Perform Fault Detection Research Articles

An intelligent fault diagnosis method for an electromechanical actuator based on sparse feature and long short-term network

Electromechanical actuators (EMAs), as the new generation of actuators, have an important impact on the safety of aircraft. With the development of measurement technology, a large amount of data provides a broad prospect for the data-based fault diagnosis method. However, the existence of redundant data increases the burden of software and hardware. Therefore, a semi-supervised sparse auto-encoder (SSAE) is employed to prune observed data based on sparsity analysis. Moreover, temporal and spatial relationships are explored by a multi-channel long short-term network to build a time series model, so as to perform fault detection and isolation based on the difference between its estimated and observed values. Due to its sparse feature extraction capability, the SSAE can improve the fault isolation accuracy while pruning observed data. Verification results confirm that the proposed method can effectively diagnose EMA faults.

Predictive Approach to Perform Fault Detection in Electrical Submersible Pump Systems.

It has been a great challenge for the oil and gas industry to timely identify any electrical submersible pump (ESP) abnormal performance to avoid ESP failure. Given the high cost of the ESP failure, more and more real-time surveillance systems are applied to monitor ESP performance to generate alarms in the case of failures. This paper presents a robust principal component analysis (PCA) model to perform fault detection for ESP systems continuously. A three-dimensional plot of scores of principal components was used to observe different patterns during the stable and failure periods. 47 cases of actual failure events and 40 cases of stable operating events were tested on the robust PCA model to generate prediction results. The testing results demonstrate that the robust PCA model has managed to identify 20 failure events before the actual failure time out of the 47 failure cases and has successfully distinguished all the 40 stable operating wells. This study has concluded that PCA has the potential to be used as a monitoring platform to recognize dynamic change and therefore to predict the developing failures in the ESP system.

Improved Statistical Pattern Analysis Monitoring for Complex Multivariate Processes Using Empirical Likelihood

This article developed an improved statistical pattern analysis (SPA) monitoring strategy for fault detection of complex multivariate processes using empirical likelihood. The technique based on statistical pattern analysis performs fault detection by inspecting change in the statistics of process variables (e.g., mean value, correlation coefficient, variance, kurtosis, etc.). It is capable of monitoring non-Gaussian or even nonlinear processes. However, the original SPA framework explicitly computes all the high-order statistics, which significantly increases the scale and dimensionality of the problem, especially in the case of complex multivariate processes. To alleviate this difficulty, we propose monitoring changes in the statistics with the same order using empirical likelihood, which is a widely used estimation method to construct confidence limits or regions for parameters with similar properties. As a result, changes in statistics of the same order can be translated into a single index; hence more information on the faulty conditions can be observed. Furthermore, by considering statistics of the same order, the scale of the problem is reduced significantly. The improved statistical pattern analysis monitoring strategy is suitable for monitoring complex multivariate processes. The performance of the improved method is illustrated by an application study to fault detection of the Tennessee Eastman (TE) process.

Machine learning‐based model for lithium‐ion batteries in BMS of electric/hybrid electric aircraft

SummaryReliable operation and control of battery packs can lead to increasing applications of batteries as energy sources for mobile power systems such as electric/hybrid electric aircraft. If the operation of a battery pack is controlled and monitored thoroughly, the safety in the battery system of an electrified aircraft can be guaranteed. The battery model has many applications in battery management systems such as battery performance analysis and fault detection. To achieve an accurate fault diagnosis for electric aircraft, an intelligent fault detection scheme within an accurate battery cell model is required. In this study, an adaptive lithium‐ion battery model is proposed in which models' parameters are estimated by a supervised machine learning paradigm. This adaptive battery model is developed based on a second order equivalent circuit model, which has a good representation of lithium‐ion batteries dynamics. Comparative verification experiments show good accuracy and robustness of the machine learning‐based parameter estimator lead to an accurate battery model with an average error less than 0.4%. Moreover, to see the effectiveness of this machine learning‐based model in fault detection applications, a model‐based fault diagnosis scheme is developed. Finally, the analysis of fault diagnosis tests under different test conditions proves that the proposed adaptive battery model can significantly improve the fault diagnosis accuracy of batteries.

Feasibility of Rotor Fault Detection from a Fluid Dynamics Perspective

The majority of condition monitoring techniques employed today consider the acquisitioning and analysis of structural responses as a means of profiling machine condition and performing fault detection. Modern research and newer technologies are driving towards non-contact and non-invasive methods for better machine characterisation. Yet current literature lacks investigations into the monitoring and detection of anomalous conditions using fluid dynamic behaviour. If one considers unshrouded rotors which are exposed to a full field of fluid interaction such as helicopter rotors and wind turbines amongst others, such an approach could potentially be beneficial. In this work, time-dependent fluid dynamic data is numerically simulated around a helicopter tail rotor blade using URANS CFD with the Open FOAM software package. Pressures are probed at locations in the field of the rotor and compared to results attained in an experimental investigation where good correlation is seen between the results. A blade is modelled with a seeded fault in the form of a single blade out of plane by 4°. Comparisons are drawn between the blade in its ‘healthy’ and ‘faulty’ configurations. It is observed that the fault can be detected by deviations in the amplitudes of the pressure signals for a single revolution at the probed locations in the field. These deviations manifest as increases in the frequency spectrum at frequencies equivalent to the rotational rate (1 per revolution frequencies). The results described are assessed for their fidelity when the pressure is probed at different locations in the domain of the rotor. Deviations in the pressure profiles over the surface of the blades are also seen for the asymmetric rotor configuration, but may prove too sensitive for practical application.

New trends in modeling and control of hybrid systems

New trends in modeling and control of hybrid systems

Data-Driven Incipient Fault Detection and Diagnosis for the Running Gear in High-Speed Trains

Incipient fault detection and diagnosis (FDD) is an important measure to improve the efficient, safe and stable operation of high-speed trains. This paper proposes a data-driven FDD method, namely deep slow feature analysis and belief rule base method (DSFA-BRB), for the running gears of high-speed trains. The method uses two kinds of statistics to perform fault detection on the multi-dimensional data of the running gears. In addition, the characteristics of more accurate data are extracted, which greatly reduces the complexity of constructing a diagnostic and quantitative model. Further, by constructing a BRB model combining expert knowledge and data, it is possible to avoid misjudgment caused by data incompleteness. Compared with the traditional methods, the DSFA-BRB algorithm has better performance in reducing fault alarm probability. Finally, the validity of the algorithm is verified by the actual running gears system.

Fault Diagnosis of a Granulator Operating under Time-Varying Conditions Using Canonical Variate Analysis

Granulators play a key role in many pharmaceutical processes because they are involved in the production of tablets and capsule dosage forms. Considering the characteristics of the production processes in which a granulator is involved, proper maintenance of the latter is relevant for plant safety. During the operational phase, there is a high risk of explosion, pollution, and contamination. The nature of this process also requires an in-depth examination of the time-dependence of the process variables. This study proposes the application of canonical variate analysis (CVA) to perform fault detection in a granulation process that operates under time-varying conditions. Beyond this, a different approach to the management of process non-linearities is proposed. The novelty of the study is in the application of CVA in this kind of process, because it is possible to state that the actual literature on the theme shows some limitations of CVA in such processes. The aim was to increase the applicability of CVA in variable contexts, with simple management of non-linearities. The results, considering process data from a pharmaceutical granulator, showed that the proposed approach could detect faults and manage non-linearities, exhibiting future scenarios for more performing and automatic monitoring techniques of time-varying processes.

Data-driven Performance Fault Detection in Commercial Wind Turbines

The operation and maintenance costs of wind parks make up a major fraction of a park's overall lifetime costs. They also include opportunity costs of lost revenue from avoidable power generation underperformance. We present a machine-learning based decision support method that minimizes these opportunity costs. By analyzing the stream of telemetry sensor data from the turbine operation, estimating highly accurate power reference relations and benchmarking, we can detect performance-related operational faults in a turbine- and site-specific manner. The most accurate power reference model is selected based on combinations of machine learning algorithms and regressor sets. Operating personnel can be alerted if a normal operating state boundary is exceeded. We demonstrate the performance fault detection method in a case study for a commercial grid-connected onshore wind turbine. Diagnosing a detected underperformance event, we find that the observed power generation deficiencies coincide with rotor blade misalignment related to low hydraulic pressure of the turbine's blade actuators.

Fault Isolation for A Complex Decentralized Waste Water Treatment Facility

SummaryDecentralized waste water treatment facilities monitor many features that are complexly related. The ability to detect the onset of a fault and to identify variables accurately that have shifted because of the fault are vital to maintaining proper system operation and high quality produced water. Various multivariate methods have been proposed to perform fault detection and isolation, but the methods require data to be independent and identically distributed when the process is in control, and most require a distributional assumption. We propose a distribution-free retrospective change-point-detection method for auto-correlated and non-stationary multivariate processes. We detrend the data by using observations from an in-control time period to account for expected changes due to external or user-controlled factors. Next, we perform the fused lasso, which penalizes differences in consecutive observations, to detect faults and to identify shifted variables. To account for auto-correlation, the regularization parameter is chosen by using an estimated effective sample size in the extended Bayesian information criterion. We demonstrate the performance of our method compared with a competitor in simulation. Finally, we apply our method to waste water treatment facility data with a known fault, and the variables identified by our proposed method are consistent with the operators’ diagnosis of the fault's cause.

Real-Time Adaptive Differential Feature-Based Protection Scheme for Isolated Microgrids Using Edge Computing

Inhabited isolated areas require independent power systems called isolated microgrids. Conventional protection schemes are not suitable for isolated microgrids due to their dependence on significant fault current. This article introduces an adaptive differential feature-based protection scheme for fault detection and faulty phase identification for isolated microgrids. Instead of differential currents, it proposes differential features, where edge computing is used to extract the features at point of measurement. The most informative one-dimensional feature is extracted from each phase at both ends of the line. The line differential feature is compared with an adaptive threshold to perform fault detection. The threshold is updated every second and is immune to system noises. Faulty phase identification is performed with a logical comparison. An in-memory circular buffer is proposed to store the local features in case of communication delay. Communication and synchronization are experimentally validated, and the worst-case latency is reported along with individual processing time for data acquisition, feature extraction, communication, and fault detection. Extensive validations are done with IEC-based and IEEE 13-node-based isolated microgrid. The test results are compared with state-of-the-art methods and proved that the proposed scheme is practical to be adopted in real- world isolated microgrids of different sizes and topologies.

Incipient fault detection of the high-speed train air brake system with a combined index

Reliability of the high-speed train air brake system is very critical to ensure a safe and comfortable operation environment for passengers. In our recently published work, a new monitoring strategy based on the concept of inter-variable variance (IVV) was proposed. It was illustrated by theoretical analysis and extensive experiments that this strategy is effective for several kinds of faults, and superior to the KNORR logic which is adopted by most high-speed trains in practice. However, the aforementioned strategy based on conventional IVV still suffers from two drawbacks. More specifically, due to the specific property of IVV, this strategy is unable to detect certain faults even if the fault is very serious; in addition, the detectability of conventional IVV for incipient faults with relatively small magnitudes still needs to improve. In this paper, a fault detection method incorporating the idea of four stages partition and involving a new combined statistic is presented to perform fault detection for high-speed train air brake systems. Through theoretical reasoning, it is pointed out that the proposed method can overcome drawbacks of the conventional IVV method to a large extent. To demonstrate the effectiveness and merits of the proposed fault detection method, in comparison with existing methods, experimental studies are carried out on a practical high-speed train air brake test bench.

A Simple Condition Monitoring Method for Gearboxes Operating in Impulsive Environments.

Reliable condition indicators are necessary to perform effective diagnosis and prognosis. However, the vibration signals are often corrupted with non-Gaussian noise and rotating machines may operate under time-varying operating conditions. This impedes the application of conventional condition indicators. The synchronous average of the squared envelope is a relatively simple yet effective method to perform fault detection, fault identification and fault trending under constant and time-varying operating conditions. However, its performance is impeded by the presence of impulsive signal components attributed to impulsive noise or the presence of other damage modes in the machine. In this work, it is proposed that the synchronous median of the squared envelope should be used instead of the synchronous average of the squared envelope for gearbox fault diagnosis. It is shown on numerical and experimental datasets that the synchronous median is more robust to the presence of impulsive signal components and is therefore more reliable for estimating the condition of specific machine components.

An enhanced rolling bearing fault detection method combining sparse code shrinkage denoising with fast spectral correlation

Rolling bearings are important supporting components widely used in rotating machinery and are prone to failure, it is thus important to perform fault detection of rolling bearing quickly and accurately. Aiming at the problem that it is difficult to extract the weak impulses buried in strong background noise in rolling bearing fault diagnosis, this paper proposes an enhanced fault detection method combining sparse code shrinkage denoising with fast spectral correlation according to the cyclic statistical properties of defective bearing vibration signals. First, in view of the non-Gaussian statistical properties of the periodic impulses caused by the localized bearing defect in vibration signals, the sparse code shrinkage algorithm is employed to denoise the original noisy signal, thereby highlighting the periodic impulses. Then, the Fast Spectral Correlation (Fast-SC) algorithm is used to process the denoised signal to get the cyclic spectral correlation. Finally, the squared enhanced envelope spectrum (SEES) is presented to effectively detect and identify the rolling bearing faults. Experimental results demonstrate the validity and superiority of the proposed method in rolling bearing fault detection through the comparison with the Fast-SC, spectral kurtosis and Infogram.

Online sensor performance monitoring and fault detection for discrete linear parameter-varying systems

Online sensor performance monitoring and fault detection for discrete linear parameter-varying systems

Online sensor performance monitoring and fault detection for discrete linear parameter-varying systems

This paper proposes a fault detection algorithm to identify online sensor performance degradation and failure, where the sensor faults are characterised by variations of the sensor measurement noise covariance matrix. That is, the proposed algorithm estimates the slowly-varying sensor measurement noise covariance and detects the abrupt and/or intermittent change of sensor measurement noise covariance. A memory-based technique is used to detect the abrupt (or intermittent) change of sensor noise covariance matrix. The memory-based technique is adopted due to its simplicity and online applicability. The proposed algorithm originally is designed for discrete linear time-varying (DLTV) systems and applied to discrete linear parameter-varying (DLPV) systems. Simulation results show that the proposed algorithm is capable of estimating the slowly-varying and detecting the abrupt (or intermittent) change of sensor measurement noise covariance for multiple-input and multiple-output discrete linear parameter-varying systems, where the scheduling parameters lie within a compact set. Furthermore, the proposed estimation algorithm shows a reasonable rate of convergence.

Detection of sensor precision degradation by monitoring second-order statistics

For industrial processes, there are usually a number of measurement sensors equipped for monitoring and control purposes. In practice, sensors may suffer from the precision degradation phenomenon due to several aspects such as aging and ambient interference. This phenomenon may lead to imprecise or even incorrect control commands and indications, so the corresponding fault detection task is of vital importance. In this paper, inspired by the fact that precision degradation of a sensor can result in the increase of the variable’s variance, an algorithm based on second-order statistics analysis is proposed to accomplish the detection task for sensor precision degradation faults. By employing the sliding window technique, second-order statistics of process variables are first extracted. Then, conventional principal component analysis (PCA) is used as a dissimilarity quantification tool, with detection statistics and corresponding control limits established, to perform fault detection. Finally, simulations on a numerical example and the continuous stirred tank reactor (CSTR) benchmark process are performed to illustrate the effectiveness and advantages of the proposed method, in comparison with some existing methods such as PCA, dynamic PCA, and dissimilarity (DISSIM).

A Robust State Estimation Method for Unknown, Time-Varying and Featureless Aircraft Sensor Failures

Fault-tolerant state estimation is necessary for analytical redundancy in aviation safety, but complicated fault conditions pose a major challenge to reliable estimations. In this paper, we propose a new state-estimation method based on an intermittent-measurement Kalman filter, a maximum-likelihood estimation rule, and the Gaussian mixture reduction. This method is robust to time-varying and featureless faults, and requires no past history of innovation errors to perform fault detection and isolation, leading to the easy implementation and improved accuracy under severe sensor-failure conditions. With an example application of monitoring the rotation speeds of a two-spool jet engine in simulations, the effectiveness of the proposed method is verified.

Fault detection in flotation processes based on deep learning and support vector machine

Effective fault detection techniques can help flotation plant reduce reagents consumption, increase mineral recovery, and reduce labor intensity. Traditional, online fault detection methods during flotation processes have concentrated on extracting a specific froth feature for segmentation, like color, shape, size and texture, always leading to undesirable accuracy and efficiency since the same segmentation algorithm could not be applied to every case. In this work, a new integrated method based on convolution neural network (CNN) combined with transfer learning approach and support vector machine (SVM) is proposed to automatically recognize the flotation condition. To be more specific, CNN function as a trainable feature extractor to process the froth images and SVM is used as a recognizer to implement fault detection. As compared with the existed recognition methods, it turns out that the CNN-SVM model can automatically retrieve features from the raw froth images and perform fault detection with high accuracy. Hence, a CNN-SVM based, real-time flotation monitoring system is proposed for application in an antimony flotation plant in China.

Advancing Building Fault Diagnosis Using the Concept of Contextual and Heterogeneous Test

Fault diagnosis and maintenance of a whole-building system is a complex task to perform. A wide range of available building fault detection and diagnosis (FDD) tools are only capable of performing fault detection using behavioral constraints analysis. However, the validity of the detected symptom is always questionable. In this work, we introduce the concept of the contextual test with validity constraints, in the context of building fault diagnostics. Thanks to a common formalization of the proposed heterogeneous tests, rule-, range-, and model-based tests can be combined in the same diagnostic analysis that reduces the whole-building modeling effort. The proposed methodology comprises the minimum diagnostic explanation feature that can significantly improve the knowledge of the building facility manager. A bridge diagnosis approach is used to describe the multiple fault scenarios. The proposed methodology is validated on an experimental building called the center for studies and design of prototypes (CECP) building located in Angers, France.