Gearbox Bearings Research Articles

A mixed adversarial adaptation network for intelligent fault diagnosis

Behind the brilliance of the deep diagnosis models, the issue of distribution discrepancy between source training data and target test data is being gradually concerned for catering to more practical and urgent diagnostic requirements. Consequently, advanced domain adaptation algorithms have been introduced to the field of fault diagnosis to address this problem. However, in performing domain adaptation, most existing diagnosis methods only focus on the minimization of marginal distribution divergences and do not consider conditional distribution differences at the same time. In this paper, we present a mixed adversarial adaptation network (MAAN) based intelligent framework for cross-domain fault diagnosis of machinery. In this approach, differences in marginal distribution and conditional distribution are reduced together by the adversarial learning strategy, moreover, a simple adaptive factor is also endowed to dynamically weigh the relative importance of two distributions. Extensive experiments on two kinds of mechanical equipment, i.e. planetary gearbox and rolling bearing, are built to validate the proposed method. Empirical evidence demonstrates that the proposed model outperforms popular deep learning and deep domain adaptation diagnosis methods.

Early stage white etching crack identification using artificial neural networks

White Etching Cracks (WEC) in gearbox bearings is a major concern in the wind turbine industry, which can lead to a premature failure of the gearbox. Though many hypotheses regarding the generation of WEC have been proposed over the decades, the answer is still disputable. To trace back the failures to earlier stages before they occur, an innovative sensor-set has been utilized on a test rig to monitor the influencing factors that lead to WEC. This paperwork seeks to recognize abnormal patterns from recorded sensor data and derive statements of sensible sensor combinations in WEC early detection. A Long Short Term Memory (LSTM) network-based autoencoder is proposed for the anomaly detection (AD) task. Employing an auto-associative sequence-to-sequence predictor, a model is trained to reconstruct the normal time series data without WEC. The reconstruction error of testing time series data is evaluated for the determination of its anomaly. The results show that the specified LSTM autoencoder framework can qualitatively distinguish anomalies from collected multivariate time series data. Moreover, the anomaly score evaluated via reconstruction-error-based metrics can discriminate normal and abnormal behaviors in the study. This investigation’s results entail a significant step towards early WEC risk detection and more cost-efficient wind turbine technology if this approach can be further applied on stream data with plausible thresholds in monitoring system.

A proposed criteria to identify wind turbine drivetrain bearing loads that induce roller slip based white-Etching cracks

In this article, the type of roller slip behavior that may result in the formation of white-etching cracks (WECs) in wind turbine gearbox bearings is identified. A new hypothesis based on the inner raceway normal contact load magnitude at the time of roller slip is proposed as the probable cause of WECs. For this purpose, the maximum normal contact loads are identified when roller slip occurs in high-speed shaft bearings at different mean wind speeds. Subsequently, the annual probability of occurrence of the maximum normal loads are obtained. The probability of maximum load under slip exceeding a limit probability is hypothesized as a probable cause for WEC. In order to apply the proposed hypothesis, two different wind turbines high-speed shaft bearings are used: the cylindrical roller bearing of the General Electric 1.5 SLE turbine and the tapered roller bearing of Vestas V52 turbine. Both the chosen bearings are on the generator side of the high-speed shaft. For both turbines, measurement data together with analytical models are used for identifying the slip and the maximum normal contact loads. We propose forecasting the probability of exceedance of a threshold maximum normal contact load level during slip to identify the possibility for inducing WECs.

Simulative investigation of ring creep on a\xa0planetary bearing of a\xa0wind turbine gearbox

Wind turbines (WT) must be further optimized concerning availability and reliability. One of the major reasons of WT downtime is the failure of gearbox bearings. Some of these failures occur, due to the ring creep phenomenon, which is mostly detected in the planetary bearings. The ring creep phenomenon describes a relative movement of the outer ring to the planetary gear. In order to improve the understanding of ring creep, the finite element method (FEM) is used to simulate ring creep in planetary gears. First, a sensitivity analysis is carried out on a small bearing size (NU205), to characterize relevant influence parameters for ring creep—considered parameters are teeth module, coefficient of friction, interference fit and normal tooth forces. Secondly, a full-scale planetary bearing (SL185030) of a 1MW WT is simulated and verified with experimental data.

Data-driven wear monitoring for sliding bearings using acoustic emission signals and long short-term memory neural networks

Driven by the potential applications of sliding bearings in planetary gearboxes for wind turbines, the wear prognosis of heavy loaded sliding bearings under low rotational speeds is an important aspect. The aims of this study are to identify an adequate condition monitoring technique and demonstrate the potential of data-driven wear monitoring for scenarios, where transient wear data for data-driven monitoring is not available.In a first step, Acoustic Emission (AE) technique has been applied to a special test rig for planetary gearbox sliding bearings. It was demonstrated that AE can be used to distinguish between wear-critical mixed friction and hydrodynamic regime. In a second step, a data-driven method for wear monitoring was developed and applied to a component test rig for sliding bearings. For validation and generation of condition monitoring data, sliding bearings from the same bronze material were subjected to steady speed conditions in mixed lubrication regime with different loads as well as runtime. The results from these experiments and results from validated physical wear simulations are the input parameters for the proposed data-driven approach. The wear prognosis is performed with recurrent neural networks, which can predict the transient degradation. With the developed data-driven approach to wear monitoring, a good accuracy can be achieved that is capable of real-time wear monitoring.

Detection of Lubrication State in a Field Operational Wind Turbine Gearbox Bearing Using Ultrasonic Reflectometry

Fully flooded lubrication is the ideal state for a rolling bearing; this is especially true in the aggressive environment of a wind turbine transmission where bearings are subject to intermittent operation and highly variable loading. In this paper, a novel ultrasonic reflection method is used to detect the presence of oil between rollers in the bearing. Ultrasonic sensors were instrumented on the static inner (lab) and outer (field) bearing raceways and reflections were captured as the rollers travelled past the sensor. The proportion of the sound wave reflected (known as the reflection coefficient, R) is dependent on the acoustic mismatch of the materials either side of the interface. Changes in R indicate either a steel–air or steel–oil interface as R values transitioned from 1 to 0.95, respectively, and even lower for a steel–roller interface. Consequently, it was possible to detect the presence of lubricant on the raceway between roller passes. From the laboratory measurements, the recurring reflection coefficient patterns between roller passes were used to identify the lubrication condition of the raceway. An absence of these patterns between roller passes indicated the absence of lubricant on the bearing surface. For the field measurements, three bearing lubrication conditions (partial, insufficient, and fully lubricated) were observed. Partially and insufficiently lubricated datasets were found to occur mostly during transient operation. As transient operation is often accompanied by overloading and torque reversals, coupled with the lubrication issues, these all act to increase the risk of premature bearing failure.

A Hybrid Intelligent Model for the Condition Monitoring and Diagnostics of Wind Turbines Gearbox

Wind turbines (WTs) are often operated in harsh and remote environments, thus making them more prone to faults and costly repairs. Additionally, the recent surge in wind farm installations have resulted in a dramatic increase in wind turbine data. Intelligent condition monitoring and fault warning systems are crucial to improving the efficiency and operation of wind farms and reducing maintenance costs. Gearbox is the major component that leads to turbine downtime. Its failures are mainly caused by the gearbox bearings. Devising condition monitoring approaches for the gearbox bearings is an effective predictive maintenance measure that can reduce downtime and cut maintenance cost. In this paper, we propose a hybrid intelligent condition monitoring and fault warning system for wind turbine's gearbox. The proposed framework encompasses the following: a) clustering filter- (based on power, rotor speed, blade pitch angle, and wind speed signals)-using the automatic clustering model and ant bee colony optimization algorithm (ABC), b) prediction of gearbox bearing temperature and lubrication oil temperature signals- using variational mode decomposition (VMD), group method of data handling (GMDH) network, and multi-verse optimization (MVO) algorithm, and c) anomaly detection based on the Mahalanobis distances and wavelet transform denoising approach. The proposed condition monitoring system was evaluated using 10 min average SCADA datasets of two 2 MW on-shore wind turbines located in the south of Sweden. The results showed that this strategy can diagnose potential anomalies prior to failure and inhibit reporting alarms in healthy operations.

Research Progress of Rotating Machinery Fault Diagnosis Based on Deep Learning

In modern production, the precision and the importance of rotating machinery is higher and higher in the direction of large-scale, high speed and automation development, so that the traditional fault diagnosis methods are insufficient to deal with massive, multi-source and high-dimensional data, cannot meet the requirements of security and reliability. Therefore, several typical deep learning models are briefly introduced at first and the application of deep learning in fault diagnosis of rotor system, gear box and rolling bearing in recent years is studied and analyzed based on its strong feature extraction ability and advantages of clustering analysis. Finally, the advantages and disadvantages of deep learning model are summarized and the fault diagnosis methods of rotating machinery are summarized and prospected based on engineering practice.

Cross-Domain Intelligent Fault Diagnosis Method of Rotating Machinery Using Multi-Scale Transfer Fuzzy Entropy

A novel rotating machine fault diagnosis method based on multi-scale transfer fuzzy entropy (MTFE) and support vector machine (SVM) is proposed in this paper. Compared with traditional machine learning methods, our proposed method can identify various fault types of rotating machinery with different data distribution by learning the transfer knowledge from different distribution source domains. First, multi-scale fuzzy entropy (MFE) features of all samples are extracted as input. Second, build a transfer learning model to find a projection matrix to map the MFE features of the source and target domains into a common subspace, called MTFE features. In this process, the distribution structure of training data is maintained and the distribution difference between training data and test data is reduced. Finally, the SVM classifier identifies the fault type of the test data. Two cases including gearbox and rolling bearing are used for validation. Experimental results demonstrate that our proposed MTFE method performs best in recognizing various fault types comparing with other six methods.

The Enhancement of Weak Bearing Fault Signatures by Stochastic Resonance with a Novel Potential Function

As the critical parts of wind turbines, rolling bearings are prone to faults due to the extreme operating conditions. To avoid the influence of the faults on wind turbine performance and asset damages, many methods have been developed to monitor the health of bearings by accurately analyzing their vibration signals. Stochastic resonance (SR)-based signal enhancement is one of effective methods to extract the characteristic frequencies of weak fault signals. This paper constructs a new SR model, which is established based on the joint properties of both Power Function Type Single-Well and Woods-Saxon (PWS), and used to make fault frequency easy to detect. However, the collected vibration signals usually contain strong noise interference, which leads to poor effect when using the SR analysis method alone. Therefore, this paper combines the Fourier Decomposition Method (FDM) and SR to improve the detection accuracy of bearing fault signals feature. Here, the FDM is an alternative method of empirical mode decomposition (EMD), which is widely used in nonlinear signal analysis to eliminate the interference of low-frequency coupled signals. In this paper, a new stochastic resonance model (PWS) is constructed and combined with FDM to enhance the vibration signals of the input and output shaft of the wind turbine gearbox bearing, make the bearing fault signals can be easily detected. The results show that the combination of the two methods can detect the frequency of a bearing failure, thereby reminding maintenance personnel to urgently develop a maintenance plan.

Prognosis of Wind Turbine Gearbox Bearing Failures using SCADA and Modeled Data


 
 
 Predictive maintenance and condition monitoring systems for wind turbines have seen increased adoption to minimize downtime, reducing operation and maintenance costs. On today’s wind power plants, the integrated supervisory control and data acquisition (SCADA) system provides low- frequency operational data that can be leveraged to quantify a wind turbine’s health. The aim of this study is to utilize machine-learning techniques to predict axial cracking failures in wind turbine gearbox bearings up to 1 month ahead of time. The failures are assumed to have occurred when the investigated bearing was replaced. While current SCADA systems show the overall condition of a wind turbine, often they do not allow for the investigation of specific gearbox bearings’ health. To enrich bearing fault signatures, additional data are computed through physics-based models using gearbox design information. Based on SCADA data, modeled data, and bearing failure log data from an actual wind plant, the performances of different machine-learning models on unseen data are then evaluated using industry-standard metrics such as precision, recall, and F1 score. Results show the overall system performance enhancement in predicting bearing failure when modeled data are included with SCADA data. The reduction in terms of false alarms is about 50%, and improvement in terms of precision and F1 score is about 33% and 12% respectively, based on the best modeling case in this study.
 
 

Feature Extraction for Bearing Fault Detection Using Wavelet Packet Energy and Fast Kurtogram Analysis

An integrated method for fault detection of bearing using wavelet packet energy (WPE) and fast kurtogram (FK) is proposed. The method consists of three stages. Firstly, several commonly used wavelet functions were compared to select the appropriate wavelet function for the application of WPE. Then the analyzed signal is decomposed using WPE and the energy of each decomposed signal is calculated and selected for signal reconstruction. Secondly, the reconstructed signal is analyzed by FK to select the best central frequency and bandwidth for the band-pass filter. Finally, the filtered signal is processed using the squared envelope frequency spectrum and compared with the theoretical fault characteristic frequency for fault feature extraction. The procedure and performance of the proposed approach are illustrated and estimated by the simulation analysis, proving that the proposed method can effectively extract the weak transients. Moreover, the analysis results of gearbox bearing and rolling bearing cases show that the proposed method can provide more accurate fault features compared with the individual FK method.

Simulation of Vibration Propagated by Rolling Element Bearings in Mechanical Gearbox

Simulation of Vibration Propagated by Rolling Element Bearings in Mechanical Gearbox

False Alarms Analysis of Wind Turbine Bearing System

Wind turbines are complex systems that use advanced condition monitoring systems for analyzing their health status. The gearbox is one of the most critical components due to its elevated downtime and failure rate. Supervisory Control and Data Acquisition systems are employed in wind farms for condition monitoring and control in real time. The volume and variety of the data require novel and robust techniques for data analysis. The main novelty of this work is the development of a new modelling of the temperature curve of the gearbox bearing versus wind speed to detect false alarms. An approach based on data partitioning and data mining centers is employed. The wind speed range is divided into intervals to increase the accuracy of the model, where the centers are considered representative samples in the modelling. A method based on the alarm detection is developed and studied together with the alarms report provided by a real case study. The results obtained allow the identification of critical alarm periods outside the confidence interval. It is validated that the study of alarm identification, pre-filtered data, state variable, and output power contribute to the detection of the false alarms.

Fault Diagnosis of Bearing in Wind Turbine Gearbox Under Actual Operating Conditions Driven by Limited Data With Noise Labels

The fault characteristics of the rolling bearings of wind turbine gearboxes are unstable under actual operating conditions. Problems such as inadequate fault sample data, imbalanced data types, and noise labels (error labels) in historical data occur. Consequently, the accuracy of wind turbine gearbox bearing fault diagnosis under actual operating conditions is insufficient. Hence, a new method for the fault diagnosis of wind turbine gearbox bearings under actual operating conditions is proposed. It uses an improved label-noise robust auxiliary classifier generative adversarial network (rAC-GAN) driven by the limited data. The improved rAC-GAN realizes a batch comparison between the generated and real data to ensure the quality of the generated data and improve the generalization capability of the model in scenarios of actual operating conditions. It can be used to generate a large number of multitype fault data that satisfy the characteristics of the probability distribution of real samples and display higher robustness to label noises. Experiments indicate that, compared with other methods, the new method exhibits a higher accuracy in the multistate classification of rolling bearings under actual operating conditions when driven by the limited data with noise labels.

Wind turbine condition monitoring based on a novel multivariate state estimation technique

With the development of the wind energy industry, condition monitoring (CM) has become one of the important ways to improve the reliability of wind turbines (WTs). A data driven WTCM method based on a novel multivariate state estimation technique (MSET) is proposed, which can realize the fault early warning of WT components. In order to reduce the information redundancy of operational parameters, the maximum conditional mutual information (CMI) feature selection algorithm is applied to the training data of MSET. To improve the performance and flexibility of MSET, a dynamic memory matrix (MM) construction method based on k-nearest neighbor algorithm is proposed, which can provide a dynamic MM that varies in real-time with the current operational condition. A fast calculation method of the dynamic MM is also proposed, which can reduce the computation time by 15–26%. Two residual-based CM methods are proposed to realize fault early warning. The real-time method is based on the real-time residuals and if several consecutive residuals exceed the threshold, the fault alerts will be issued. The long-term method divides the historical residuals into day-level and analyzes them based on control charts. The proposed method is applied to two real cases of the gearbox and generator bearing overheating faults. The experimental results show that the maximum CMI algorithm and the dynamic MM significantly improve the estimation error of MSET. Compared with the recorded fault information, the proposed method realizes the fault early warning about 2–20 days in advance.

Geometric and mechanical parameters for the adjustment of the preload of differential bearings in gearboxes of automobile axles

This article discusses the construction of bearing units of different cars differentials and provides the analysis of technical requirements for the adjustment of the preload of bearings. There are four most commonly used methods for the adjustment of the preload. This paper defines a method for the adjustment of the preload based on deformation of bearing seats, which is the most acceptable in production and which is characterized by the least indirect adjustment and consequently the least number of failures. A formula is given for the calculation of the required preload force based on the failure in opening the joint in the unloaded bearing. A description of industrial equipment for the implementation of this method is presented.

Fault Diagnosis for Rotating Machinery Using Multiscale Permutation Entropy and Convolutional Neural Networks.

In view of the limitations of existing rotating machine fault diagnosis methods in single-scale signal analysis, a fault diagnosis method based on multi-scale permutation entropy (MPE) and multi-channel fusion convolutional neural networks (MCFCNN) is proposed. First, MPE quantitatively analyzes the vibration signals of rotating machine at different scales, and obtains permutation entropy (PE) to construct feature vector sets. Then, considering the structure and spatial information between different sensor measurement points, MCFCNN constructs multiple channels in the input layer according to the number of sensors, and each channel corresponds to the MPE feature sets of different monitored points. MCFCNN uses convolutional kernels to learn the features of each channel in an unsupervised way, and fuses the features of each channel into a new feature map. At last, multi-layer perceptron is applied to fuse multi-channel features and identify faults. Through the health monitoring experiment of planetary gearbox and rolling bearing, and compared with single channel convolutional neural networks (CNN) and existing CNN based fusion methods, the proposed method based on MPE and MCFCNN model can diagnose faults with high accuracy, stability, and speed.

Calculating loads and life-time reduction of wind turbine gearbox and generator bearings due to shaft misalignment

During the last two decades, wind turbine industries have faced high failure rates, downtimes and costly repairs. Gearbox and generator have contributed to this, especially, because their high speed shaft bearings have often failed. In this article, an analytical method is proposed to calculate the reaction loads of flexible connecting couplings installed between wind turbine gearbox and generator. Raction loads are determined from joint kinematics and metal disk pack deformations as well as axial and angular shaft misalignment. The calculations are executed for both flexible connecting couplings and a universal joint shaft and applied to the gearbox high speed shaft. The performance of flexible connecting couplings and universal joint shaft is compared with respect to the bearing loads and life-time of the gearbox high speed shaft. It is shown that the early, unplanned bearing failures of gearbox and generator high speed shaft can often be attributed to the flexible connecting couplings installed between gearbox and generator.

Validation of combined analytical methods to predict slip in cylindrical roller bearings

This paper presents a combination of models that together calculate the cage and roller speeds of a cylindrical roller bearing. The models consider elastohydrodynamic lubrication and contact elasticity between the roller and raceway, roller centrifugal forces, hydrodynamic lubrication at the cage pocket, and frictional forces. Using these models, the predicted cage and roller speeds and the extent of slip are compared to measurements acquired on cylindrical roller bearings in a commercial gearbox in steady-state and transient operating conditions of a wind turbine. In steady-state conditions at low wind speeds and low lubricant temperatures, cage and roller slip up to 60% occur in the loaded zone of the bearing. In the unloaded zone, up to 80% roller slip occurs. Cage and roller slip then decrease as the lubricant temperature and wind speed increases. In general, the analytical model results match experimental measurements within 10% for lubricant temperatures above 40°C and wind speeds over 10 meters per second. The analytical model is further evaluated during a transient start-up event and highly dynamic emergency stop event and also used to examine changes in the bearing design or lubricant properties. Roller and cage slip in cylindrical bearings is a combined effect of the bearing design, applied load, shaft speed, and lubricant properties and temperature, and can be quickly evaluated with the combined analytical models.