Fault Signal Detection Research Articles

A pattern recognition system based on acoustic signals for fault detection on composite materials

Abstract The use of composite materials in industry applications is constantly growing. However, fault detection and prediction on these materials is not as simple as in traditional materials. Thus, the development of a methodology for fault detection is strictly necessary to ensure the integrity of a structure. This paper proposes a pattern recognition system that implements an Artificial Neural Network classifier to detect and classify damage on composite beams. Classifiers were trained and tested using acoustic signals emitted by healthy and damaged beams after an impulsive load was applied to them. Singular Value Decomposition was used to filter the acoustic signals whereas Principal Component Analysis was implemented to extract relevant information from the filtered signal. The extracted information was used as inputs to the classifier that was able to predict four different levels of damage on glass fiber and carbon fiber beams with more than 97% accuracy. These results suggest that the proposed methodology can be further investigated to develop a robust system for automatic detection of damage on composite structures.

Fault Detection of High-Speed Train Wheelset Bearing Based on Impulse-Envelope Manifold

A novel fault detection method employing the impulse-envelope manifold is proposed in this paper which is based on the combination of convolution sparse representation (CSR) and Hilbert transform manifold learning. The impulses with different sparse characteristics are extracted by the CSR with different penalty parameters. The impulse-envelope space is constructed through Hilbert transform on the extracted impulses. The manifold based on impulse-envelope space (impulse-envelope manifold) is executed to learn the low-dimensionality intrinsic envelope of vibration signals for fault detection. The analyzed results based on simulations, experimental tests, and practical applications show that (1) the impulse-envelope manifold with both isometric mapping (Isomap) and locally linear coordination (LLC) can be successfully used to extract the intrinsic envelope of the impulses where local tangent space analysis (LTSA) fails to perform and (2) the impulse-envelope manifold with Isomap outperforms those with LLC in terms of strengthening envelopes and the number of extracted harmonics. The proposed impulse-envelope manifold with Isomap is superior in extracting the intrinsic envelope, strengthening the amplitude of intrinsic envelope spectra, and enlarging the harmonic number of fault-characteristic frequency. The proposed technique is highly suitable for extracting intrinsic envelopes for bearing fault detection.

Recursive observer‐based fault detection for a class of nonlinear uncertain systems with output constraints

SummaryThis paper investigates the fault detection (FD) problem for a class of nonlinear uncertain systems in strict feedback form with an output constraint. The key idea is to design an observer to generate the FD signals and the output estimate, which also satisfies the output constraint. To facilitate constraint handling, the constraints on the output and the output estimate are transformed into the output estimation error constraint. Then, the FD observer is designed in a recursive framework. By employing a barrier Lyapunov function, the output estimation error constraint is incorporated in the last step of the recursive observer design algorithm to prevent constraint violation. It is shown that the output estimation error is uniformly bounded and satisfies the constraint for the fault‐free case. Furthermore, the residual signal is constructed by the output estimation error, and its corresponding bound is used as threshold. Compared with the FD method without considering the constraints, the proposed FD scheme provides a smaller threshold and characterizes a larger set of faults, which can be detected. Finally, simulation results are presented to illustrate the benefits of the proposed FD scheme. Copyright © 2016 John Wiley & Sons, Ltd.

Model-based real-time thermal fault diagnosis of Lithium-ion batteries

Ensuring safety and reliability is a critical objective of advanced Battery Management Systems (BMSs) for Li-ion batteries. In order to achieve this objective, advanced BMS must implement diagnostic algorithms that are capable of diagnosing several battery faults. One set of such critical faults in Li-ion batteries are thermal faults which can be potentially catastrophic. In this paper, a diagnostic algorithm is presented that diagnoses thermal faults in Lithium-ion batteries. The algorithm is based on a two-state thermal model describing the dynamics of the surface and the core temperature of a battery cell. The residual signals for fault detection are generated by nonlinear observers with measured surface temperature and a reconstructed core temperature feedback. Furthermore, an adaptive threshold generator is designed to suppress the effect of modelling uncertainties. The residuals are then compared with these adaptive thresholds to evaluate the occurrence of faults. Simulation and experimental studies are presented to illustrate the effectiveness of the proposed scheme.

Phase‐space‐based symmetrical fault detection during power swing

This study proposes an effective symmetrical fault detection technique during power swing in distance relaying using phase-space (PS) method. This is based on an embedding theorem which shows that a time series can be mapped to a higher-dimensional space called PS through embedding. Thus, a sampled signal can be transformed into the PS so that its features can be more clearly viewed. The process starts at retrieving the current signal samples and process through PS, the original signal is decoupled into two parts such as power swing and symmetrical fault part. The symmetrical fault detection during power swing is easily achieved as the symmetrical fault part of the signal produces an irregular shape compared with the shape produced during from the power swing condition. The symmetrical fault detection signal using PS (FDPS) detects the symmetrical fault during power swing in the transmission line with wide variations in operating conditions and the results from extensive studies indicate that the proposed FDPS can reliably detect the faults during power swing.

Characterisation of Major Fault Detection Features and Techniques for the Condition-Based Monitoring of High-speed Centrifugal Blowers

This paper investigates and characterises the major fault detection signal features and techniques for the diagnostics of rotating element bearings and air leakage faults in high-speed centrifugal blowers. The investigation is based on time domain and frequency domain analysis, as well as on process information, vibration, and acoustic emission fault detection techniques. The results showed that the data analysis method applied in this study is effective, as it yielded a detection accuracy of 100%. A lookup table was compiled to provide an integrated solution for the developer of Condition-Based Monitoring (CBM) applications of centrifugal blowers. The major contribution of this paper is the integration and characterisation of the major fault detection features and techniques.

The joint optimal filtering and fault detection for multi-rate sensor fusion under unknown inputs

In multi-sensor fusion, it is hard to guarantee that all sensors have an identical sampling rate, especially in the distributive and/or heterogeneous case. Meanwhile, stochastic noise, unknown inputs (UIs), and faults may coexist in complex environment. To this end, we propose the problem of joint optimal filtering and fault detection (FD) for multi-rate sensor fusion subject to UIs, stochastic noise with known covariance, and faults imposed on the actuator and sensors. Furthermore, the new scheme of optimal multi-rate observer (MRO) is presented and applied to detect faults. The observer parameters are determined optimally in pursuit of the UI decoupling and maximizing noise attenuation under the causality constraint due to multi-rate nature. Finally, the output estimation error of the MRO is used as a residual signal for FD via a hypothesis test in which the threshold is adaptively designed according to the MRO parameters. One numerical example is given to show the effectiveness of our proposed method.

Application of an improved dynamic time synchronous averaging method for fault diagnosis in conditions of speed fluctuation and no tachometer

Time synchronous averaging has been widely used for machinery fault diagnosis. However, it cannot reveal signal characteristics accurately in conditions of speed fluctuation and no tachometer due to the phase accumulation error. In this paper, an improved dynamic-time synchronous averaging method is proposed to extract the periodic feature signal from the fluctuated vibration signal for fault detection when no tachometer signal is available. In this method, empirical mode decomposition, dynamic time warping, and time synchronous averaging are performed on gear vibration signals to detect fault characteristic information. First, empirical mode decomposition is performed on the vibration signal and a series of intrinsic mode functions are produced. The sensitive intrinsic mode functions providing fault-related information are selected and reconstructed and the corresponding envelop signals are equal-space intercepted. Then, the phase accumulation error among the envelop signal segments is estimated by the dynamic time warping, which is further used to compensate the phase accumulation error between the intrinsic mode function segments of the reconstructed signal. Finally, the compensated intrinsic mode function segments are averaged to obtain the feature signal. Simulation analysis shows the advantages of the proposed method in extracting faulty feature signal from speed fluctuation signal without tachometer and identifying gear fault. Experiments with both normal and faulty gear were conducted and the vibration signals were captured. The proposed method is applied to identify the gear damage and the diagnosis results demonstrate its superiority than other methods.

Normalization and source separation of acoustic emission signals for condition monitoring and fault detection of multi-cylinder diesel engines

A signal processing technique is presented in this paper to normalize and separate the source of non-linear acoustic emission (AE) signals of a multi-cylinder diesel engine for condition monitoring applications and fault detection. The normalization technique presented in the paper overcomes the long-existing non-linearity problem of AE sensors so that responses measured by different AE sensors can be quantitatively analysed and compared. A source separation algorithm is also developed in the paper to separate the mixture of the normalized AE signals produced by a multi-cylinder diesel engine by utilising the system parameters (i.e., wave attenuation constant and the arrival time delay) of AE wave propagation determined by a standard pencil lead break test on the engine cylinder head. It is shown that the source separation algorithm is able to separate the signal interference of adjacent cylinders from the monitored cylinder once the wave attenuation constant and the arrival time delay along the propagation path are known. The algorithm is particularly useful in the application of AE technique for condition monitoring of small-size diesel engines where signal interference from the neighbouring cylinders is strong.

Fault Signal Detection System of Large Scale Integrated Circuit Based on Single Chip Microcomputer

In the process of fault signal detection of large-scale integrated circuit, the fault signal detection can improve the working performance of large-scale integrated circuit. The traditional detection system is simple, time-consuming, and the error is large, the fault signal detection haslow accuracy. In view of this situation, a new design method of large scale integrated circuit fault signal detection system is proposed based on single chip microcomputer. S3C2410 is taken as hardware design basis, and the signal sensor is used, the software algorithm uses three B spline wavelet transform for ORB wave detection method. The detection is completed.The experiment experimental results show that the system can improve the detection accuracy greatly, and effectively improve the work efficiency.

Fault Signal Detection Model for Large-Scale Circuit Communication System

In the process of fault signal detection for large-scale circuit communication systems, with traditional methods to process detection, the fault detection method is more conservative. A fault signal detection for large-scale circuit communication system based on QRS wave group detection method is proposed. The signal to be measured is transformed appropriately in the time domain or frequency domain to strengthen or separate the QRS component, in order to suppress interference from various noise to signals, and the fault point of circuit communication system fault signal is identified, the filter is utilized as representative to process multiscale decomposition for fault signals of circuit communication systems. Experiments show that QRS wave group detection method can determine the occurrence time of the circuit system fault signal, and further estimate the nature of the fault signal, thus, the fault point of communication system fault signal is found to improve the efficiency of detection.

Model-based robust estimation and fault detection for MEMS-INS/GPS integrated navigation systems

In micro-electro-mechanical system based inertial navigation system (MEMS-INS)/global position system (GPS) integrated navigation systems, there exist unknown disturbances and abnormal measurements. In order to obtain high estimation accuracy and enhance detection sensitivity to faults in measurements, this paper deals with the problem of model-based robust estimation (RE) and fault detection (FD). A filter gain matrix and a post-filter are designed to obtain a RE and FD algorithm with current measurements, which is different from most of the existing priori filters using measurements in one-step delay. With the designed filter gain matrix, the H-infinity norm of the transfer function from noise inputs to estimation error outputs is limited within a certain range; with the designed post-filter, the residual signal is robust to disturbances but sensitive to faults. Therefore, the algorithm can guarantee small estimation errors in the presence of disturbances and have high sensitivity to faults. The proposed method is evaluated in an integrated navigation system, and the simulation results show that it is more effective in position estimation and fault signal detection than priori RE and FD algorithms.

Accurate Detection of Fault Signal in Large-Scale Communication

Deep data-mining methods of fault signal in large-scale communication system are researched. Although with the characteristic of frequency uniformity as signals distribute in each reaction zone, common method of fault signal detection based on shortwave dispersing is invalid employing in large-scale communication system, which presents the absence or instability of fault signal. For this reason, a method based on particle swarm optimization is proposed for fault signal detection in large-scale communication system. As reaction speed and activity scope within the whole particle swarm are replaced, accurate results are achieved. Taking particle swarm optimization, it is detected that whether there is a fault in communication systems. The experimental results show that proposed method in signal fault detection process can greatly increase accuracy of signal fault detection, as plays a greater role in future.

Fault detection in an induction motor drive based on the behavior of the power electronics devices

In this work the fault diagnosis technique is applied to the one phase induction motor drive system. To carry out the fault detection, the voltage and current signals obtained directly from the converter behavior (residues) are used. The faults of interest to analyze can happen when control signals of the switches that compose the converter do vary. These faults are designated as lost of commutation sequence.

Approach Signal for Rotor Fault Detection in Induction Motors

In this paper, two approach signals are used for broken rotor bar fault diagnosis. One is based on the spectrum analysis, such as the fast Fourier transform, which utilizes the steady-state spectral components of the stator quantities. The accuracy of this technique depends on the loading conditions and constant speed of the machine. The second approach is based on the discrete wavelet transform which is considered an ideal tool for this purpose due to its suitability for the analysis of signals, the frequency spectrum of which is variable in time. These two approaches are tested in simulation and validated experimentally.

확률분포추정기법을 이용한 와이어로프의 결함진단

A large number of wire rope has been used in various inderstiries as Cranes and Elevators from expanding the scale of the industrial market. But now, the management of wire rope is used as manually operated by rope replacement from over time or after the accident.It is caused to major accidents as well as economic losses and personal injury. Therefore its time to need periodic fault diagnosis of wire rope or supply of real-time monitoring system. Currently, there are several methods has been reported for fault diagnosis method of the wire rope, to find out the feature point from extracting method is becoming more common compared to time wave and model-based system. This method has implemented a deterministic modeling like the observer and neural network through considering the state of the system as a deterministic signal. However, the out-put of real system has probability characteristics, and if it is used as a current method on this system, the performance will be decreased at the real time. And if the random noise is occurred from unstable measure/experiment environment in wire rope system, diagnostic criterion becomes unclear and accuracy of diagnosis becomes blurred. Thus, more sophisticated techniques are required rather than deterministic fault diagnosis algorithm. In this paper, we developed the fault diagnosis of the wire rope using probability density estimation techniques algorithm. At first, The steady-state wire rope fault signal detection is defined as the probability model through probability distribution estimate. Wire rope defects signal is detected by a hall sensor in real-time, it is estimated by proposed probability estimation algorithm. we judge whether wire rope has defection or not using the error value from comparing two probability distribution.

Fault Signal Classification using Adaptive Boosting Algorithm

In recent years, researchers seldom investigate how to boost the classification performance of any learning algorithm for fault signal detection. We propose a fault signal classification method based on adaptive boosting (adaboost) in this paper. Adaboost is able to select an optimal linear combination of classifiers to form an ensemble whose joint decision rule has relatively high accuracy on the training set. First, we extract statistical features from sample signals. And then we make use of a decision tree to identify optimal features, which are used to classify the sample set by adaboost algorithm. To verify its accuracy, we set up the roller bearing experiment. Practical results show that the method can precisely identify fault signals, and be comparable to SVM based traditional method.DOI: http://dx.doi.org/10.5755/j01.eee.18.8.2635

Gear fault detection using adaptive morphological gradient lifting wavelet

Wavelet transform is one of the most acceptable tools to analyze vibration signals for gear fault detection. However, there are still some limitations of the traditional wavelet transforms due to the utilization of fixed linear filters. This investigation presents an adaptive morphological gradient lifting wavelet (AMGLW) to remedy the shortcomings of traditional wavelet transform schemes. A novel nonlinear filter, named morphological gradient filter, is designed for enhancing the impulsive features of the original signal. Then the adaptability of AMGLW is implemented by selecting between two filters, namely the average filter and the morphological gradient filter, to update the approximation signal dependent upon the local gradient of the analyzed signal. This new scheme is evaluated on a simulated signal and a practical vibration signal measured from a gearbox. Experimental results demonstrate that the presented AMGLW outperforms the traditional linear wavelet (LW) transform obviously for detecting gear defects. Furthermore, the computational cost of AMGLW is much less than the traditional LW. Thus the AMGLW scheme is quite suitable for the online condition monitoring of gears.

Imbalance Fault Detection of Direct-Drive Wind Turbines Using Generator Current Signals

Imbalance faults constitute a significant portion of all faults in wind turbine generators (WTGs). WTG imbalance fault detection using generator current measurements has advantages over traditional vibration-based methods in terms of cost, implementation, and system reliability. However, there are challenges in using current signals for imbalance fault detection due to low signal-to-noise ratio of the useful information in current signals and nonstationary characteristic frequencies of imbalance faults. This paper proposes a method of using generator stator currents for imbalance fault detection of direct-drive WTGs. In the proposed method, the variable shaft rotating frequency of a WTG is estimated from one phase stator current measured from the generator terminal by using a phase-locked loop method. The estimated shaft rotating frequency is then processed by using appropriate upsampling and variable-rate downsampling algorithms. Consequently, the variable characteristic frequencies of imbalance faults in the spectrum of the estimated shaft rotating frequency are converted to constant values. Therefore, the signatures of wind turbine imbalance faults can be clearly identified from power spectral density analysis of the converted shaft rotating frequency signal. Simulation and experimental results show that the proposed method is effective to detect various imbalance faults in direct-drive WTGs.

Shannon wavelet spectrum analysis on truncated vibration signals for machine incipient fault detection

Although a variety of methods have been proposed in the literature for machine fault detection, it still remains a challenge to extract prominent features from random and nonstationary vibratory signals, a typical representative of which are the resonance signatures generated by incipient defects on the rolling elements of ball bearings. Due to its random and nonstationary nature, the involved signal generally possesses a low signal-to-noise ratio, where the classical signal processing methods cannot be effectively applied and the extracted features are usually submerged into the severe background noise. In this paper, a novel random and nonstationary vibratory signature analysis (R&N-VSA) technique is presented to address this challenge. The original vibration signal is decomposed into fault-related and non-fault-related signal segments, and multi-level exponential moving average power filtering is suggested to guide this decomposition. Instead of analyzing the whole vibratory signal, the developed Shannon wavelet spectrum analysis is more efficiently applied on the truncated fault-related signal segments so as to enhance the features' characteristics. The effectiveness of the proposed technique is examined through a series of tests with two experimental setups, and the investigation results show that the developed R&N-VSA technique is an effective signal processing technique for incipient machine fault detection.