Identification Of Multiple Faults Research Articles

Adaptive mode decomposition method based on fault feature orientation and its application to compound fault diagnosis of planetary gearboxes

Abstract Planetary gearboxes often experience multiple component failures during service, which can accelerate the degradation and failure of industrial equipment. Accurate separation and identification of multiple faults is an important means of ensuring the safe and stable operation of equipment. However, different faults can interact with each other, along with the influence of background noise, making it challenging to accurately extract faults with relatively weak energy among multiple faults. This difficulty leads to the problems of potential misdiagnosis and underdiagnosis. To address this issue, an adaptive mode decomposition method based on fault feature orientation (AMD-FF) is proposed in this paper. Initially, a fault impact indicator (FII) is constructed based on period-weighted kurtosis of envelope spectral and correlated combination negentropy to effectively characterize the impulsiveness and periodicity of fault features. Furthermore, with the objective of maximizing the FII, an adaptive decomposition of the original signal is designed based on blind convolution theory using a finite-impulse response filter group. Subsequently, a variable weight particle swarm optimization is employed to adaptively optimize the key decomposition parameters. Finally, the data of industrial-grade planetary gear transmission test rig are collected to validate the proposed method for compound fault diagnosis of planetary gearboxes. The results indicate that the AFMD-FF can effectively separate and extract compound faults in planetary gearboxes, demonstrating superior fault separation and diagnostic performance compared to the fault mode decomposition (FMD) and adaptive FMD. This method offers a novel approach to diagnosing compound faults in rotating equipment in industrial scenarios.

Analysis and implementation of certain multivariate statistical process monitoring Tools for fault detection and isolation (FDI) task in a laboratory scale shell-tube heat exchanger

Early detection and diagnosis of faulty events in industrial processes can represent economic, social and environmental profits. When the process has a great quantity of sensors or actuators, the Fault Detection and Isolation (FDI) task is very difficult. Advanced statistical based FDI methods are extensively used for fault detection and isolation purposes. In this work, three multivariate statistical techniques such as neural network based Principal Component Analysis (PCA), neural network-based Fischer Discriminant Analysis (FDA) and Correspondence Analysis (CA) was applied to the multivariate data extracted from laboratory scale shell and tube heat exchanger. Performance metric such as detection delay, estimated time of occurrence of fault, misclassification rate was computed for those three techniques for the detection and isolation of biases in sensors and actuators. Correspondence Analysis was proven to perform better when compared to PCA and FDA. CA was observed to perform FDI with minimal detection delay (which is less than or equal to 7 seconds) and lower misclassification rate (which is less than or equal to 6%) in case of both sensor & actuator faults. PCA and FDA showed significant detection delay and missed alarm rate for single and multiple fault identification.

Geometric Analysis of Signals for Inference of Multiple Faults in Induction Motors.

Multiple fault identification in induction motors is essential in industrial processes due to the high costs that unexpected failures can cause. In real cases, the motor could present multiple faults, influencing systems that classify isolated failures. This paper presents a novel methodology for detecting multiple motor faults based on quaternion signal analysis (QSA). This method couples the measured signals from the motor current and the triaxial accelerometer mounted on the induction motor chassis to the quaternion coefficients. The QSA calculates the quaternion rotation and applies statistics such as mean, variance, kurtosis, skewness, standard deviation, root mean square, and shape factor to obtain their features. After that, four classification algorithms are applied to predict motor states. The results of the QSA method are validated for ten classes: four single classes (healthy condition, unbalanced pulley, bearing fault, and half-broken bar) and six combined classes. The proposed method achieves high accuracy and performance compared to similar works in the state of the art.

Feature extraction of rolling bearing multiple faults based on correlation coefficient and Hjorth parameter

The paper has brought in the Hjorth Complexity parameter and combined it with Intrinsic time decomposition (ITD) algorithm as characteristic parameter index in order to implement accurate identification of multiple faults of rolling bearings. Firstly, concerning about the much uncertainty in manual setting of decomposition layer number in ITD, overdoing of automatic decomposition and the fact that a larger correlation coefficient of signal relates with the greater correlation of signals before and after decomposition and vice versa, the paper has carried out self-adaptive determination of the number of ITD decomposition layers. Secondly, regarding the insensitivity of Hjorth Complexity parameter to noise and the fact that with larger Complexity parameter, signals are simpler and it becomes more available to dig out characteristic information of fault from signals. With Complexity parameter as the index of characteristic parameter, option of optimal Proper rotation component (PRC) is made after ITD. Finally, through the comparison with other methods and the analysis of multiple faults of bearings, it indicates that correlation coefficient can self-adaptively determine the number of ITD decomposition layers and prevent from overdoing and underdoing of decomposition. The Hjorth Complexity parameter can be treated as index parameter to implement optimal PRC option, based on which multiple fault characteristics of bearings can be effectively extracted and the type precisely determined.

Assessment of Dynamic Bayesian Models for Gas Turbine Diagnostics, Part 2: Discrimination of Gradual Degradation and Rapid Faults

There are many challenges that an effective diagnostic system must overcome for successful fault diagnosis in gas turbines. Among others, it has to be robust to engine-to-engine variations in the fleet, it has to discriminate between gradual deterioration and abrupt faults, and it has to identify sensor faults correctly and be robust in case of such faults. To combine their benefits and overcome their limitations, two diagnostic methods were integrated in this work to form a multi-layer system. An adaptive performance model was used to track gradual deterioration and detect rapid or abrupt anomalies, while a series of static and dynamic Bayesian networks were integrated to identify component degradation, component abrupt faults, and sensor faults. The proposed approach was tested on synthetic data and field data from a single-shaft gas turbine of 50 MW class. The results showed that the approach could give acceptable accuracy in the isolation and identification of multiple faults, with 99% detection and isolation accuracy and 1% maximum error in the identified fault magnitude. The approach was also proven robust to sensor faults, by replacing the faulty signal with an estimated value that had only 3% error compared to the real measurement.

Comparative Analysis of Signal Processing Techniques for Fault Detection in Three Phase Induction Motor

Signal processing is considered as an efficient technique to detect the faults in three-phase induction motors. Detection of different varieties of faults in the rotor of the motor are widely studied at the industrial level. To extend further, this research article presents the analysis on various signal processing techniques for fault detection in three-phase induction motor due to the damages in rotor bar. Usually, Fast Fourier Transform (FFT) and STFT are used to analyze the healthy and faulty motor conditions based on the signal characteristics. The proposed study covers the advantages and limitations of the proposed wavelet transform (WT) and each technique for detecting the broken bar of induction motors. The good frequency information can be collected from FFT techniques for handling multiple faults identification in three-phase induction motor. Despite the hype, the detection accuracy gets reduced during the dynamic condition of the machine because the frequency information on sudden time changes cannot be employed by FFT. The WT method signal analysis is compared with FFT to propose fault detection method for induction motor. The WT method is proving better accuracy when compared to all existing methods for signal information analysis. The proposed research work has simulated the proposed method with MATLAB / SIMULINK and it helps to effectively detect the healthy and faulty conditions of the motor.

Transverse Vibration of Geared-Rotor Integrated With Active Magnetic Bearings in Identification of Multiple Faults

Abstract This paper presents a novel concept of the modeling, active control of transverse vibration responses, and identification of fault parameters in a geared-rotor system integrated with active magnetic bearings (AMBs). The sources of error in gears while in the operation are the gear mesh deformation, transmission error, and runout, resulting in dynamic forces, excessive vibration, and noise. To avoid any undesirable effect on the gear-pair and other supporting structures, it is essential to investigate these forced vibrations in time and frequency domain. Hence, an approach to monitor and control the transverse vibration of mating gears is presented with the help of AMBs. The AMBs are capable of suppressing the vibration of the system (transients as well as steady-state) by controlled electromagnetic forces considering the rotor vibrational displacement with a closed-loop feedback system. A mathematical model has been developed with geared rotor faults, like the mesh deformation, gear run-out, and asymmetric transmission error. The transmission error has been modeled as the sum of mean and varying components of error in two orthogonal transverse directions. Based on the mathematical model, an identification algorithm has been developed. Considering full spectrum analysis of the rotor vibration and AMB current information, estimation of system parameters, i.e., the equivalent mesh stiffness, mesh damping, gear runouts, the mean and varying transmission error magnitude and phase angles, and the current and displacement constants of AMBs has been performed. Gaussian noise in responses and modeling errors in mathematical models have been added to test the robustness of the proposed algorithm to comply with the experimental settings.

Multiple faults detection and identification of three phase induction motor using advanced signal processing techniques

In this paper, we have presented the multiple fault detection and identification system for three-phase induction motor. Fast Fourier Transform (FFT) is the most used signal processing technique that offers good frequency information but failing in providing time information and handling multiple faults identification with their occurrence time. FFT also fails to detect non-stationary condition of the signal and unable to convey sudden changes, start and end of the events, drifts and trends. To obtain simultaneous time frequency information and to deal with non-stationary signals Short Time Fourier Transform (STFT) is considered optimal technique that can clearly provide time and frequency information both. In this research work, the multiple fault detection and identification system is presented by employing Short Time Fourier Transform (STFT) signal processing technique. The proposed model is designed using current signature analysis method (CSAM) for three major faults including three phase supply imbalance, single phasing condition and breakage of rotor bars. The system is simulated in MATLAB/SIMULINK and simulation is performed based on healthy and unhealthy conditions of the motor. Comparative analysis between FFT and STFT, shows STFT as a promising approach.

Identification of Multiple Faults in Gearbox Based on Multipoint Optional Minimum Entropy Deconvolution Adjusted and Permutation Entropy.

The weak compound fault feature is difficult to extract from a gearbox because the signal components are complex and inter-modulated. An approach (that is abbreviated as MRPE-MOMEDA) for extracting the weak fault features of a transmission based on a multipoint optimal minimum entropy deconvolution adjustment (MOMEDA) and the permutation entropy was proposed to solve this problem in the present paper. The complexity of the periodic impact signal was low and the permutation entropy was relatively small. Moreover, the amplitude of the impact was relatively large. Based on these advantages, the multipoint reciprocal permutation entropy (MRPE) was proposed to track the impact fault source of the weak fault feature in gearbox compound faults. The impact fault period was indicated through MRPE. MOMEDA achieved signal denoising. The optimal filter coefficients were solved using MOMEDA. It exhibits an outstanding performance for noise suppression of gearbox signals with a periodic impact. The results from the transmission show that the proposed method can identify multiple faults simultaneously on a driving gear in the 4th gear of the transmission.

A Novel Method for Multi-Fault Feature Extraction of a Gearbox under Strong Background Noise.

Strong background noise and complicated interfering signatures when implementing vibration-based monitoring make it difficult to extract the weak diagnostic features due to incipient faults in a multistage gearbox. This can be more challenging when multiple faults coexist. This paper proposes an effective approach to extract multi-fault features of a wind turbine gearbox based on an integration of minimum entropy deconvolution (MED) and multipoint optimal minimum entropy deconvolution adjusted (MOMEDA). By using simulated periodic transient signals with different noise to signal ratios (SNR), it evaluates the outstanding performance of MED in noise suppression and reveals the deficient in extract multiple impulses. On the other hand, MOMEDA can performs better in extracting multiple pulses but not robust to noise influences. To compromise the merits of them, therefore the diagnostic approach is formalized by extracting the multiple weak features with MOMEDA based on the MED denoised signals. Experimental verification based on vibrations from a wind turbine gearbox test bed shows that the approach allows successful identification of multiple faults occurring simultaneously on the shaft and bearing in the high speed transmission stage of the gearbox.

Adaptive Filter for Automatic Identification of Multiple Faults in a Noisy OTDR Profile

We present a novel methodology able to distinguish meaningful level shifts from typical signal fluctuations. A two-stage regularization filtering can accurately identify the location of the significant level-shifts with an efficient parameter-free algorithm. The developed methodology demands low computational effort and can easily be embedded in a dedicated processing unit. Our case studies compare the new methodology with current available ones and show that it is the most adequate technique for fast detection of multiple unknown level-shifts in a noisy OTDR profile.

Identification of multiple faults in rotating machinery based on minimum entropy deconvolution combined with spectral kurtosis

Due to the complexity of mechanical system, multiple faults may co-exist in a rotating machinery, where vibration is commonly used for diagnosis. The measured vibration signal could be considered as a result of convolution process of malfunction induced periodic impact signal and resonant response of the mechanical component, and deconvolution is an effective way to restore impulses. The minimum entropy deconvolution (MED) has been shown to be an effective deconvolution method and has been employed in rotating machinery fault diagnosis. Nevertheless, the simulation in this paper shows that the MED is unable to identify multi-faults of rotating machinery fully when different faults excite different resonance frequencies. To overcome this shortcoming, a new multi-faults detection method based on Spectral kurtosis (SK) and MED is proposed. The effectiveness of the proposed method is validated by simulation data and field signals from a vacuum pump.

Variable Elimination-Based Contribution for Accurate Fault Identification

We propose a new fault identification method, which can describe the contribution of each process variable to a detected fault and identify a faulty variable more accurately than conventional methods. In the proposed method, in addition to a fault detection model that describes normal operating condition (NOC), multiple fault identification models that describe the same NOC are also constructed by eliminating one variable from all monitored variables at a time. After a fault is detected with the fault detection model, the fault detection index, e.g. a combined index of the T2 and Q statistics, is calculated by using each of the fault identification models. When the faulty variable is eliminated, the index does not change before and after the fault occurs. On the other hand, when the normal variable is eliminated, the index is affected by the fault and increases after the fault occurs. Thus, the eliminated variable corresponding to the index that does not increase after the occurrence of the fault is identified as a faulty variable. In the proposed method, the ratio of the average index in NOC to the current index is used as a fault identification index or a contribution. To validate the proposed method, VEC was compared with the reconstruction-based contribution (RBC) through numerical examples. The results have demonstrated that VEC outperformed RBC in fault identification performance both in the linear case and in the nonlinear case.

Multiple Fault Identification Using Vibration Signal Analysis and Artificial Intelligence Methods

Paper addresses the implementation of feature based artificial neural networks and self-organized feature maps with the vibration analysis for the purpose of automated faults identification in rotating machinery. Unlike most of the research in this field, where a single type of fault has been treated, the research conducted in this paper deals with rotating machines with multiple faults. Combination of different roller elements bearing faults and different gearbox faults is analyzed. Experimental work has been conducted on a specially designed test rig. Frequency and time domain vibration features are used as inputs to fault classifiers. A complete set of proposed vibration features are used as inputs for self-organized feature maps and based on the results they are used as inputs for supervised artificial neural networks. The achieved results show that proposed set of vibration features enables reliable identification of developing bearing and gear faults in geared power transmission systems.

Detection of multiple faults in single-frequency differential GPS measurements

For reliable application of GPS, multiple fault detection is one of the most serious concerns demanding a practical solution. Multiple faults occurring at the same time cannot be detected by simple position domain residual detectors or parity vectors since they bear the possibility of cancelling each other. For differential positioning based on single-frequency receivers, this study proposes an efficient fault detection method. The proposed method consists of three different detectors: the range domain signal fault detector, the abnormal ephemeris detector and the double differenced position domain residual detector. Owing to diverse viewpoints provided by the three different detectors, detection and identification of multiple faults is facilitated. To evaluate the performance of the proposed method, simulation and experiment results are demonstrated.

Multiple Faults Identification for Nonlinear Control Systems based on T-S Fuzzy Models

AbstractThe identification of multiple faults in a control system where some sensors and/or actuators may be faulty simultaneously is of great importance for enhancing the control system reliability. In this paper, a novel method for multiple faults identification of nonlinear systems based on T-S fuzzy model and parameter estimation is presented. A series of fully decoupled parity equations sensitive to special groups, which are called the susceptible groups, are designed for the local linear models. Then the residuals are generated by fusing the residuals generated from the parity equations of local linear models using the T-S model. A parameter estimator based on recursive least square is designed to identify the multiple faults from the information contained in the residuals. A simulation example on an induction motor of a railway traction system is given for illustration. Results show that the new approach can be used to identify multiple faults in nonlinear systems.

Fault diagnosis of the polypropylene production process (UNIPOL PP) using ANFIS

The performance of a chemical process plant can gradually degrade due to deterioration of the process equipment and unpermitted deviation of the characteristic variables of the system. Hence, advanced supervision is required for early detection, isolation and correction of abnormal conditions. This work presents the use of an adaptive neuro–fuzzy inference system (ANFIS) for online fault diagnosis of a gas-phase polypropylene production process with emphasis on fast and accurate diagnosis, multiple fault identification and adaptability. The most influential inputs are selected from the raw measured data sets and fed to multiple ANFIS classifiers to identify faults occurring in the process, eliminating the requirement of a detailed process model. Simulation results illustrated that the proposed method effectively diagnosed different fault types and severities, and that it has a better performance compared to a conventional multivariate statistical approach based on principal component analysis (PCA). The proposed method is shown to be simple to apply, robust to measurement noise and able to rapidly discriminate between multiple faults occurring simultaneously. This method is applicable for plant-wide monitoring and can serve as an early warning system to identify process upsets that could threaten the process operation ahead of time.

Experimental identification of multiple faults in rotating machines

The aim of this paper is to define the required measurements and processing tools necessary for developing a maintenance approach applied to rotating machines in the presence of multiple faults. The system responses measured were accelerations and transmission errors. Acceleration measurements provide most of the information on bearing conditions, while transmission error measurements provide pertinent information on gear conditions. The measurements were carried out for several operating conditions (loads and speeds). System responses were processed in several analyzing domains (Time, Spectrum, and Cepstrum domains). The approach developed enables the detection and identification of combined faults and it can be applied to other types of rotating machines once the critical elements and their associated faults have been defined.

Maximal diagnosis of interconnects of random access memories

This paper presents an approach for the maximal diagnosis of all faults (stuck-at, open and short) in the interconnect of a random access memory (RAM); and the interconnect includes data and address lines. This approach accomplishes maximal diagnosis under a complex model in which the lines in the interconnect of the RAM can be affected by multiple faults. Maximal diagnosis consists of detection and location of all diagnosable faults as well as type identification of multiple faults affecting each line. The proposed algorithm (referred to as the Improved Maximal Diagnosis Algorithm, or IMDA) requires max{n,m-1}+n+3 WRITE and max{n,m}+2n READ, where n is the number of address lines and m is the number of data lines. IMDA executes in three different steps: the first step diagnoses the data lines (and in particular the stuck-at faults); the second step accomplishes maximal diagnosis of the shorts (involving either the data lines only, or the data and address lines); and the third step completes the diagnosis of the address lines.

IDENTIFICATION OF MULTIPLE FAULTS IN ROTOR SYSTEMS

Many papers are available in the literature about identification of faults in rotor systems. However, they generally deal only with a single fault, usually an unbalance. Instead, in real machines, the case of multiple faults is quite common: the simultaneous presence of a bow (due to several different causes) and an unbalance or a coupling misalignment occurs often in rotor systems. In this paper, a model-based identification method for multiple faults is presented. The method requires the definition of the models of the elements that compose the system, i.e., the rotor, the bearings and the foundation, as well as the models of the faults, which can be represented by harmonic components of equivalent force or moment systems. The identification of multiple faults is made by a least-squares fitting approach in the frequency domain, by means of the minimization of a multi-dimensional residual between the vibrations in some measuring planes on the machine and the calculated vibrations due to the acting faults. Some numerical applications are reported for two simultaneous faults and some experimental results obtained on a test-rig are used to validate the identification procedure. The accuracy and limits of the proposed procedure have been evaluated.