Fault Symptoms Research Articles

Analyse resilience risks in microservice architecture systems with causality search and inference algorithms

The microservice architecture has already become the mainstream architecture pattern of web service applications in recent years. However, compared with traditional software architectures, the microservice architecture has a more sophisticated deployment structure, which makes it have to face more potential risks with greater diversity of fault symptoms. Microservice practitioners started to use the word 'resilience' to describe the capability of coping with different unexpected conditions. How to judge whether a system environment disruption is a risk of microservice resilience, and how to analyse resilience risks before the system is released, are the research questions in microservice development. As the practice of chaos engineering has solved the problem of resilience risk identification, this paper focuses on how to analyse identified resilience risks in microservice architecture systems, and a resilience risk analysis method is proposed. Based on performance monitoring data collected during chaos experiments, the analysis method uses the causality search algorithm to build causality graphs of performance indicators, and generates causality chains to system operators by the causality inference algorithm. The effectiveness of the proposed approach is proved by conducting a case study on a microservice architecture system.

Early Fault Diagnosis and Classification of Ball Bearing Using Enhanced Kurtogram and Gaussian Mixture Model

Ball bearing failure is one of the major obstacles to the effective operation of large mechanical systems. During maintenance, the initial diagnosis of a fault within the bearing is key to reducing repair costs and improving the efficiency of the system. However, such faults are difficult to accurately diagnose due to noise and the unusual and unpredictable phenomena that they cause in the peaks of the measured signal. In this paper, we present an effective analytical technique for the early diagnosis of ball bearing faults based on vibration data derived from the bearings. We apply a feature extraction technique based on spectral kurtosis (SK) and then filter the results using statistical approaches. The actual defects in the bearings are evaluated in terms of a Gaussian mixture model; principal component analysis is then used to reduce the misclassifications caused by noise and weak fault symptoms. We verified the proposed algorithm experimentally and compared the results of our diagnostic technique to those obtained using the root mean square (rms) of the vibration data to evaluate the performance of the SK-based technique. The rms-based method can only be used to identify defects, whereas the SK-based method can evaluate the level and severity of the fault. In addition, the hybrid statistical SK-based process can diagnose faults without data on the rotational speed, actual load, or design specifications of the bearings or equipment that they are used in.

A dynamic Bayesian network based methodology for fault diagnosis of subsea Christmas tree

A subsea Christmas tree (XT) is an extremely important part of a subsea production system. The safety-fault of subsea XT indicates that no major safety incidents are difficult to diagnose. To identify the faulty components and distinguishing the fault types, including the blocking, leakage, and especially safety-fault, we present a dynamic Bayesian networks (DBN)-based fault diagnosis methodology of subsea XT considering component degradation and safety-fault. As the performance of components degrades over time, the diagnosis results can differ at different times for the given identical fault symptoms. DBNs are established to model the dynamic degradation of components in a system under additional information by using the failure rate, and fault diagnosis is conducted through a backward analysis of DBNs. Three fault diagnosis cases of subsea XT system are investigated. In case 1, when safety-fault occur on surface control subsea safety valve (SCSSV) and production main valve (PWV) components, the absolute difference in the posterior and prior probabilities of safety-fault for SCSSV and PWV was >50%. In case 2, when the blocking and leakage occur in SCSSV and annular main valve (AMV) components, respectively, the absolute difference between the posterior and prior probabilities of blocking for the SCSSV was >30%, and the absolute difference between the posterior and prior probabilities of leakage for AMV was >30%. In case 3, when the fault occur in production control valve (PCV) and chemical injection valve 1 (CIV1) components, respectively, the absolute difference between the posterior and prior probabilities of leakage for PCV and CIV1 was >60%. Three fault diagnosis cases validate the accuracy and effectiveness of the proposed methodology. This method is appropriate in providing maintenance instructions to engineers.

Gearbox Fault Prediction of Wind Turbines Based on a Stacking Model and Change-Point Detection

The fault diagnosis and prediction technology of wind turbines are of great significance for increasing the power generation and reducing the downtime of wind turbines. However, most of the current fault detection approaches are realized by setting a single alarm threshold. Considering the complicated working conditions of wind farms, such methods are prone to ignore the fault, send out a false alarm, or leave insufficient troubleshooting time. In this work, we propose a gearbox fault prediction approach of wind turbines based on the supervisory control and data acquisition (SCADA) data. A stacking model composed of Random Forest (RF), Gradient Boosting Decision Tree (GBDT), and Extreme Gradient Boosting (XGBOOST) was constructed as the normal behavior model to describe the normal conditions of the wind turbines. We used the Mahalanobis distance (MD) instead of the residual to measure the deviation of the current state from the normal conditions of the turbines. By inputting the MD series into the proposed change-point detection algorithm, we can obtain the change point at which the fault symptom begins to appear, and thus achieving the fault prediction of the gearbox. The proposed approach is validated on the historical data of 5 wind turbines in a wind farm, which proves its effectiveness to detect the fault in advance.

Selected Rolling Bearing Fault Diagnostic Methods in Wheel Embedded Permanent Magnet Brushless Direct Current Motors

In recent years, the number of outer rotor permanent magnet brushless direct current (PM BLDC) motor drives has been intensively growing. Due to the specifics of drive operation, bearing faults are especially common, which results in motor stoppage. In a number of these types of motor applications, the monitoring and diagnostics of bearing conditions is relatively rare. This article presents the results of research aimed at searching for simple and simultaneously effective methods for assessing the condition of bearings that can be built into the drive control system. In the experimental research, four vibration signal processing methods were analysed with regards to the identification accuracy of fault symptoms in the geometric elements of bearings (characteristic frequencies). The results are presented for three cases of bearing faults and compared with a new bearing, they were obtained based on a vibration signal analysis using the classical fast Fourier transform (FFT), Fourier transform of signal absolute values, Fourier transform of an envelope signal obtained using the Hilbert transform, and the Fourier transform of a signal filtered with the Teager–Kaiser energy operator (TKEO).

Combination of condition monitoring and prognosis systems based on current measurement and PSO-LS-SVM method for wind turbine DFIGs with rotor electrical asymmetry

An important and strategic instrument in energy market is wind turbine (WT). WT consists of many components, whose operation and maintenance of them are very sensitive due to their unstable dynamic and environmental conditions. Condition monitoring (CM) and fault prognosis approaches are typical methods in order to decrease the production cost of energy and WT downtime. In this paper, a new combinatory system of CM and fault prognosis based on feature extraction and prediction algorithm is proposed. Firstly, 3 phase current signals are collected. In the next step, proper features selection is done in time and frequency domain. The CM operation is based on a feature amplitude value crossing of an alarm threshold. The fault prognosis is conducted with improved support vector machine (SVM) method. The least squares (LS) type of SVM method is used for fault prediction and particle swarm optimization (PSO) is used for the optimum selection of SVM and kernel parameters. When the trend of fault symptom crosses through the specific alarm threshold, the prognostics unit is activated and remaining useful life (RUL) prediction is performed using PSO-LSSVM. For a scenario of the fault, generator rotor winding failure based on rotor electrical asymmetry is created with different percentages in the doubly-fed induction generator. The effectiveness of the proposed prognosis structure is evaluated using a WT simulation setup. For this setup, a 2.5 Mw WT experimental data and equipment parameters are used. The statistical investigations show the precision of PSO-LSSVM in RUL prediction. The different simulations confirm that improved condition-based maintenance performances can be obtained.

Fuzzy logic based open-circuit fault diagnosis in IGBT for CMLI fed PMSM drive

For the objective of enhancing the reliability in power electronics, a fault diagnosis technique has been reported for a three-phase cascaded multilevel inverter (CMLI). This paper presents a fuzzy based fault diagnosis technique based on the average current Parks vector technique for the three-phase CMLI fed permanent-magnet synchronous motor (PMSM) drive. The three-phase normalized currents divided into positive (Eq+) and negative groups (Eq−). Then the average current Park's vector technique is used to measure the fault symptom variables by using the phase current information. The proposed fault diagnosis technique can detect and locate the single or multiple open-circuit faults, and also the intermittent faults in IGBT's at any condition (ex., change in mechanical speed, change in load torque and change in system parameters), which can improve the reliability of the motor drive system. The effectiveness of the proposed method is validated by the simulation and experimental results.

Characterizing long-term wear and tear of ion-selective pH sensors.

Today, the development and testing of methods for fault detection and identification in wastewater treatment research relies on two important assumptions: (i) that sensor faults appear at distinct times in different sensors and (ii) that any given sensor will function near-perfectly for a significant amount of time following installation. In this work, we show that such assumptions are unrealistic, at least for sensors built around an ion-selective measurement principle. Indeed, long-term exposure of sensors to treated wastewater shows that sensors exhibit fault symptoms that appear simultaneously and with similar intensity. Consequently, this suggests that future research should be reoriented towards methods that do not rely on the assumptions mentioned above. This study also provides the first empirically validated sensor fault model for wastewater treatment simulation, which is useful for effective benchmarking of both fault detection and identification methods and advanced control strategies. Finally, we evaluate the value of redundancy for remote sensor validation in decentralized wastewater treatment systems.

Research on an enhanced scale morphological-hat product filtering in incipient fault detection of rolling element bearings

Incipient vibration signals of rolling element bearing are usually characterized by weak fault symptoms and multiple interference source components, which imply that it is difficult to recognize effectively the defects of rolling element bearing at an early stage. To address the issue, a novel early fault detection strategy based on an enhanced scale morphological-hat product filtering (ESMHPF) is proposed in this paper. Firstly, motivated by the existing morphology theory, the concept of morphology-hat product operation (MHPO) is presented to handle the collected weak fault signal, which can extract efficiently periodic impulse characteristics closely linked to the bearing defects. Subsequently, diagonal slice spectra (DSS) are incorporated into morphological analysis, which can achieve the efficacy of noise rejection and feature enhancement. Ultimately, the optimal scale morphological filtering results are determined by using a sensitive index termed as fault feature ratio (FFR) for identifying weak damage feature and completing early fault detection. Simulated signal and two experimental cases of run-to-failure are performed to assess the efficacy of the proposed algorithm. The analysis results achieved show that the formulated algorithm can identify clearly early fault symptoms immersed in bearing vibration data. Moreover, the availability of superiority of our designed approach is demonstrated by comparing with traditional multiscale morphological filtering and some existing algorithm. This study provides a new ideafor the improvement of incipient damage identification of rolling element bearings.

Model-Based Diagnosis of Hybrid Systems Using Satisfiability Modulo Theory

Currently, detecting and isolating faults in hybrid systems is often done manually with the help of human operators. In this paper we present a novel model-based diagnosis approach for automatically diagnosing hybrid systems. The approach has two parts: First, modelling dynamic system behaviour is done through well-known state space models using differential equations. Second, from the state space models we calculate Boolean residuals through an observer-pattern. The novelty lies in implementing the observer pattern through the use of a symbolic system description specified in satisfiability theory modulo linear arithmetic. With this, we create a static situation for the diagnosis algorithm and decouple modelling and diagnosis. Evaluating the system description generates one Boolean residual for each component. These residuals constitute the fault symptoms. To find the minimum cardinality diagnosis from these symptoms we employ Reiter’s diagnosis lattice.For the experimental evaluation we use a simulation of the Tennessee Eastman process and a simulation of a four-tank model. We show that the presented approach is able to identify all injected faults.

Intelligent Fault Inference of Inverters Based on a Three-Layer Bayesian Network

A three-layer Bayesian intelligent fault inference model (BIFIM) for an inverter is established to infer the probable uncertain faults. The topological structure of the BIFIM includes the inverter’s operation conditions for the first layer, the inverter’s faults for the second layer, and the fault symptoms for the third layer, which combines the field technicians’ knowledge and experiences with historical running data. The prior probability table of the root node is acquired by the method of basic probabilities corrected historical operation data. The conditional probability parameter table of the BIFIM is obtained by the improved maximum expectation algorithm. Four kinds of incomplete evidence were reasoned and verified, including simple evidence with obvious support, incomplete evidence information, complex evidence without obvious support, and evidence with information conflict. The proposed strategy can make use of the available evidences to inference the probabilities of faults, indicating different reasoning abilities under the different degree of completeness of evidence, especially demonstrating the same inference result under some incomplete evidence information as under complete evidence information.

High Voltage Circuit Breaker Fault Diagnosis Based on Reasoning chain and Bayes

In order to find the reason of the high voltage circuit breaker rejection quickly, an fault diagnosis method based on reasoning chain and Bayesian network (BN) is presented in the paper. Firstly, the concept of event set is introduced. A fault symptom set is constructed based on the measured circuit breaker characteristic parameters and warning signals. And the direct cause of circuit breaker failure, such as a part of the circuit breaker components or mechanism failure constitute the fault cause set. After that, when the high voltage circuit breaker fails, a subset of fault symptoms and fault causes is obtained according to the information. Then an reasoning chain is constructed according to the causal relationship between the two. Finally, a corresponding Bayesian network model for reasoning chain is built. The probability of fault causes is obtained by Bayesian backward reasoning, so as to realize the rapid analysis of circuit breaker faults. And the effectiveness of the method is verified by the result of practical fault diagnosis examples.

A Transformer Fault Diagnosis Method Based on Bayesian Network

This paper summarizes the fault types and fault hazards of the transformer, and combs the corresponding relationship between the transformer fault types and the fault symptoms, which provides reference for the transformer fault diagnosis and state maintenance. Based on the analysis of the existing transformer fault diagnosis, the transformer fault diagnosis based on Bayesian network is mainly studied. Bayesian network has advantages in machine learning, data reasoning and so on. It has been widely applied in the field of fault diagnosis. In this paper, the QMR model of transformer fault diagnosis is established by using the principle of Bayesian network and the relationship between fault type and fault feature, and the model is used to diagnose the specific fault. Then the paper further analyzes the result of fault diagnosis in order to prove the correctness and validity of Bayesian network.

Data-driven Detection and Diagnosis of Incipient Faults in Electrical Drives of High-Speed Trains

Incipient faults in electrical drives can corrupt overall performance of high-speed trains; however, they are difficult to discover because of their slight fault symptoms. By sufficiently exploiting the distribution information of incipient faults, this paper presents the reason why incipient faults cannot be detected by the existing fault detection and diagnosis (FDD) methods. Under principal component analysis (PCA) framework, we propose a new data-driven FDD method, which is named probability-relevant PCA (PRPCA), for electrical drives in high-speed trains. The salient strengths of the PRPCA-based FDD method are: 1) it can greatly improve the fault detectability; it is suitable for non-Gaussian electrical drives; 2) based on the improved fault detectability, it can achieve accurate fault diagnosis via support vector machine; and 3) it can be easily applied to electrical drives even if neither physical models or parameters nor expert knowledge of drive systems is given; and it is of highly computational efficiency that can meet requirements on the real-time FDD. A set of experiments on a dSPACE platform-based traction system of the CRH2A-type high-speed train are carried out to demonstrate the effectiveness of the proposed method.

A Study of Probabilistic Diagnosis Method for Three Kinds of Internal Combustion Engine Faults Based on the Graphical Model

A strategy for increasing the accuracy rate of internal combustion engine (ICE) fault diagnosis based on the probabilistic graphical model is proposed. In this method, a three-layer network with inference of probability is constructed, and both the material conditions and the signals collected from different engine parts are considered as the inputs of the system. Machine signals measured by sensors were processed in order to diagnose potential faults, which were presented as probabilities based on the components in layer 1, fault categories in layer 2, and fault symptoms in layer 3. The diagnosis model was built by using nodes and arcs, and the results depended on the connections between the fault categories and symptoms. The parameters of the network represented quantitative probabilistic relationships among all layers, and the conditional probabilities of each type of fault and relevant symptoms were summarized. Fault cases were simulated on a 12-cylinder diesel engine, and three fault types that often occur on ICEs were tested based on five different fault symptoms with different loads, respectively. The diagnostic capability of the method was investigated, reporting high accuracy rates.

An adaptive pre-warning method based on trend monitoring: Application to an oil refining process

The feedstock and products in high-risk processes like oil refining are often flammable, corrosive and explosive, thus any faults that may occur during the operation can lead to significant economic losses and possibly human lives. One expects that the fault symptoms are detected as early as possible by properly monitoring the process status with pre-warning functions to limit unexpected abnormal behaviors. To accomplish this task, an adaptive pre-warning method based on trend monitoring is proposed in this paper. It captures the varying trend of process variables through online monitoring and an adaptive pre-warning scheme is used to provide early warnings when the observable signals are recognized to be non-steady. The proposed method also includes an adaptive weighting scheme to provide a processing priority for variables for which a pre-warning signal is triggered in a short time period. The effectiveness of the method is illustrated with two field case studies taken from oil refining applications.

Periodic impulses extraction based on improved adaptive VMD and sparse code shrinkage denoising and its application in rotating machinery fault diagnosis

Abstract The presence of periodic impulses in vibration signals is a typical symptom of localized faults of rotating machinery. It is of great significance to study how to effectively extract the periodic impulses in vibration signals for realizing the fault diagnosis of rotating machinery. Variational mode decomposition (VMD) provides a feasible tool for non-stationary signal analysis. However, the reasonable selection of algorithm parameters and under- or over-decomposition problem in VMD hinder its application in engineering signals processing to some extent. Therefore, a new periodic impulses extraction method based on improved adaptive VMD and adaptive sparse code shrinkage denoising is proposed for the fault diagnosis of rotating machinery. The method can adaptively determine the mode number and the penalty factor depending on different signals. Meanwhile, the decomposition results are clustered and combined by using the spectrum overlap coefficient and kurtosis index to eliminate the over decomposition phenomenon and realize the effective extraction of the periodic impulses. The adaptive sparse code shrinkage algorithm is developed to denoise the mode component containing the periodic impulses, further highlighting the impulses and improving the accuracy of fault identification. Simulation data and real data acquired from rolling bearing and gearbox are adopted to verify the effectiveness and superiority of the proposed method compared with other methods.

Fault Identification of Chemical Processes Based on k-NN Variable Contribution and CNN Data Reconstruction Methods.

Data-driven fault detection and identification methods are important in large-scale chemical processes. However, some traditional methods often fail to show superior performance owing to the self-limitations and the characteristics of process data, such as nonlinearity, non-Gaussian distribution, and multi-operating mode. To cope with these issues, the k-NN (k-Nearest Neighbor) fault detection method and extensions have been developed in recent years. Nevertheless, these methods are primarily used for fault detection, and few papers can be found that examine fault identification. In this paper, in order to extract effective fault information, the relationship between various faults and abnormal variables is studied, and an accurate “fault–symptom” table is presented. Then, a novel fault identification method based on k-NN variable contribution and CNN data reconstruction theories is proposed. When there is an abnormality, a variable contribution plot method based on k-NN is used to calculate the contribution index of each variable, and the feasibility of this method is verified by contribution decomposition theory, which includes a feasibility analysis of a single abnormal variable and multiple abnormal variables. Furthermore, to identify all the faulty variables, a CNN (Center-based Nearest Neighbor) data reconstruction method is proposed; the variables that have the larger contribution indices can be reconstructed using the CNN reconstruction method in turn. The proposed search strategy can guarantee that all faulty variables are found in each sample. The reliability and validity of the proposed method are verified by a numerical example and the Continuous Stirred Tank Reactor system.

Analysis of the periodic breathing of a transverse annular crack propagated in a real rotating machine

This paper shows the numerical results of a model-based analysis carried out to investigate the experimental dynamic behaviour of the shaft of a real steam turbine in which a deep annular – or circumferential – crack propagated. The shape of this transverse crack was caused by the generation of multiple initiation points, equally spaced over the whole circumferential surface of the shaft cross-section located at the first high pressure stage of the steam turbine.The periodic breathing of the crack caused changes in the area moments of inertia of the damaged cross-section and fault symptoms. Conversely, the shaft heating that affected the turbine during the load rises performed at the end of the runups carried out from a cold state pointed out an abnormal sensitivity of the synchronous vibrations to the machine thermal state. As a result, the fault symptoms caused by this crack were quite different from those induced by common shaft transverse cracks. The breathing of this crack, caused by gravity, has been studied by means of a mathematical model in order to investigate the influence of the crack opening and closure on the harmonic content of the turbine vibration.A new approach is used in this paper, considering a more appropriate 3D FE model that allowed the non-linear distribution of the stresses in the cracked section to be estimated. The method used for the present study has proved that the non-linear distribution of the axial stresses in the area close to the crack tip gives only minor effects on the breathing mechanism. This is an original contribution, which could have not been observed in previous authors' papers that employed simplified models for the periodic breathing.The good accordance between numerical results and experimental data confirms the adequacy of the investigation method used for this study. With reference to diagnostic purposes, this involves that the early on-line detection of annular cracks propagated in the shafts of rotating machinery needs to use more accurate techniques, based on the analysis of vibration data, which can point out fault symptoms, even in the case of very deep transverse cracks.

Fault Prediction via Symptom Pattern Extraction Using the Discretized State Vectors of Multisensor Signals

Fault prediction and early degradation detection have received considerable attention in many engineering disciplines. Fault symptoms can be identified by abnormal values or unusual trends in the monitored sensor signals over a certain period prior to fault occurrence. However, how to extract abnormal pattern, particularly those with conditional relations among multiple sensor signals, remains unclear. Pattern extraction is further difficult particularly when there is no gradient relationship between measurements and operational states due to highly scattered data and unclear boundaries for distinguishing operational states. Additionally, defining the time period for symptom periods is challenging. To resolve these issues, we define the terms symptom pattern and symptom period, and then present a symptom pattern extraction method that collects all evidence of potential fault occurrence from multiple sensor signals. We postulate that, given time markers of fault occurrences, a symptom period precedes the occurrence of a fault. Symptom patterns are defined as either only found in the symptom periods or not found in the given time series, but similar to fault patterns. We further discuss an iterative search procedure for determining the length of symptom periods and propose a severity assessment method for symptom patterns. Finally, we apply the symptom pattern extraction and severity assessment methods to an online fault prediction procedure. By assessing the total severity of patterns in the monitoring window, early warning decision can be made. The procedure is tested in the early detection of abnormal cylinder temperature in a marine diesel engine and automotive gasoline engine knocking.