Fault Signal Detection Research Articles

Fault Diagnosis Method of the Construction Machinery Hydraulic System Based on Artificial Intelligence Dynamic Monitoring

This paper aims to study the fault diagnosis method of the mechanical hydraulic system based on artificial intelligence dynamic monitoring. According to the characteristics of functional principal component analysis (FPCA) and neural network in the fault diagnosis method in the feature extraction process, the fault diagnosis method combining functional principal component analysis and BP neural network is studied and it is applied to the fault of the coordinator hydraulic system diagnosis. This article mainly completed the following tasks: analyzing the structure and working principle of the mechanical hydraulic system, studying the failure mechanism and failure mode of the mechanical hydraulic system, summarizing the common failures of the hydraulic system and the individual failures of the mechanical hydraulic system, and establishing the mechanical hydraulic system. Description of failure mode and effects analysis (FMEA): then, a joint simulation model of the mechanical hydraulic system was established in ADAMS and AMESim, and the fault detection signal of the hydraulic system was determined and compared with the experimental data. At the same time, the simulation data of the cosimulation model were compared with the simulation data of the hydraulic model in MATLAB to further verify the correctness of the model. The functional principal component analysis is used to perform functional processing on sample data, feature parameters are extracted, and the BP neural network is used to train the mapping relationship between feature parameters and fault parameters. The consistency is verified, and the fault diagnosis method is finally completed. The experimental results show that the diagnostic accuracy rates are 0.9848 and 0.9927, respectively, the reliability is significantly improved, close to 100%, and the uncertainty is basically 0, which significantly improves the accuracy of fault diagnosis.

Rotating machinery weak fault signal detection method based on QPSO and stochastic resonance

Due to the weak early vibration signal of rotating machinery, which is nonlinear and non-stationary, the traditional filtering method has some limitations. To this end, a weak signal detection method combining Quantum Particle Swarm Optimization (QPSO) and Stochastic Resonance (SR) is proposed. First, according to the adiabatic approximation theory, make the original signal meet the SR requirements, and optimize the structural parameters of the SR through QPSO; secondly, the signal-to-noise ratio is used as the objective function to make the input signal and the output signal achieve the best match; finally, simulation and experimental data are used to verify the feasibility of the proposed method. Comparative analysis shows that the proposed method increases the amplitude by 46% compared with the characteristic frequency obtained by the PSO optimized SR model.

An adaptive cumulative sum based method for unblocking distance relays in TCSC compensated transmission lines

Fault detection in transmission lines with thyristor controlled series compensators (TCSC) can be challenging, in particular during power swing conditions. A reliable unblocking method based on the adaptive cumulative Sum (ACUSUM) is presented in this paper. This method alleviates the limitations of conventional methods and has the capability to detect all asymmetric and symmetric faults with high accuracy and speed. The proposed method measures the negative sequence of the current signal for fault detection. Two simulation case studies consisted of the single machine and infinite bus (SMIB) and the IEEE standard 9-bus and three machine systems are simulated in PSCAD/EMTDC to evaluate the proposed method. It is shown that the non-linear behavior of TCSC will not impact the accurate performance of this method.

Compressive Sensing-Based Missing-Data-Tolerant Fault Detection for Remote Condition Monitoring of Wind Turbines

Compared with traditional onsite wind turbine condition monitoring systems (CMSs), the remote CMSs can use better computational resources to process data with more advanced algorithms and, thus, can provide more advanced condition monitoring capabilities, but may suffer from a data loss problem, especially when wireless data transmission is used. To solve this problem, this article proposes a compressive sensing-based missing-data-tolerant fault detection method for remote condition monitoring of wind turbines. First, the condition monitoring signals collected from wind turbines are conditioned to increase their sparsity. Then, a compressive-sensing-based sampling algorithm is designed to sample the conditioned signals. The resulting data samples, called measurements of the conditioned signals are transmitted wirelessly during which some data samples are possibly lost. At the data receiving end, the conditioned signals are reconstructed from the received data samples, which might be incomplete, via a compressive-sensing-based signal reconstruction algorithm. Finally, spectrum analysis is performed on the reconstructed signals for wind turbine fault detection via fault characteristic frequency identification. The proposed method is validated for bearing fault detection of a Skystream 3.7 wind turbine and an Air Breeze wind turbine by using the data of a generator current signal collected from each wind turbine remotely while considering different data loss rates.

Sinusoidal FM patterns of fault-related vibration signals for planetary gearbox fault detection under non-stationary conditions

Vibration signals related to planetary gearbox faults under non-stationary conditions will be equipped with sinusoidal modulation laws in their amplitude modulation (AM) and frequency modulation (FM) modes. Extracting fault features implied in the AM and FM modes is the key issue to detect inner gear faults occurring in planetary gearboxes. However, there exist few signal analysis approaches that are able to track the fast oscillating nature implied in the fault-induced sinusoidal FM patterns. To address such a challenging issue, this paper conducts a detailed study on sinusoidal FM patterns of the fault-related vibration signal model and proposes a novel method to extract them successfully. To be specific, the proposed method firstly separates the AM and FM modes via Hilbert transform to avoid their mutual interference. The intrinsic modulation features implied in the AM and FM modes can then be separately extracted by a developed signal decomposition method named iterative-joint chirp mode decomposition (I-JCMD) with a composite mode optimization scheme. Compared with the existing fault detection techniques, the proposed method is not only able to avoid the complicated sideband analysis but also identify the fast and time-varying oscillating nature of the FM mode with a high accuracy. The performance of our method is finally verified by three simulated cases and four groups of real vibration signals related to four types of sun gear faults occurring in a planetary gearbox test-rig, which also reveals a potential scaling relation between load strength and the oscillating coefficient of the FM mode.

Detecting Partial Demagnetization in AFPM Generators by Monitoring Speed and EMF Induced in a Supplemental Winding

In this article, a novel method to detect the partial demagnetization in air-cored permanent-magnet axial-flux generators is proposed. Its principle is to monitor a speed-normalized fault detection signal. This signal depends only on the demagnetization severity, as it was extracted using a flux-sensor with a span that negates the induced voltage under balanced conditions. The generator has been designed for marine renewable applications working under variable speed conditions rending PM health monitoring under nonstationary conditions imperative. Initially, a mathematical equation for the electromotive force is derived under partial demagnetization conditions with the aim to extract the frequency components related to the fault. To estimate the demagnetization severity, an approach was developed based on the magnetic flux monitoring, which penetrates throughout the sensor. A 3-D finiteelement model of the installed winding in the generator is employed to verify the proposed method under steady and variable speed conditions.

The intermittent fault diagnosis of analog circuits based on EEMD-DBN

Aiming at the intermittent faults widely existing in analog circuits, caused by poor soldering and performance degradation of components, a method for the intermittent fault diagnosis of analog circuits based on ensemble empirical mode decomposition (EEMD) and deep belief network (DBN) is proposed. The amplitude and frequency anomaly signals in the frequency domain is regarded as triggering signals for intermittent fault detection, so the signal segments containing intermittent faults can be detected. The signal segment with intermittent fault is decomposed into multiple intrinsic mode functions (IMFs) by EEMD, and the IMF and fault features are optimized by using Pearson correlation coefficient and feature separability. The optimization feature set is established, which is used as feature vector to transmitted to DBN for fault diagnosis. The proposed method can autonomously realize the feature selection and the diagnosis of intermittent fault. The optimization features make DBN improve diagnostic accuracy, reduce diagnostic time, and can locate intermittent faults to the circuit branch with intermittent faults. The simulation experiments show that the proposed method has strong fault diagnosis capability. And the comparative experiments prove that compared with other commonly used methods, the EEMD-DBN method has higher diagnostic accuracy in intermittent fault diagnosis of analog circuits.

Real-Time Abnormal-Signal Detection Under a Noisy Environment Using a Resonance Filter

In this article, we present improved short-time Fourier transform for fault detection of field coils, among the most important parts in rotating machinery. It is generally difficult to detect a fault signal of a field coil due to the low signal-to-noise ratio (SNR) associated with these signals. To solve this problem, we constructed a parallel resonance circuit using the inductance of the field coil and improved the SNR of the fault detection signal by using the bandpass filter characteristics of the resonance circuit. The transfer function curve upon parameter changes of the proposed resonance circuit is presented, and the governing equation is derived so that the resonance frequency can be calculated. We also carried out a fault detection test to verify the feasibility of the proposed method. As a result, we demonstrated that the proposed method can enhance the SNR and enable fault detection using the extracted abnormal signal. Also, it was confirmed that the proposed system responds instantaneously to an abnormal signal for fault detection. Therefore, it is proved that the methodology presented in this article is efficient for fault detection in rotating machinery.

Modulated signal detection method for fault diagnosis

Envelope analysis is a dominant approach in detecting modulated fault signals in rotating machinery. The Hilbert transform is a popular method used to obtain envelopes. However, envelopes based on the Hilbert transform are only suitable for narrowband signals. Signals are usually filtered before envelope demodulation, which leads to the challenging issue of filter selection. A modulated fault signal detection method using the upper and lower envelopes of the signals is proposed. This method can demodulate the modulated fault signals directly from the raw data, which can include wideband and narrowband signals. The proposed method uses a simple algebraic operation instead of a transform function, so it is convenient in practical applications. It is tested with some simulated signals and some real-measured data. The test results show that the method can effectively demodulate a modulated fault signal directly from the raw data.

Transient feature identification from internal encoder signal for fault detection of planetary gearboxes under variable speed conditions

Due to the harsh circumstances, planetary gearboxes as the important transmission component of mechanical equipment inevitably incur unexpected failures. Due to the merits of encoder signal, the research of encoder signal for fault detection have recently received much attention. However, the research on how to apply encoder signal to the fault detection of planetary gearbox under variation speed conditions is still not sufficient. Therefore, this paper proposes a three-stage variable speed encoder signal analysis approach for extracting fault-related transient features in original encoder signal and detecting potential fault of planetary gearboxes. In the proposed method, we first employ difference method to original encoder signal to convert it into more meaningful instantaneous angular speed (IAS) in time domain. Then, self-resampling technique is introduced to transform IAS in time domain into the new one in angular domain. At last, low-pass filter and sparsity based algorithm (LpfSpaA) is established for separating fault-induced IAS fluctuation from noisy IAS in angular domain. In the proposed LpfSpaA that is the core of the paper, a unique convex optimization problem is constructed and an iterative algorithm is derived to solve it. Meanwhile, to obtain the perfect performance of proposed LpfSpaA, an adaptive parameter determination scheme is also analyzed. The efficacy of proposed method in feature extraction and fault detection is assessed using synthetic and actual signals.

Versatile relaying algorithm for detection and classification of fault on transmission line

This article presents a new and fast approach for detection and classification of the fault on a transmission line. The proposed method can effectively detect and classify a fault on a regular uncompensated transmission line. Moreover, the same can detect and classify faults when these lines are subjected to fixed series compensation, with the same accuracy and without any functional customization of the algorithm. Therefore, this method gets rid of the new setting requirement for the inclusion of series compensation. Vice versa can provide protection to a series compensated line during maintenance as well as in the eventuality of bypass of the compensator. This makes the developed algorithm versatile. The proposed algorithm normally converges for fault detection within half cycle from the fault inception, and requires no data after generating fault detection signal for fault classification; which enables the application of faster circuit breaking devices. The two-stage approach uses Wavelet Transform (WT) with Chebyshev Neural Network (ChNN) and uses only measured three-phase current signals at relaying end. The accuracy, speed, and effectiveness of the scheme have been verified with a fault data generation system developed on PSCAD/EMTP with different system parameter variations like fault resistance, load angle, fault inception angle, and types of faults. The results obtained show that the proposed scheme is accurate and fast as well.

An efficient acoustic-based diagnosis of inter-turn fault in interior mount LSPMSM

This paper investigates the use of acoustic signals in diagnosing stator inter-turn faults in Line Start Permanent Magnet Synchronous Motors (LSPMSMs). Singular spectrum analysis (SSA) is proposed to analyze the acoustic signal for fault detection. The proposed methodology involves collection of experimental acoustic data using a smartphone from an interior mount LSPMSM under different loading levels, healthy as well as faulty motor cases where different number of turns are shorted to emulate different fault levels. Acoustic signal for each case is analyzed using SSA, which results in decomposition of the signal into periodic components and noise. Fast Fourier Transform (FFT) of each component is carried out to find the unique frequency representative for each case. The results with the proposed methodology show that it is capable to detect the occurrence of inter-turn fault under different loading levels with the capability of distinguishing fault severity with readily available acoustic sensor of a smart phone. In addition, it is possible to differentiate between the load and no-load modes of operation.

A typical stochastic resonance in Landau-type potential field

Considering the significant influence of the asymmetric structure of the one-dimensional potential field on the stochastic resonance (SR), the research is extended to the two-dimensional asymmetric potential field. The Brownian particles moving in the Landau-type potential field served as the research object. According to the linear response theory and probability flow method, the analytical formula for the spectrum power amplification (SPA) of the physical system concerning the driving frequency is derived. When the SPA is used as a measurement index, the theoretical analysis shows that an atypical SR phenomenon occurs in the system. The reason is that as the noise intensity increases, the response mode changes due to the asymmetric structure of the Landau-type potential field. Moreover, the linear responses and the SPA are inversely proportional to the driving frequency. The conclusions of the theoretical analysis are verified by numerical simulation. Finally, the Landau-type potential system is applied to the bearing fault signal detection. The experimental results show that the Landau-type potential system can effectively diagnose the faults of the inner and outer ring of the bearings, which proves the application value of the Landau-type potential system.

Fault-Tolerant Model Predictive Control Algorithm for Path Tracking of Autonomous Vehicle.

The fault detection and isolation are very important for the driving safety of autonomous vehicles. At present, scholars have conducted extensive research on model-based fault detection and isolation algorithms in vehicle systems, but few of them have been applied for path tracking control. This paper determines the conditions for model establishment of a single-track 3-DOF vehicle dynamics model and then performs Taylor expansion for modeling linearization. On the basis of that, a novel fault-tolerant model predictive control algorithm (FTMPC) is proposed for robust path tracking control of autonomous vehicle. First, the linear time-varying model predictive control algorithm for lateral motion control of vehicle is designed by constructing the objective function and considering the front wheel declination and dynamic constraint of tire cornering. Then, the motion state information obtained by multi-sensory perception systems of vision, GPS, and LIDAR is fused by using an improved weighted fusion algorithm based on the output error variance. A novel fault signal detection algorithm based on Kalman filtering and Chi-square detector is also designed in our work. The output of the fault signal detector is a fault detection matrix. Finally, the fault signals are isolated by multiplication of signal matrix, fault detection matrix, and weight matrix in the process of data fusion. The effectiveness of the proposed method is validated with simulation experiment of lane changing path tracking control. The comparative analysis of simulation results shows that the proposed method can achieve the expected fault-tolerant performance and much better path tracking control performance in case of sensor failure.

Vibration signal fusion using improved empirical wavelet transform and variance contribution rate for weak fault detection of hydraulic pumps

This paper presents a novel vibration signal fusion algorithm using improved empirical wavelet transform and variance contribution rate to fuse three-channel vibration signals for weak fault detection of hydraulic pumps. Firstly, empirical wavelet transform (EWT) is utilized to decompose the three-channel signals into several AM–FM components. Then in accordance with the statistical characteristics of these component data, variance contribution rate is defined to measure the weight of component data points. A series of fusion coefficients are computed and assigned to every component point. Finally, these component points are fused into one single signal and Hilbert transform is conducted to demodulate the fault characteristic frequency for weak fault detection. Moreover, to address the issue of improper EWT spectrum segmentation, we introduce Density-Based Spatial Clustering of Applications with Noise (DBSCAN) to improve EWT in the full space and the frequencies corresponding to outlier points are taken as the boundaries of spectrum segmentation. Therefore, the number of boundaries is more reasonable and the AM–FM components are more consistent with inherent components existing in the vibration signals of pumps. Results of simulation and experiment analysis demonstrate the good performance of the exhibited fusion algorithm in weak fault detection of hydraulic pumps.

An adaptive empirical mode decomposition and stochastic resonance system in high efficient detection of terahertz radar signal

Interference noise affects the ranging capability of terahertz radar signal detection. Aiming at the above problem, a method based on adaptive empirical mode decomposition and stochastic resonance system is proposed. With the preprocessing method of twice sampling, adaptive empirical mode decomposition, stochastic resonance system, and scale recovery, the optimal intrinsic mode function can be screened automatically, the optimal parameters can be obtained automatically and the ranging calculation can be completed. Experimental results demonstrate the effectiveness and superiority of the proposed adaptive system in achieving weak fault signal detection under heavy background noise.

Data processing and augmentation of acoustic array signals for fault detection with machine learning

This paper proposes a novel method to detect and localize disturbances in dynamical systems using non-collocated acoustic arrays by processing their data such that established artificial intelligence techniques can be used. Methods are proposed to compress time-domain based separate microphone data into frequency-domain based images. Furthermore, a data augmentation solution is given to significantly reduce the amount of required training data by using augmentation in the time domain. The use of artificial intelligence in the field of condition monitoring and diagnostics is becoming increasingly popular. However, the vast majority of work is based on sensors collocated in the machines. Therefore, the possibility of using acoustic arrays as a non-intrusive contactless sensor is emerging. The goal of this work is to develop data processing methods that allow for analysis of acoustic array images of vibration patterns. These methods have been demonstrated on a newly created experimental dataset where the a vibrating plate is measured using a 32 by 32 microphone array. On this plate, a disturbance mass placed in different positions to modify the dynamic properties of the system. The experimental results show that the proposed methods are able to determine the position of the disturbance mass even with low amounts of training data. They show to be promising for applications where space-frequency information is of essence.

Research on weak fault signal detection technology based on lv system

The detection of weak fault signals is realized by applying non-resonant parametric excitation to Lu chaotic system. Periodic signals far larger than the characteristic frequency of the system are taken as input, the sharp change of the system attractor is observed, and the existence of weak signals is detected. The conditions that the parameters of the detection system should meet are deduced by using the averaging method and Lyapunov method. Numerical research shows that the detection system can quickly reach a stable state. When the amplitude of the detected weak signals changes slightly, the attractor state changes abruptly, and the system is sensitive to the input. Through the "butterfly effect" of the system, feature signals that are difficult to extract by traditional methods can be detected. And the system has strong anti-interference performance.

An adaptive smooth unsaturated bistable stochastic resonance system and its application in rolling bearing fault diagnosis

An adaptive smooth unsaturated bistable stochastic resonance (ASUBSR) system for bearing fault signal detection is established. Based on the problem of output saturation and poor low-frequency suppression performance of classical bistable stochastic resonance (CBSR) system, an SUBSR with unsaturated characteristics is proposed. An ASUBSR system is designed by extracting the envelope spectrum of the input signal and resampling it to satisfy the adiabatic approximation condition, combining high-pass filter to filter out low-frequency interference, and using genetic algorithm to select the optimal system parameters. Through simulations and experiments, we found that the system can effectively suppress the interference of low-frequency and high-frequency, indicates that the system performs like a band-pass filter, and the output signal-to-noise ratio is better than that of the CBSR system. The proposed ASUBSR system has great application in the field of fault detection of rolling bearings.

Loss of Excitation Detection in Synchronous Generators Based on Dynamic State Estimation

In this article, we present a new approach to the detection of Loss-Of-Excitation (LOE), a typical failure of synchronous generators. Unlike most of the algorithms proposed in the literature, which only use the information available at the point of connection, we also take advantage of prior knowledge of the generator model. To track the field voltage and the other state variables, we have chosen the Constrained Unscented Kalman Filter (CUKF) as the core estimation technique, with phasor measurements as the input for this filtering algorithm. Detection of LOE is then performed by using the Faulty Modes Detection and Diagnosis (FMDD) algorithm, which combines the normal operation and a LOE-based model to decide in real-time whether an LOE has occurred or not. Results of simulations using a small two-area power system and the IEEE 39-bus system show that LOE detection times can be significantly reduced as compared to conventional and state-of-the-art approaches. Moreover, we observe that the new fault detection signal used to trip the generator can avoid short-term voltage stability problems.