fault origin can be determined accordingly by inspection. Furthermore, it is also possible to define a formal diagnostic language on the basis of this diagnoser. Every string (word) in the language is then encoded into an IF-THEN rule and, consequently, a comprehensive fuzzy inference system can be synthesized for on-line diagnosis. The language generation steps are illustrated with a series of simple examples in this paper. The feasibility and effectiveness of this approach has been tested in extensive numerical simulation studies.     

Statistical root cause analysis of novel faults based on digraph models

This paper investigates the challenging problem of diagnosing novel faults whose fault mechanisms and relevant historical data are not available. Most existing fault diagnosis systems are incapable to explain root causes for unanticipated, novel faults, because they rely on either models or historical data of known faulty conditions. To address this issue we propose a new framework for novel fault diagnosis, which integrates causal reasoning on signed digraph models with multivariate statistical process monitoring. The prerequisites for our approach include historical data of normal process behavior and qualitative cause–effect relationships that can be derived from process flow diagrams. In this new approach, a set of candidate root nodes is identified first via qualitative reasoning on signed digraph; then quantitative local consistency tests are implemented for each candidate based on multivariate statistical process monitoring techniques; finally, using the resulting multiple local residuals, diagnosis is performed based on the exoneration principle. The cause–effect relationships in the digraph enable automatic variable selection and the local residual interpretations for statistical monitoring. The effectiveness of this new approach is demonstrated using numerical examples based on the Tennessee Eastman process data.     

A novel dynamic bayesian network‐based networked process monitoring approach for fault detection,propagation identification,and root cause diagnosis

A novel networked process monitoring, fault propagation identification, and root cause diagnosis approach is developed in this study. First, process network structure is determined from prior process knowledge and analysis. The network model parameters including the conditional probability density functions of different nodes are then estimated from process operating data to characterize the causal relationships among the monitored variables. Subsequently, the Bayesian inference‐based abnormality likelihood index is proposed to detect abnormal events in chemical processes. After the process fault is detected, the novel dynamic Bayesian probability and contribution indices are further developed from the transitional probabilities of monitored variables to identify the major faulty effect variables with significant upsets. With the dynamic Bayesian contribution index, the statistical inference rules are, thus, designed to search for the fault propagation pathways from the downstream backwards to the upstream process. In this way, the ending nodes in the identified propagation pathways can be captured as the root cause variables of process faults. Meanwhile, the identified fault propagation sequence provides an in‐depth understanding as to the interactive effects of faults throughout the processes. The proposed approach is demonstrated using the illustrative continuous stirred tank reactor system and the Tennessee Eastman chemical process with the fault propagation identification results compared against those of the transfer entropy‐based monitoring method. The results show that the novel networked process monitoring and diagnosis approach can accurately detect abnormal events, identify the fault propagation pathways, and diagnose the root cause variables. © 2013 American Institute of Chemical Engineers AIChE J, 59: 2348–2365, 2013     

A fault magnitude based strategy for effective fault classification

A common approach in fault diagnosis is monitoring the deviations of measured variables from the values at normal operations to identify the root causes of faults. When the number of conceivable faults is larger than that of predictive variables, conventional approaches can yield ambiguous diagnosis results including multiple fault candidates. To address the issue, this work proposes a fault magnitude based strategy. Signed digraph is first used to identify qualitative relationships between process variables and faults. Empirical models for predicting process variables under assumed faults are then constructed with support vector regression (SVR). Fault magnitude data are projected onto principal components subspace, and the mapping from scores to fault magnitudes is learned via SVR. This model can estimate fault magnitudes and discriminate a true fault among multiple candidates when different fault magnitudes yield distinguishable responses in the monitored variables. The efficacy of the proposed approach is illustrated on an actuator benchmark problem.     

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.     

OBJECT-BASED AUTOMATED FAULT DIAGNOSIS

The implementation of a model-based fault diagnosis methodology is presented. An object-based representation is used and shown to allow for easy access to and maintenance of the process model. The diagnostic methodology employed fits easily into this object-based approach. The methodology also offers other advantages including explicit use of simple model equations, non-Boolean reasoning, sensitivity weighting of evidence, and the possibility of multiple fault explanations. Additionally, methods of improving fault discrimination are discussed such as the cautious use of heuristics and the single fault hypothesis. Testing of the diagnostic system on a process simulation has shown it to be successful at locating various fault conditions.     

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 ...     

基于建筑瓷砖烧成缺陷专家系统的模糊推理原型分析

针对建筑瓷砖烧成缺陷诊断问题及其特点,提出了一种基于诊断知识的模糊描述和模糊推理方法,阐述了建筑瓷砖烧成缺陷诊断专家系统中前向推理,后向推理及正反向混合推理模糊断言可信度的计算方法,并给出了相应的实例。     

Topological preservation techniques for nonlinear process monitoring

This work proposes a novel approach for the offline development and online implementation of data-driven process monitoring (PM) using topological preservation techniques, specifically self-organizing maps (SOM). Previous topological preservation PM applications have been restricted due to the lack of monitoring and diagnosis tools. In the proposed approach, the capabilities of SOM are further extended so that all aspects of PM can be performed in a single environment. First for fault detection, using SOM's vector quantization abilities, an SOM-based Gaussian mixture model (GMM) is proposed to define the normal region. For identification, an SOM-based contribution plot is proposed to identify the variables most responsible for the fault. This is done by analyzing the residual of the faulty point and an SOM model of the normal region used in fault detection. The data points are projected on the model by locating the best matching unit (BMU) of the point. Finally, for fault diagnosis a procedure is formulated involving the concept of multiple self-organizing maps (MSOM), creating a map for each fault. This allows the ability to include new faults without directly affecting previously characterized faults. A Tennessee Eastman Process (TEP) application is performed on dynamic faults such as random variations, sticky valves and a slow drift in kinetics. Previous studies of the TEP have considered particular feed-step-change faults. Results indicate an excellent performance when compared to linear and nonlinear distance preservation techniques and standard nonlinear SOM approaches in fault diagnosis and identification.     

GMM and optimal principal components-based Bayesian method for multimode fault diagnosis

Principal component analysis (PCA) serves as the most fundamental technique in multivariate statistical process monitoring. However, other than determining contributions to a fault from each variable based on the pre-selected major principal components (PCs), the PCA-based fault diagnosis with an optimal selection of PCs is seldom investigated. This paper presents a novel Gaussian mixture model (GMM) and optimal principal components (OPCs)-based Bayesian method for efficient multimode fault diagnosis. First, the GMM and Bayesian inference is utilized to identify the operating mode, and then local PCA model is established in each mode. Second, given that the various principal components (PCs) may contain distinct fault signatures, the behavior of each PC in local PCA is examined and the OPCs are selected through stochastic optimization algorithm. Based on the OPCs, a Bayesian diagnosis system is then formulated to identify the fault statuses in a probability manner. Performance of GMM–OPC Bayesian diagnosis is examined through a numerical example and the Tennessee Eastman challenge process. The efficiency and feasibility are demonstrated.     

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.     

Fast and robust fault diagnosis for a class of nonlinear systems: detectability analysis

In recent years, robust fault diagnosis of nonlinear systems has received much more attention due to the universal existence of nonlinearities and model uncertainties in practice. By introducing a new adaptive law and sliding mode observers with boundary layer control into Polycarpou's online approximator, we propose a fast and robust fault diagnosis strategy for a class of nonlinear systems in this article. The robustness and stability are proved theoretically by the Lyapunov method and the detectability conditions as well as the upper bound of detection time are given, which demonstrate that the detection time of our strategy is much shorter than that of Polycarpou's approach. Simulation results on the three-tank system “DTS200” show the effectiveness and fastness of the proposed strategy.     

Fault magnitude estimation for processes

In this paper, the fault isolation and fault magnitude estimate methods are proposed. In the original fault isolation methods, contribution plots are popular. However, it is not accurate. In the original fault estimation methods, the authors assume that the fault magnitude is far greater than the normal measurement. However, the assumption is too strong. To avoid the above two problems, in this paper, the fault is isolated by determining the fault direction and the fault magnitude is estimated using the fault reconstruction. The proposed methods are used to Monte Carlo simulation and the electrical smelting magnesium furnace. From the simulation results, we can see that the proposed methods can solve the problems mentioned above effectively.     

Data-driven causal inference based on a modified transfer entropy

Causality inference and root cause analysis are important for fault diagnosis in the chemical industry. Due to the increasing scale and complexity of chemical processes, data-driven methods become indispensable in causality inference. This paper proposes an approach based on the concept of transfer entropy which was presented by Schreiber in 2000 to generate a causal map. To get a better performance in estimating the time delay of causal relations, a modified form of the transfer entropy is presented in this paper. Case studies on two simulated chemical processes, including the benchmark Tennessee Eastman process are performed to illustrate the effectiveness of this approach.     

A combined canonical variate analysis and Fisher discriminant analysis (CVA–FDA) approach for fault diagnosis

This paper proposes a combined canonical variate analysis (CVA) and Fisher discriminant analysis (FDA) scheme (denoted as CVA–FDA) for fault diagnosis, which employs CVA for pretreating the data and subsequently utilizes FDA for fault classification. In addition to the improved handling of serial correlations in the data, the utilization of CVA in the first step provides similar or reduced dimensionality of the pretreated datasets compared with the original datasets, as well as decreased degree of overlap. The effectiveness of the proposed approach is demonstrated on the Tennessee Eastman process. The simulation results demonstrate that (i) CVA–FDA provides better and more consistent fault diagnosis than FDA, especially for data rich in dynamic behavior; and (ii) CVA–FDA outperforms dynamic FDA in both discriminatory power and computational time.     

化工过程监测系统的传感器网络设置问题研究

化工过程监测系统的性能与其传感器网络的设置方式有直接联系。通过建立化工过程有向图模型,结合故障传播模式分析和传感器网络搜索算法,从过程监测的角度研究了传感器设置问题。给出了考虑故障可观性与分辨率下的传感器设置方法。并以CSTR-heat exchanger过程为例,说明了统计过程监测方法与传感器网络设置问题相结合的重要性。     

空分设备粗氩塔氮塞的诊断与估计方法

为提高空分产品质量,降低氮塞故障造成的不利影响,应用互信息方法建立了空分系统的符号有向图模型,改进了基于有向图的故障诊断方法,提出了故障严重程度的判别方法,实现了空分系统粗氩塔氮塞故障的快速诊断以及氮塞严重程度的准确估计.     

An algorithm for diagnosis of system failures in the chemical process

An attempt was made to apply graph theory to the diagnosis of the system failures in the chemical process. A signed digraph is used for a mathematical model representing the influences among elements of the system. The concept of a pattern on the signed digraph is introduced for representing a state of the system. In order to eliminate carrying out the complicated and inefficient quantitative simulation, the mathematical model of the system structure to represent the rpopagation of failures is simplified in a qualitative fashion. The origin of the system failure can be located at the maximal strongly-connected component in the cause-effect graph reflecting the pattern of abnormality. Even when the pattern is observed only partially, the assumption of single origin of the failure reduces, to some extent, the range of possible candidates to be the first cause of the failure.