Improved canonical correlation analysis-based fault detection methods for industrial processes

Recent research has emphasized the successful application of canonical correlation analysis (CCA) to perform fault detection (FD) in both static and dynamic processes with additive faults. However, dealing with multiplicative faults has not been as successful. Thus, this paper considers the application of CCA to deal with the detection of incipient multiplicative faults in industrial processes. The new approaches incorporate the CCA-based FD with the statistical local approach. It is shown that the methods are effective in detecting incipient multiplicative faults. Experiments using a continuous stirred tank heater and simulations on the Tennessee Eastman process are provided to validate the proposed methods.     

Fault detection of process correlation structure using canonical variate analysis-based correlation features

This paper proposes a canonical variate analysis (CVA) approach based on feature representation of canonical correlation for the monitoring of faults associated with changes in process correlations, which involves two new metrics, R_s and R_r, corresponding to the state and residual spaces. The utilization of the canonical correlation feature can improve the monitoring proficiency by providing more application-dependent representations compared with the original data, as well as a decreased degree of redundancy in the feature space. A physical interpretation is provided for the canonical correlation-based method. The effectiveness of the proposed approach for the monitoring of process correlation changes is demonstrated for both abrupt (step change) and incipient (slow drift) types of faults in simulation studies of a network system. In the simulation results, the canonical correlation-based method has superior performance over both the causal dependency-based method and the traditional variable-based method.     

Analysis of performance criteria in sampled-data fault detection

Sampled-data control systems are widely used in industry. In this paper the problem of fault detection and isolation (FDI) in sampled-data systems is studied. Many existing methods to design a robust sampled-data FDI are based on optimization of a norm based performance index. Our focus in this study is on the selection of the performance index. It is shown that the existing performance indices are not appropriate choices in the sense that they do not satisfy some expected intuitive properties. To resolve this, an alternative performance index is defined after converting the FDI problem to a standard control problem. This performance index is shown to satisfy the expected properties.     

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.     

Norm invariant discretization for sampled-data fault detection

In this paper, the problem of fault detection in sampled-data systems is studied. It is shown that norms of a sampled system are equal to the corresponding norms of a certain discrete time system. Based on this discretization, the sampled-data fault detection problem can be converted to an equivalent discrete-time problem. A framework that unifies the H₂ and H_∞ optimal residual generators in sampled-data systems is then proposed.     

On linear observers and application to fault detection in synchronous generators

This work introduces an observer structure and highlights its distinct advantages in fault detection and isolation. Its application to the issue of shorted turns detection in synchronous generators is demonstrated. For the theoretical foundation, the convergence and design of Luenberger-type observers for disturbed linear time-invariant （LTI） single-input single-output （SISO） systems are reviewed with a particular focus on input and output disturbances. As an additional result, a simple observer design for stationary output disturbances that avoids a system order extension, as in classical results, is proposed.     

A frequency domain approach to fault detection in sampled-data systems

This paper deals with fault detection problems in sampled-data (SD) systems. A tool is first introduced for the analysis of intersample behavior of SD systems in the frequency domain from the viewpoint of fault detection and isolation. Based on it, a direct design approach of fault detection systems for SD systems is proposed, and further the problem of full decoupling from unknown disturbances is studied.     

Semi-supervised kernel canonical correlation analysis with application to human fMRI

Kernel canonical correlation analysis (KCCA) is a general technique for subspace learning that incorporates principal components analysis (PCA) and Fisher linear discriminant analysis (LDA) as special cases. By finding directions that maximize correlation, KCCA learns representations that are more closely tied to the underlying process that generates the data and can ignore high-variance noise directions. However, for data where acquisition in one or more modalities is expensive or otherwise limited, KCCA may suffer from small sample effects. We propose to use semi-supervised Laplacian regularization to utilize data that are present in only one modality. This approach is able to find highly correlated directions that also lie along the data manifold, resulting in a more robust estimate of correlated subspaces.Functional magnetic resonance imaging (fMRI) acquired data are naturally amenable to subspace techniques as data are well aligned. fMRI data of the human brain are a particularly interesting candidate. In this study we implemented various supervised and semi-supervised versions of KCCA on human fMRI data, with regression to single and multi-variate labels (corresponding to video content subjects viewed during the image acquisition). In each variate condition, the semi-supervised variants of KCCA performed better than the supervised variants, including a supervised variant with Laplacian regularization. We additionally analyze the weights learned by the regression in order to infer brain regions that are important to different types of visual processing.     

Design of unknown input functional observers for nonlinear systems with application to fault diagnosis

This paper considers the design of low-order unknown input functional observers for robust fault detection and isolation of a class of nonlinear Lipschitz systems subject to unknown inputs. The proposed functional observers can be used to generate residual signals to detect and isolate actuator faults. By using the generalized inverse approach, the effect of the unknown inputs can be decoupled completely from the residual signals. Conditions for the existence and stability of reduced-order unknown input functional observer are derived. A design procedure for the generation of residual signals to detect and isolate actuator faults is presented using the proposed unknown-input observer-based approach. A numerical example is given to illustrate the proposed fault diagnosis scheme in nonlinear systems subject to unknown inputs.     

Concurrent quality and process monitoring with canonical correlation analysis

Canonical correlation analysis (CCA) is a well-known data analysis technique that extracts multidimensional correlation structure between two sets of variables. CCA focuses on maximizing the correlation between quality and process data, which leads to the efficient use of latent dimensions. However, CCA does not focus on exploiting the variance or the magnitude of variations in the data, making it rarely used for quality and process monitoring. In addition, it suffers from collinearity problems that often exist in the process data. To overcome this shortcoming of CCA, a modified CCA method with regularization is developed to extract correlation between process variables and quality variables. Next, to handle the issue that CCA focuses only on correlation but ignores variance information, a new concurrent CCA (CCCA) modeling method with regularization is proposed to exploit the variance and covariance in the process-specific and quality-specific spaces. The CCCA method retains the CCA's efficiency in predicting the quality while exploiting the variance structure for quality and process monitoring using subsequent principal component decompositions. The corresponding monitoring statistics and control limits are then developed in the decomposed subspaces. Numerical simulation examples and the Tennessee Eastman process are used to demonstrate the effectiveness of the CCCA-based monitoring method.     

Incipient fault detection with smoothing techniques in statistical process monitoring

In modern industry, detecting incipient faults timely is of vital importance to prevent serious system performance deterioration and ensure optimal process operation. Recently, multivariate statistical process monitoring (MSPM) techniques have been extensively studied and widely applied to modern industrial systems. However, conventional fault detection indices utilized in statistical process monitoring are not sensitive to incipient faults with small magnitude. In this paper, by introducing two representative smoothing techniques, novel incipient fault detection strategies based on a generic fault detection index in MSPM are proposed. Fault detectability for each proposed strategy is analyzed. In addition, the effects of the smoothing parameters on fault detection, including advantages and disadvantages, are also investigated. Finally, case studies on a numerical example and two practical industrial processes are carried out to demonstrate the effectiveness of the proposed incipient fault detection strategies.     

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...     

High-gain observation with disturbance attenuation and application to robust fault detection

One approach to the problem of residual generation in a purpose of fault detection is to use an observer. One particular difficulty is to distinguish between faults and disturbances. Various observers have already been inspected in that direction, generally based on exact decoupling w.r.t. unknown disturbances. Here the use of high-gain observer techniques is inspected, with a purpose of attenuation of disturbances rather than exact decoupling: conditions allowing some “robust partial estimation” are first presented, and their possible use in fault detection is then discussed.     

Outlier detection based on Gaussian process with application to industrial processes

Due to the extensive usage of data-based techniques in industrial processes, detecting outliers for industrial process data become increasingly indispensable. This paper proposes an outlier detection scheme that can be directly used for either process monitoring or process control. Based on traditional Gaussian process regression, we develop several detection algorithms, of which the mean function, covariance function, likelihood function and inference method are specially devised. Compared with traditional detection methods, the proposed scheme has less postulation and is more suitable for modern industrial processes. The effectiveness of the proposed scheme is verified by experiments on both synthetic and real-life data sets.     

Subspace decomposition approach of fault deviations and its application to fault reconstruction

In the present work, a new subspace decomposition approach of fault deviations is developed in the context of principal component analysis (PCA) based monitoring system for fault diagnosis via reconstruction. The fault effects are decomposed in different monitoring subspaces, principal subspace (PCS) and residual subspace (RS), and the significant fault deviations that are responsible for the concerned alarming monitoring statistic are calculated. This is achieved by designing a two-step feature decomposition procedure in each monitoring subspace. In the first step, the relative fault deviations are sorted by comparing the fault variations with the normal variations. All possible fault deviations that may contribute to the out-of-control monitoring statistics are collected. In the second step, PCA is performed on the chosen fault information where the largest fault deviation directions are decomposed in order. By the two-step decomposition, in each monitoring subspace, two different parts are separated for the purpose of fault reconstruction. One is composed of the concerned fault deviations that contribute to alarming monitoring statistics which are thus significant to remove the out-of-control signals. The other is composed of general variations that are deemed to follow normal rules and thus insignificant to remove alarming monitoring statistics. Theoretical support is framed and the related statistical characteristics are analyzed. Its feasibility and performance are illustrated with data from the three-tank system and the Tennessee Eastman (TE) benchmark process.     

On observer-based fault detection for nonlinear systems

This paper addresses analysis and integrated design of observer-based fault detection (FD) for nonlinear systems. To gain a deeper insight into the observer-based FD framework, definitions and existence conditions for nonlinear observer-based FD systems are studied first. Then, a scheme for an integrated design of observer-based FD systems for affine nonlinear systems is proposed. Our work is considerably inspired by the study on input–output stability and stabilization of nonlinear systems. Examples are given at the end of the paper to illustrate the theoretical results.     

Identification of neural dynamic models for fault detection and isolation: the case of a real sugar evaporation process

The paper deals with problems of fault detection of industrial processes using dynamic neural networks. The considered neural network has a feed-forward multi-layer structure and dynamic characteristics are obtained by using dynamic neuron models. Two optimisation problems are associated with neural networks. The first one is selection of a proper network structure which is solved by using information criteria such as the Akaike Information Criterion or the Final Prediction Error. In turn, the training of the network is performed by a stochastic approximation algorithm. The effectiveness of the proposed fault detection and isolation system is checked using real data recorded in Lublin Sugar Factory, Poland. Additionally, a comparison with alternative approaches is presented.     

Relaxed fault detection and isolation: An application to a nonlinear case study

Given a number of possibly concurrent faults (and disturbances) that may affect a nonlinear dynamic system, it may not be possible to solve the standard fault detection and isolation (FDI) problem, i.e., to detect and isolate each single fault from all other, possibly concurrent faults and disturbances, due to the violation of the available necessary conditions of geometric nature. Motivated by a robotic application where this negative situation structurally occurs, we propose some relaxed formulations of the FDI problem and show how necessary and sufficient conditions for their solution can be derived from those available for standard FDI. The design of a hybrid residual generator follows directly from the fulfillment of the corresponding solvability conditions. In the considered nonlinear case study, a robotic system affected by possible actuator and/or force sensor faults, we detail the application of these results and present experimental tests for validation.     

Robust fault detection in linear systems based on PI observers

A parametric approach for robust fault detection in linear systems with unknown disturbances is presented. The residual is generated using full-order proportional-integral (PI) observers. The approach is based on a result for PI observer design recently proposed. In terms of the design degrees of freedom provided by the parametric PI observer design and a group of introduced parameter vectors, a sufficient and necessary condition for PI observer design with disturbance decoupling is established. By properly constraining the design parameters according to this proposed condition, the effect of the disturbance to the residual signal is decoupled, and a simple algorithm is presented. The presented approach offers all the degrees of design freedom. A numerical example illustrates the effect of the proposed approach.     

Survey and application of sensor fault detection and isolation schemes

Model-based sensor fault detection, isolation and accommodation (SFDIA) is a direction of development in particular with UAVs where sensor redundancy may not be an option due to weight, cost and space implications. SFDIA via neural networks (NNs) have been proposed over the years due to their nonlinear structures and online learning capabilities. The majority of papers tend to consider single sensor faults. While useful, this assumption can limit application to real systems where sensor faults can occur simultaneously or consecutively. In this paper we consider the latter scenario, where it is assumed that a 1 s time gap is present between consecutive faults. Furthermore few applications have considered fixed-wing UAVs where full autonomy is most needed. In this paper an EMRAN RBF NN is chosen for modelling purposes due to its ability to adapt well to nonlinear environments while maintaining high computational speeds. A nonlinear UAV model is used for demonstration, where decoupled longitudinal motion is considered. System and measurement noise is also included in the UAV model as wind gust disturbances on the angle of attack and sensor noise, respectively. The UAV is assumed to operate at an initial trimmed condition of speed, 32 m/s and altitude, 1000 m. After 30 separate SFDIA tests implemented on a 1.6 GHz Pentium processor, the NN-SFDIA scheme detected all but 2 faults and the NN processing time was 97% lower than the flight data sampling time.