Showing papers on "Condition monitoring published in 2019"
TL;DR: This paper reviews the recent literature on machine learning models that have been used for condition monitoring in wind turbines and shows that most models use SCADA or simulated data, with almost two-thirds of methods using classification and the rest relying on regression.
482 citations
TL;DR: Experimental results on two wind turbine datasets show that the proposed fault diagnosis framework is able to do fault classification effectively from raw time-series signals collected by single or multiple sensors and outperforms state-of-art approaches.
323 citations
TL;DR: A systemic and pertinent state-of-art review on WT planetary gearbox condition monitoring techniques on the topics of fundamental analysis, signal processing, feature extraction, and fault detection is provided.
312 citations
TL;DR: Experimental results have shown that the proposed predictive modeling approach is capable of predicting the surface roughness of 3D printed components with high accuracy.
Abstract: Additive manufacturing (AM), also known as 3D printing, has been increasingly adopted in the aerospace, automotive, energy, and healthcare industries over the past few years. While AM has many advantages over subtractive manufacturing processes, one of the primary limitations of AM is surface integrity. To improve the surface integrity of additively manufactured parts, a data-driven predictive modeling approach to predicting surface roughness in AM is introduced. Multiple sensors of different types, including thermocouples, infrared temperature sensors, and accelerometers, are used to collect temperature and vibration data. An ensemble learning algorithm is introduced to train the predictive model of surface roughness. Features in the time and frequency domains are extracted from sensor-based condition monitoring data. A subset of these features is selected to improve computational efficiency and prediction accuracy. The predictive model is validated using the condition monitoring data collected from a set of AM tests conducted on a fused filament fabrication (FFF) machine. Experimental results have shown that the proposed predictive modeling approach is capable of predicting the surface roughness of 3D printed components with high accuracy.
226 citations
TL;DR: A novel fault diagnosis approach integrating Convolutional Neural Networks and Extreme Learning Machine, which can detect different fault types and outperforms other methods in terms of classification accuracy is proposed.
207 citations
TL;DR: The experimental results indicate that the proposed method achieves high accuracy in bearing fault diagnosis under complex operational conditions and is superior to traditional methods and standard deep learning methods.
Abstract: Recently, research on data-driven bearing fault diagnosis methods has attracted increasing attention due to the availability of massive condition monitoring data. However, most existing methods still have difficulties in learning representative features from the raw data. In addition, they assume that the feature distribution of training data in source domain is the same as that of testing data in target domain, which is invalid in many real-world bearing fault diagnosis problems. Since deep learning has the automatic feature extraction ability and ensemble learning can improve the accuracy and generalization performance of classifiers, this paper proposes a novel bearing fault diagnosis method based on deep convolutional neural network (CNN) and random forest (RF) ensemble learning. Firstly, time domain vibration signals are converted into two dimensional (2D) gray-scale images containing abundant fault information by continuous wavelet transform (CWT). Secondly, a CNN model based on LeNet-5 is built to automatically extract multi-level features that are sensitive to the detection of faults from the images. Finally, the multi-level features containing both local and global information are utilized to diagnose bearing faults by the ensemble of multiple RF classifiers. In particular, low-level features containing local characteristics and accurate details in the hidden layers are combined to improve the diagnostic performance. The effectiveness of the proposed method is validated by two sets of bearing data collected from reliance electric motor and rolling mill, respectively. The experimental results indicate that the proposed method achieves high accuracy in bearing fault diagnosis under complex operational conditions and is superior to traditional methods and standard deep learning methods.
182 citations
TL;DR: This paper presents the state of the art review describing different type of IM faults and their diagnostic schemes, and several monitoring techniques available for fault diagnosis of IM have been identified and represented.
Abstract: There is a constant call for reduction of operational and maintenance costs of induction motors (IMs) These costs can be significantly reduced if the health of the system is monitored regularly This allows for early detection of the degeneration of the motor health, alleviating a proactive response, minimizing unscheduled downtime, and unexpected breakdowns The condition based monitoring has become an important task for engineers and researchers mainly in industrial applications such as railways, oil extracting mills, industrial drives, agriculture, mining industry etc Owing to the demand and influence of condition monitoring and fault diagnosis in IMs and keeping in mind the prerequisite for future research, this paper presents the state of the art review describing different type of IM faults and their diagnostic schemes Several monitoring techniques available for fault diagnosis of IM have been identified and represented The utilization of non-invasive techniques for data acquisition in automatic timely scheduling of the maintenance and predicting failure aspects of dynamic machines holds a great scope in future
155 citations
TL;DR: In order to make the filtering results of Condition Monitoring (CM) data smoother and avoid misjudgment of status when the degradation speed is negative, the measurement error parameter is selected as the standard deviation of CM data in the degradation stage.
146 citations
TL;DR: A new fault diagnosis strategy based on the synchrosqueezing transform (SST) and the deep convolutional neural network (DCNN) is proposed in this paper, which automatically recognizes the planet bearing fault type, which is free from artificially capturing fault characteristic frequencies in spectrum or time-frequency spectrum that contain many interference items.
130 citations
TL;DR: This paper reviews recent onboard condition monitoring sensors, systems, methods and techniques, aiming to define the present state of the art and its potential application for freight wagons without onboard electric power.
Abstract: Given the constant demand for heavier, longer, faster, and more efficient rail freight vehicles, onboard fault detection systems appear as a good approach for enhanced railway asset exploitation. Real-time condition monitoring reduces inefficient preventive and reactive maintenance actions, decreases waste from replacing parts that still have a useful life, and improves availability and safety by real-time rolling stock diagnosis. There have been considerable advances in wayside monitoring applications, but these cannot achieve real-time continuous monitoring. With the price reduction and miniaturization trends of electronic devices, the cost of deploying wireless sensor networks onboard freight trains continues to become more feasible and accessible. On the other hand, the lack of onboard electric power availability on freight wagons appears as the major limitation for the implementation of these technologies. This paper reviews recent onboard condition monitoring sensors, systems, methods and techniques, aiming to define the present state of the art and its potential application for freight wagons without onboard electric power.
101 citations
TL;DR: A deep-learning-based model termed multiresolution & multisensor fusion network for motor fault diagnosis, through multiscale analysis of motor vibration and stator current signals is presented.
Abstract: Condition monitoring and fault diagnosis are of significance to improve the safety and reliability of motors, given their widespread applications in virtually every branch of the industry. Sequential data modeling based on recurrent neural network and its variants have drawn increasing attention because the temporal nature of motor signals can be well leveraged for motor analysis. One common drawback of prior research is that signals measured on motors are typically analyzed with a fixed time window, making it difficult to tradeoff between global state estimation and local feature extraction. This paper presents a deep-learning-based model termed multiresolution & multisensor fusion network for motor fault diagnosis, through multiscale analysis of motor vibration and stator current signals. Specifically, vibration and current signals are first segmented by analysis windows of varying lengths to create a new data stream for the joint representation and temporal encoding of the original sensor signals, based on two network structures: convolutional neural network and long short-term memory. The advantage of the developed method is that it automatically learns the discriminative features through the network training process, without requiring manual feature selection as is typically the case in prior methods. By considering the temporal dependence of the signals being analyzed, the developed multiresolution fusion technique not only improves the effectiveness of feature extraction but is also adaptive to varying motor speed. Two case studies demonstrate the advantages of the developed method.
TL;DR: A novel approach, namely physics-based convolutional neural network (PCNN), for fault diagnosis of rolling element bearings is proposed and the performance of PCNN in machinery fault diagnosis is compared with that of traditional machine learning- and deep learning-based approaches reported in the literature.
Abstract: During the past few years, deep learning has been recognized as a useful tool in condition monitoring and fault detection of rolling element bearings. Although existing deep learning approaches are able to intelligently detect and classify the faults in bearings, they still face one or both of the following challenges: 1) most of these approaches rely exclusively on data and do not incorporate physical knowledge into the learning and prediction processes and 2) the approaches often focus on the fault diagnosis of a single bearing in a rotating machine, while in reality, a rotating machine may contain multiple bearings. To address these challenges, this paper proposes a novel approach, namely physics-based convolutional neural network (PCNN), for fault diagnosis of rolling element bearings. In PCNN, an exclusively data-driven deep learning approach, called CNN, is carefully modified to incorporate useful information from physical knowledge about bearings and their fault characteristics. To this end, the proposed approach 1) utilizes spectral kurtosis and envelope analysis to extract sidebands from raw sensor signals and minimize non-transient components of the signals and 2) feeds the information about the fault characteristics into the CNN model. With the capability to process signals from multiple sensors, the proposed PCNN approach is capable of concurrently monitoring multiple bearings and detecting faults in these bearings. The performance of PCNN in machinery fault diagnosis is compared with that of traditional machine learning- and deep learning-based approaches reported in the literature.
TL;DR: A transfer learning-convolutional neural network (TLCNN) based on AlexNet is proposed for bearing fault diagnosis that has higher accuracy and has much better robustness against noise than other deep learning and traditional methods.
Abstract: Traditional methods used for intelligent condition monitoring and diagnosis significantly depend on manual feature extraction and selection. To address this issue, a transfer learning-convolutional neural network (TLCNN) based on AlexNet is proposed for bearing fault diagnosis. Firstly, a 2D image representation method converts vibration signals to 2D time-frequency images. Secondly, the proposed TLCNN model extracts the features of the 2D time-frequency images and achieves the classification conditions of the bearing, which is faster to train and more accurate. Thirdly, t-distributed stochastic neighbor embedding (t-SNE) is applied to visualize the feature learning process to demonstrate the feature learning ability of the proposed model. The experimental results verify that the proposed fault diagnosis model has higher accuracy and has much better robustness against noise than other deep learning and traditional methods.
TL;DR: This paper presents a deep learning based solution for defect pattern recognition by the use of aerial images obtained from unmanned aerial vehicles that significantly improves the efficiency and accuracy of asset inspection and health assessment for large-scale PV farms in comparison with the conventional solutions.
Abstract: The efficient condition monitoring and accurate module defect detection in large-scale photovoltaic (PV) farms demand for novel inspection method and analysis tools. This paper presents a deep learning based solution for defect pattern recognition by the use of aerial images obtained from unmanned aerial vehicles. The convolutional neural network is used in the machine learning process to classify various forms of module defects. Such a supervised learning process can extract a range of deep features of operating PV modules. It significantly improves the efficiency and accuracy of asset inspection and health assessment for large-scale PV farms in comparison with the conventional solutions. The proposed algorithmic solution is extensively evaluated from different aspects, and the numerical result clearly demonstrates its effectiveness for efficient defect detection of PV modules.
TL;DR: A general overview of the available knowledge regarding vibration-based speed estimation techniques is targeted by means of a performance comparison of seven speed estimation methods on three different experimental data sets and the resulting speed estimation data of all tested methods is made publicly available such that it can help in forming a benchmark for futurespeed estimation methods.
TL;DR: Vibration analysis technique along with other condition monitoring techniques gives better results in the fault diagnostics and condition monitoring of rolling element bearing.
Abstract: Different types of machines having rotary component are linked together in process industries, to perform the process of manufacturing. The failure of any single machine rotary component in the process can result in loss of money per downtime hour. To continue the working of the machines, it is necessary to monitor the health of the machine during its operation. Bearing failure is considered to be one of the major causes of breakdown in different rotating machines operating in industry at high and low speeds. Vibration in machines is generated due to defective bearings. During use, the condition of the bearings change, hence the vibrations also change and its characteristic basically depends on the cause. Hence, this characteristic feature of the bearing makes it suitable for vibration monitoring and other monitoring techniques. This article presents the brief review of recent trends in research on bearing defects, sources of vibration and vibration measurement techniques in the time domain, frequency domain and time frequency domain. The study reveals that, the envelope analysis method and time–frequency analysis technique are able to detect the bearing fault effectively. The wavelet analysis technique along with artificial neural network and fuzzy logic is also found to be the most effective techniques for fault analysis in rolling element bearing. Finally, it is concluded that vibration analysis technique along with other condition monitoring techniques gives better results in the fault diagnostics and condition monitoring of rolling element bearing.
TL;DR: Results may be used to target the development of condition monitoring systems focusing on critical systems and to find optimal O&M strategies by understanding failure paths of main offshore wind turbine systems resulting in a lower cost of energy and a more optimal risk-return balance.
TL;DR: In this paper, an artificial neural network (ANN) and fuzzy logic are used for mapping inputs distance, time of travel of EV and outputs casing temperature, winding temperature, time to refill the bearing lubricant, percentage deterioration of magnetic flux to compute remaining useful life (RUL) of permanent magnet (PM).
Abstract: Electric mobility has become an essential part of the future of transportation. Detection, diagnosis and prognosis of fault in electric drives are improving the reliability, of electric vehicles (EV). Permanent magnet synchronous motor (PMSM) drives are used in a large variety of applications due to their dynamic performances, higher power density and higher efficiency. In this study, health monitoring and prognosis of PMSM is developed by creating intelligent digital twin (i-DT) in MATLAB/Simulink. An artificial neural network (ANN) and fuzzy logic are used for mapping inputs distance, time of travel of EV and outputs casing temperature, winding temperature, time to refill the bearing lubricant, percentage deterioration of magnetic flux to compute remaining useful life (RUL) of permanent magnet (PM). Health monitoring and prognosis of EV motor using i-DT is developed with two approaches. Firstly, in-house health monitoring and prognosis is developed to monitor the performance of the motor in-house. Secondly, Remote Health Monitoring and Prognosis Centre (RHMPC) is developed to monitor the performance of the motor remotely using cloud communication by the service provider of the EV. The simulation results prove that the RUL of PM and time to refill the bearing lubricant obtained by i-DT twins theoretical results.
TL;DR: A new signal reconstruction modeling technique is proposed using support vector regression that demonstrates improved performance in detecting wind turbine faults, and controlling false and missed alarms.
TL;DR: This paper proposed a comprehensive overview of the work on fuel cell conditions monitoring technology from two different points of view, technology, and scenario, and proposed a 5 × 5 tables for detail comparison.
TL;DR: Motivated by spectral kurtosis (SK) and extreme learning machine (ELM), a novel intelligent diagnosis method for fault classification of rotating machines is proposed and the significance of SK as a feature set is examined and improved ELM in comparison with traditional methods.
Abstract: The condition monitoring of rotating machinery systems based on effective and intelligent fault diagnosis has been widely accepted. Traditional signal processing (SP) methods are less effective due to noises and interferences from different sources and incipient faults which remain active for a short time with a particular frequency. In recent times, SP techniques along with artificial intelligence methods are being used for fault classification. Various complex approaches in SP domain have used for feature extraction of the vibration data to design a feature set. A challenging task is to select dominant features from the available feature set for improving the accuracy of fault classification. Thus, motivated by spectral kurtosis (SK) and extreme learning machine (ELM), we propose a novel intelligent diagnosis method for fault classification of rotating machines. In this paper, SK is used as an input feature set to avoid the task of finding the dominant feature set. The extracted features are fed to ELM for fault identification. However, ELM performance primarily depends upon the hidden node parameters and the number of hidden nodes. The selection of optimum ELM parameters for good performance is an open issue. Therefore, modified bidirectional search with local search method is proposed to determine the optimum set of ELM parameters. The developed method is tested on two vibration data sets of rolling element bearings. We examined the significance of SK as a feature set and improved ELM in comparison with traditional methods. The experimental results demonstrate that the proposed method efficiently optimizes the ELM parameters to provide a compact ELM architecture and also enhances the fault classification accuracy.
TL;DR: This paper presents a fast, accurate, and simple systematic approach for online condition monitoring and severity identification of ball bearings that utilizes compact one-dimensional convolutional neural networks to identify, quantify, and localize bearing damage.
Abstract: This paper presents a fast, accurate, and simple systematic approach for online condition monitoring and severity identification of ball bearings. This approach utilizes compact one-dimensional (1-D) convolutional neural networks (CNNs) to identify, quantify, and localize bearing damage. The proposed approach is verified experimentally under several single and multiple damage scenarios. The experimental results demonstrated that the proposed approach can achieve a high level of accuracy for damage detection, localization, and quantification. Besides its real-time processing ability and superior robustness against the high-level noise presence, the compact and minimally trained 1-D CNNs in the core of the proposed approach can handle new damage scenarios with utmost accuracy.
TL;DR: A machine learning based approach for detecting drifting behavior – so-called concept drifts – in continuous data streams as potential indication for defective system behavior and depict initial tests on synthetic data sets is presented.
TL;DR: A model-based insulation fault diagnosis method is proposed using a high-fidelity cell model and the Kalman filter based state observer is designed for joint estimation of both the battery voltage and state-of-charge using the identified battery model.
TL;DR: The aim of this review was to present state-of-the-art UHF sensors in PD detection and facilitate future improvements in the UHF method.
Abstract: Condition monitoring of an operating apparatus is essential for lifespan assessment and maintenance planning in a power system. Electrical insulation is a critical aspect to be monitored, since it is susceptible to failure under high electrical stress. To avoid unexpected breakdowns, the level of partial discharge (PD) activity should be continuously monitored because PD occurrence can accelerate the aging process of insulation in high voltage equipment and result in catastrophic failure if the associated defects are not treated at an early stage. For on-site PD detection, the ultra-high frequency (UHF) method was employed in the field and showed its effectiveness as a detection technique. The main advantage of the UHF method is its immunity to external electromagnetic interference with a high signal-to-noise ratio, which is necessary for on-site monitoring. Considering the detection process, sensors play a critical role in capturing signals from PD sources and transmitting them onto the measurement system. In this paper, UHF sensors applied in PD detection were comprehensively reviewed. In particular, for power transformers, the effects of the physical structure on UHF signals and practical applications of UHF sensors including PD localization techniques were discussed. The aim of this review was to present state-of-the-art UHF sensors in PD detection and facilitate future improvements in the UHF method.
TL;DR: The main contribution is to establish an adaptive empirical wavelet transform framework for fault-related mode extraction, which incorporates a novel meshing frequency modulation phenomenon to enhance the planetary gear related vibration components in wind turbine gearbox.
TL;DR: A straightforward tacho-less order tracking method based on order spectrogram visualization is proposed in this paper, which has been validated by both simulated and experimental rolling bearing vibration signals.
TL;DR: An acoustic-based fault detection in a three-phase induction motor is done by estimating the torque from the acoustic signals released by the machine by exploiting the energy possessed in the processed acoustic signal.
Abstract: Condition monitoring of electric drives play a significant role for a safe working environment. Induction motors are widely used in industries and any fault in it leads to interruption in the process or complete shutdown of the equipment. In this paper, acoustic-based fault detection in a three-phase induction motor is done by estimating the torque from the acoustic signals released by the machine. The fault detection is possible as acoustic emission is different for faults such as single phasing, bearing cage damage, and broken rotor bars. The acoustic signals are processed using rational-dilation wavelet transform (RADWT) technique to extract the fault features and thereby diagnose the fault type. The torque estimation is done using multiple regression method by extracting the energy possessed in the processed acoustic signal and the faults are diagnosed precisely. An experimental setup comprising of a three-phase induction motor with brake drum loading is used to validate this approach. The RADWT has adjustable frequency resolution in comparison with other wavelet methods. When high Q-factor filters are employed in the RADWT, better representation of different faults are obtained in the decomposed sub-bands. In addition, characteristic frequencies of different faults are calculated analytically and validated by observing the frequencies in the FFT spectrum of acoustic fault signals.
TL;DR: A novel method for accurately predicting tool wear under varying cutting conditions based on a proposed new meta-learning model which can be easily trained, updated and adapted to new machining tasks of different cutting conditions is presented.
TL;DR: A multi-objective iterative optimization algorithm (MOIOA) for multi-fault diagnosis is proposed and results indicate that MOIOA is efficient to extract weak fault features even with heavy noise and harmonic interferences.
Abstract: Rolling element bearings (REBs) play an essential role in modern machinery and their condition monitoring is significant in predictive maintenance. Due to the harsh operating conditions, multi-fault may co-exist in one bearing and vibration signal always exhibits low signal-to-noise ratio (SNR), which causes difficulties in detecting fault. In the previous studies, maximum correlated kurtosis deconvolution (MCKD) has been validated as an efficient method to extract fault feature in the fault signals. Nonetheless, there are still some challenges when MCKD is applied to fault detection owing to the rigorous requirements of multiple input parameters. To overcome limitation, a multi-objective iterative optimization algorithm (MOIOA) for multi-fault diagnosis is proposed. In this method, correlated kurtosis (CK) is taken as a criterion to select optimal Morlet wavelet filter using the whale optimization algorithm (WOA). Meanwhile, to further eliminate the effect of the inaccurate period on CK, the update process of period is incorporated. After that, the simulated and experimental signals are utilized to testify the validity and superiority of the MOIOA for multiple faults detection by the comparison with MCKD. The results indicate that MOIOA is efficient to extract weak fault features even with heavy noise and harmonic interferences.