Gearbox Vibration Research Articles

의사-레이블기법을 이용한 기계학습기반 설비상태 모니터링과 진단

This paper deals with a monitoring problem of the hot strip roughing mill that consists of lots of mechanical and electrical units. In fault diagnosis, various and complex units could be failed with unknown reasons and the main issue is to how to detect efficiently the abnormal or failed condition of the equipments. In this paper, pseudonormal and pseudo-abnormal data sets were defined and labeled to apply a classification analysis algorithm to unlabeled equipment vibration data. The Lagrangian perspective is used to compare the initial baseline pseudonormal dataset with the pseudo-abnormal dataset determined over time to monitor the classification evaluation scale for the status of the equipment. We use the support vector machine (SVM), K-NN, and discriminant analysis together to diagnose equipment fault and give alarm to maintain the facility properly applied classification algorithms and a procedure are proposed to generate alarms through test of correlation coefficient, cluster performance, and mean difference between clusters. A case study is done based on the experimental field gear box vibration data.

Frequencies estimation of gearbox vibration signal based on normalized lattice filter with RLS-based algorithm

In this paper, in order to estimate the meshing frequencies of a gearbox online, based on its vibration signal, an adaptive notch filter with a normalized lattice structure is proposed as an estimator. The notch filter exhibits a high degree of robustness and a low level of estimation errors, due to the minimal dynamic range of the node energies of the normalized lattice structure. A novel lattice recursive least square (NLRLS) algorithm, which introduces a correlation value between the vibration signal and the output signal of the estimation filter into the LRLS algorithm, is used as the adaptive algorithm for the notch filter. Theoretical analysis and simulation results show that the NLRLS algorithm exhibits greater stability than the LRLS algorithm for vibration signals where prior information is lacking. The meshing frequency estimation experiment is performed on a two-stage gearbox experimental platform. A triangular cascade structure is used to connect multiple notch filters to achieve multi-frequency estimation. The experimental results show that the notch filters can estimate meshing frequencies quickly and accurately, which verifies its effectiveness.

Gear fault detection based on instantaneous frequency estimation using variational mode decomposition and permutation entropy under real speed scenarios

AbstractStructural compactness with multiple meshing points excites multiple vibrations in a planetary gearbox of a wind turbine operating at varying speeds. The nonstationary, multicomponent vibration signals result in complex modulations challenging the effectiveness of a signal processing technique‐based fault diagnosis method. This paper aims to detect gear tooth faults under varying speed conditions based on the instantaneous frequency (IF) estimate of the decomposed component. Vibration signals were decomposed using variational mode decomposition (VMD), which is capable of frequency demodulation by exhibiting mono‐oscillatory components, followed by the IF estimation to select the most sensitive decomposed mode. The selected sensitive mode was further analyzed to deduce envelope spectrums for the extraction of frequency components highlighting fault symptoms. Furthermore, permutation entropy (PE) was evaluated to characterize the presence of randomness due to fault severity and to validate the proposed signal processing approach. In this paper, the vibration signals simulating planet gear faults were analyzed using the proposed signal processing approach. Symptoms of planet tooth faults were analyzed using the experimental investigation under real‐time varying speed to validate and exhibit the effectiveness of the proposed fault diagnosis approach for a planetary gearbox.

Novel Higher-Order Spectral Cross-Correlation Technologies for Vibration Sensor-Based Diagnosis of Gearboxes.

Novel vibration sensor-based diagnostic technologies, built on the higher order wavelet spectral cross-correlation (WSC), are proposed, investigated and applied to gearbox vibration diagnosis for the first time in worldwide terms. The proposed WSC-based technologies do not feature any constrains in selection of signal spectral components, relations between which are analysed. That is a radical improvement in comparison with the higher-order spectra (HOS). The WSC technologies are applied for an experimental diagnosis of a local gear tooth fault of a helical gearbox that is developed during a long duration gearbox endurance test. Differences between the applied technologies and advantages of the novel WSC approach over the classical HOS are explained in detail. Superiority of the WSC technologies is justified by high validity comprehensive experimental comparison with the HOS technologies: i.e., the wavelet bicoherence and the wavelet tricoherence.

Investigating parallel multi-step vibration processing pipelines for planetary stage fault detection in wind turbine drivetrains

This paper proposes a signal processing approach for wind turbine gearbox vibration signals based on employing multiple analysis pipelines. These so-called pipelines consist of combinations of various advanced signal processing methods that have been proven to be effective in literature when applied to wind turbine vibration signals. The performance of the pipelines is examined on vibration data containing different wind turbine gearbox faults in the planetary stages. Condition indicators are extracted from every pipeline to evaluate the fault detection capability for such incipient failures. The results indicate that the multipronged approach with the different pipelines increases the reliability of successfully detecting incipient planetary stage gearbox faults. The type, location, and severity of the fault influences the choice for the appropriate processing method combination. It is therefore often insufficient to only utilize a single processing pipeline for vibration analysis of wind turbine gearbox faults. Besides investigating the performance of the different processing techniques, the main outcome and recommendation of this paper is thus to employ a diversified analysis methodology which is not limited to a sole method combination, to improve the early detection rate of planetary stage gearbox faults.

Aircraft Gearbox Fault Diagnosis System: An Approach based on Deep Learning Techniques

Abstract Gearbox is one of the vital components in aircraft engines. If any small damage to gearbox, it can cause the breakdown of aircraft engine. Thus it is significant to study fault diagnosis in gearbox system. In this paper, two deep learning models (Long short term memory (LSTM) and Bi-directional long short term memory (BLSTM)) are proposed to classify the condition of gearbox into good or bad. These models are applied on aircraft gearbox vibration data in both time and frequency domain. A publicly available aircraft gearbox vibration dataset is used to evaluate the performance of proposed models. The results proved that accuracy achieved by LSTM and BLSTM are highly reliable and applicable in health monitoring of aircraft gearbox system in time domain as compared to frequency domain. Also, to show the superiority of proposed models for aircraft gearbox fault diagnosis, performance is compared with classical machine learning models.

The Effect of Parallel-Axis Helix Gear Pair on Wind Turbine Gearbox Vibration Control

This article studies the effects of some basic parameters of a parallel-axis helix gear stage on wind turbine gearbox vibration in a case study: a multibody dynamic model is constructed to simulate the drive train of a faulted multistage wind turbine gearbox with serious vibrations. The significant vibration behaviour of the drive train for typical excitations is calculated, and the results according to specified geometric parameters of the gears are analysed in detail to investigate effective solutions for vibration reduction. The results indicate that the helix angle and numbers of teeth of a gear pair are the most significant factors for solving the problem. The effectiveness of the proposed solutions and relevant mechanisms are discussed and validated by a prototype vibration test.

Translation Invariance-Based Deep Learning for Rotating Machinery Diagnosis

Discriminative feature extraction is a challenge for data-driven fault diagnosis. Although deep learning algorithms can automatically learn a good set of features without manual intervention, the lack of domain knowledge greatly limits the performance improvement, especially for nonstationary and nonlinear signals. This paper develops a multiscale information fusion-based stacked sparse autoencoder fault diagnosis method. The autoencoder takes advantage of the multiscale normalized frequency spectrum information obtained by dual-tree complex wavelet transform as input. Accordingly, the multiscale normalized features guarantee the translational invariance for signal characteristics, and the stacked sparse autoencoder benefits the unsupervised feature learning and ensures accurate and stable diagnosis performance. The developed method is performed on motor bearing vibration signals and worm gearbox vibration signals, respectively. The results confirm that the developed method can accommodate changing working conditions, be free of manual feature extraction, and perform better than the existing intelligent diagnosis methods.

Sensitivity Analysis of Online Oil Quality Monitoring for Early Detection of Water Ingress in Marine Propulsion Systems


 
 
 Gearboxes are critical equipment in the vessel propulsion system. Lubrication contamination caused by water ingress or condensation is one of the major failure modes in marine gearboxes leading to accelerated aging of lubricant, resulting in accelerated wear of gearbox components such as bearing and gears. This article presents a systematic evaluation of moisture ingress sensitivity of three commercial online oil quality monitoring sensors. A laboratory setup consisting of an industrial multi-stage planetary gearbox is utilised for the experimental studies. Furthermore, the gearbox vibration signatures are analysed to investigate the sensitivity of vibration measurements to water ingress to determine the need for oil quality monitoring when considering this specific failure mode.
 
 

Fault Diagnosis of Planetary Gearbox Based on Adaptive Order Bispectrum Slice and Fault Characteristics Energy Ratio Analysis.

The vibration of a planetary gearbox (PG) is complex and mutually modulated, which makes the weak features of incipient fault difficult to detect. To target this problem, a novel method, based on an adaptive order bispectrum slice (AOBS) and the fault characteristics energy ratio (FCER), is proposed. The order bispectrum (OB) method has shown its effectiveness in the feature extraction of bearings and fixed-shaft gearboxes. However, the effectiveness of the PG still needs to be explored. The FCER is developed to sum up the fault information, which is scattered by mutual modulation. In this method, the raw vibration signal is firstly converted to that in the angle domain. Secondly, the characteristic slice of AOBS is extracted. Different from the conventional OB method, the AOBS is extracted by searching for a characteristic carrier frequency adaptively in the sensitive range of signal coupling. Finally, the FCER is summed up and calculated from the fault features that were dispersed in the characteristic slice. Experimental data was processed, using both the AOBS-FCER method, and the method that combines order spectrum analysis with sideband energy ratio (OSA-SER), respectively. Results indicated that the new method is effective in incipient fault feature extraction, compared with the methods of OB and OSA-SER.

Dynamic Characteristics of Encased Differential Gear Train with Journal Bearing

Taking the marine encased differential gear train as an example, the relationship between the journal bearing parameters and the meshing force of the transmission system is analyzed. In this paper, the dynamic model of the encased differential gear train with journal bearing is established considering the factors of time-varying meshing stiffness and comprehensive transmission error. In this dynamic model, four stiffnesses and four damping coefficients are applied to characterize the asymmetry and interaction of the oil film stiffness and damping of planet bearing. The system responses are calculated by the Fourier series numerical algorithm. The results show that the introduction of journal bearing in encased differential gear train can contribute to gearbox vibration reduction. Moreover, the planet bearing parameters (e.g., clearance-to-radius ratio and eccentricity ratio) of the differential stage affect the meshing forces of both the differential and encased stages. In addition, the influence of the planet bearing parameters of the encased stage on the meshing force of the encased stage is more obvious than that of the differential stage. This work may develop a theoretical analysis framework for the design and manufacture of marine transmission systems in the future.

Resonance-based sparse adaptive variational mode decomposition and its application to the feature extraction of planetary gearboxes.

Due to the assumption that the VMD technique is essentially a set of adaptive Wiener filter banks and its performance depends to a large extent on the preset parameter K (the number of decomposition). A new method named resonance-based sparse adaptive variational mode decomposition (RSAVMD) is proposed for the decomposition of planetary gearbox vibration signals. Tunable Q-Factor Wavelet Transform (TQWT) and morphological component analysis (MCA) are introduced to decompose the original signal into high and low resonance components. High resonance components containing planetary gearbox signals are screened for analysis. At the same time, Quality factor is used to select the number of Variational mode decomposition (VMD) adaptively. This method was applied in fault diagnosis of planetary gearbox. Compared with VMD, RASVMD could extract fault characteristic frequency of planetary gearbox accurately, but VMD lost part of fault information, showing the superiority of RSAVMD. Simultaneously, the selection method of VMD decomposition number in literature was cited, and it was found that the decomposition number selected by the method in this paper was more accurate.

A new gearbox compound fault recognition approach based on improved double wavelet packet transform

A new approach for the diagnosis of compound faults in gearboxes is proposed in this paper. To extract characteristics in the frequency band of non-stationary raw vibration signals, a double-tree complex wavelet packet transform (DTCWPT) is used to decompose the signals. A singular value spectrum (SVP) is generated by performing singular value decomposition (SVD) on the matrix formed by all of the components. The new analysis method, DTCWPT-SVP, is used to diagnose different operating conditions of a gearbox, including single faults and compound faults, with a k-nearest neighbour (kNN) classifier. The results show that the minimum recognition accuracy using DTCWPT-SVP is 91.9% with different values of k in the kNN and that DTCWPT has better performance in signal decomposition than discrete wavelet packet transform (DWPT). Furthermore, the decomposition level used in DTCWPT is analysed in this paper. For the gearbox vibration problem, a level 2 or 3 DTCWPT can achieve good performance.

풍력발전기 상태 감시를 위한 SaaS 클라우드 인프라 내 데이터 처리 알고리즘 개선 연구

In this study, an SW for the analysis of the wind-turbine vibration characteristics was developed as an application of SaaS cloud infrastructure. A measurement system for power-performance, mechanical load, and gearbox vibration as type-test class was installed at a target MW-class wind turbine, and structural meta and raw data were then acquired into the cloud. Data processing algorithms were developed to provide cloud data to the SW. To operate the SW continuously, raw data was downloaded consistently based on the algorithms. During the SW test, an intermittent long time-delay occurred due to the communication load associated with frequent access to the cloud. To solve this, a compression service for the target raw data was developed in the cloud and more stable data processing was confirmed. Using the compression service, stable big data processing of wind turbines, including gearbox vibration analysis, is expected.

Dynamic modeling and analysis of wind turbine drivetrain considering the effects of non-torque loads

With the increase of the rotor diameter and the deterioration of operating conditions, modern wind turbines suffer from more and more significant time-varying non-torque loads, which increases the burden of turbine structures especially the gearbox. Based on an aeroelastic loose coupling approach and assembly of the finite element method, an integrated drivetrain coupling analysis model including blade module, aerodynamic module, and gearbox module is established in this study. This proposed model is validated by comparing the calculation results with previous literature. Taking National Renewable Energy Laboratory 5-MW wind turbine as the research object, the gearbox vibration responses, gear meshing forces and bearing forces under non-torque loads caused by blade gravity, wind shear (WS), tower shadow (TS) and yawed inflow are studied in detail. Results show that the y-direction displacements of the gearbox, sun gear 1, sun gear 2 and gear are larger than those in x-direction because Fy or Mx generated by blade gravity, WS and TS dominates the non-torque loads. The non-torque loads lead to a non-uniform planet load sharing especially for the planetary gear stage 1. Because of the fluctuations of non-torque loads, not only the rotation frequencies of the corresponding carrier but also the multiple frequencies of the carrier 1 are observed in the frequency spectrums. The non-torque loads are mainly borne by carrier 1 bearings. Except for the blade gravity, the bearing forces caused by other unsteady inflows have obvious fluctuations.

Spectral Kurtosis of Choi–Williams Distribution and Hidden Markov Model for Gearbox Fault Diagnosis

A combination of spectral kurtosis (SK), based on Choi–Williams distribution (CWD) and hidden Markov models (HMM), accurately identifies initial gearbox failures and diagnoses fault types of gearboxes. First, using the LMD algorithm, five types of gearbox vibration signals are collected and decomposed into several product function (PF) components and the multicomponent signals are decomposed into single-component signals. Then, the kurtosis value of each component is calculated, and the component with the largest kurtosis value is selected for the CWD-SK analysis. According to the calculated CWD-SK value, the characteristics of the initial failure of the gearbox are extracted. This method not only avoids the difficulty of selecting the window function, but also provides original eigenvalues for fault feature classification. In the end, from the CWD-SK characteristic parameters at each characteristic frequency, the characteristic sequence based on CWD-SK is obtained with HMM training and diagnosis. The experimental results show that this method can effectively identify the initial fault characteristics of the gearbox, and also accurately classify the fault characteristics of different degrees.

Condition Monitoring of Wind Turbine Planetary Gearboxes Under Different Operating Conditions

Abstract Digitally enhanced services for wind power could reduce operations and maintenance costs as well as the levelized cost of energy. Therefore, there is a continuous need for advanced monitoring techniques, which can exploit the opportunities of internet of things and big data analytics, revolutionizing the future of the energy sector. The heart of wind turbines is a rather complex epicyclic gearbox. Among others, extremely critical gearbox components, which are often responsible for machinery stops, are the rolling element bearings. The vibration signatures of bearings are rather weak compared to other components, such as gears, and as a result, an extended number of signal processing techniques and tools have been proposed during the last decades, focusing toward accurate, early, and on time bearing fault detection with limited false alarms and missed detections. Envelope analysis is one of the most important methodologies, where an envelope of the vibration signal is estimated usually after filtering around a frequency band excited by impacts due to the bearing faults. Different tools, such as Kurtogram, have been proposed in order to accurately select the optimum filter parameters (center frequency and bandwidth). Cyclic spectral correlation (CSC) and cyclic spectral coherence (CSCoh), based on cyclostationary analysis, have been proved as very powerful tools for condition monitoring. The monitoring techniques seem to have reached a mature level in case a machinery operates under steady speed and load. On the other hand, in case the operating conditions change, it is still unclear whether the change of the monitoring indicators is due to the change of the health of the machinery or due to the change of the operating parameters. Recently, the authors have proposed a new tool called improved envelope spectrum via feature optimization-gram (IESFOgram), which is based on CSCoh and can automatically select the filtering band. Furthermore, the CSCoh is integrated along the selected frequency band leading to an improved envelope spectrum (IES). In this paper, the performance of the tool is evaluated and further extended on cases operating under different speeds and different loads. The effectiveness of the methodology is tested and validated on the National Renewable Energy Laboratory (NREL) wind turbine gearbox vibration condition monitoring benchmarking dataset, which includes various faults with different levels of diagnostic complexity as well as various speed and load operating conditions.

Cyclostationary modeling for local fault diagnosis of planetary gear vibration signals

Owing to the rotation and reciprocation of mechanical equipment, their vibration signals inherently exhibit cyclic stationary characteristics. This paper provides a phenomenological signal model of the planetary gear vibrations under the cyclostationarity paradigm. The model is helpful to give comprehensive demodulation information which is the limit of classic spectral analysis of deterministic signals, so that the diagnosis of local faults in planetary gear sets can be more explicit. First, the effects of rotating speed fluctuation on periodic gear meshing vibrations are analyzed. It is shown that the stationary speed fluctuation will turn the periodicity of signals into second-order cyclostationarity. Next, an impulsive oscillation sequence with interval timing jitter is used to describe the dynamic response of the meshing stiffness change caused by local defects. Meanwhile, the influence of a vibration transfer path is also considered in the model. Based on the above analysis, the following procedure is utilized to analyze planetary gearbox vibrations: remove the periodic components by the self-adaptive noise cancellation method and then, apply the second-order frequency descriptor to reveal fault features of the signals. Both the simulation and experiment are used to validate the correctness of model derivation and the superiority of the cyclostationary analysis.

Dictionary Learning Approach to Monitoring of Wind Turbine Drivetrain Bearings

Condition monitoring is central to the efficient operation of wind farms due to the challenging operating conditions, rapid technology development and large number of aging wind turbines. In particular, predictive maintenance planning requires the early detection of faults with few false positives. Achieving this type of detection is a challenging problem due to the complex and weak signatures of some faults, particularly the faults that occur in some of the drivetrain bearings. Here, we investigate recently proposed condition monitoring methods based on unsupervised dictionary learning using vibration data recorded over 46 months under typical industrial operations. Thus, we contribute novel test results and real world data that are made publicly available. The results of former studies addressing condition monitoring tasks using dictionary learning indicate that unsupervised feature learning is useful for diagnosis and anomaly detection purposes. However, these studies are based on small sets of labeled data from test rigs operating under controlled conditions that focus on classification tasks, which are useful for quantitative method comparisons but gives little insight into how useful these approaches are in practice. In this study, dictionaries are learned from gearbox vibrations in six different turbines, and the dictionaries are subsequently propagated over a few years of monitoring data when faults are known to occur. We perform the experiment using two different sparse coding algorithms to investigate if the algorithm selected affects the features of abnormal conditions. We calculate the dictionary distance between the initial and propagated dictionaries and find the time periods of abnormal dictionary adaptation starting six months before a drivetrain bearing replacement and one year before the resulting gearbox replacement.

Dynamic modeling of planetary gear set with tooth surface wear

Wear commonly occurs in planetary gear transmission systems. Fundamentally, the tooth wear could cause tooth profile deviation, which would increase the vibration and noise of gearbox. In order to monitor and forecast the wear condition of gears via vibration-based methods, it is necessary to establish the dynamics model of a planetary gear set with tooth surface wear, which can provide a prior about the vibration characteristics of gear wear. In this study, a purely torsional dynamics model of a planetary gear set with tooth surface wear is proposed to analyze the fault mechanism of tooth surface wear. The tooth surface wear is incorporated into the dynamics model through unloaded static transmission error (USTE) and time-varying mesh stiffness (TVMS), which are evaluated by Archard’s wear equation. Subsequently, the vibration responses of the planetary gear set with tooth surface wear are analyzed. It is revealed that tooth wear would change the vibration responses in both time- and frequency-domain and the condition indicators present different trends.