Wind Turbine Failure Research Articles

A sensorless active control approach to mitigate fatigue loads arising from the torsional and blade edgewise vibrations in PMSG-based wind turbine system

Fatigue loads associated with torsional and blade-edgewise vibrations are recognized as major reasons of large scale wind turbine failures. This paper presents an active mitigating control approach to reduce torsional and edgewise vibrations in the PMSG-based wind turbine. Due to the interactions of both vibrations, the active control is designed based on the expanded dynamic model of the mechanical-structural flexibilities. In this way, the blades are modeled as flexible cantilever beams, and thus, in the drive-train model the dynamics of the blades are considered. The control input to suppress the vibrations is achieved by the power electronic converters through adding an axillary term proportional to the speed difference between the blade and hub into the power control loop. This term manipulates the generator torque to counteract the unwanted vibrations. The needed drive-train and mechanical variables for making the auxiliary term are estimated by a novel robust method based on the sliding-mode observer. Using mathematical analyses and simulation results, it is shown that the proposed active mitigating approach well alleviates the fatigue loads of the torsional and edgewise modes even under severe changes in the drive-train stiffness coefficients. Thus, the proposed sensorless control approach is an appropriate remedy action for alleviation of WT fatigue loads even under parameters uncertainties.

Yaw system restart strategy optimization of wind turbines in mountain wind farms based on operational data mining and multi-objective optimization

The wind resources of mountain wind farms are affected by more complex terrain than are found at flat wind farms. Wind turbine (WT) failures caused by frequent operation of the yaw system occur often in mountain wind farms, resulting in significantly reduced economic benefits over the WT life cycle. Considering the problem of frequent yaw motions of WTs in mountain wind farms, this study introduces multi-objective optimization theory into the yaw system restart strategy optimization of WTs for the first time and proposes an optimized yaw system restart strategy with wind speed segmentation in the wind speed region below the rated wind speed. The yaw behavior of mountain wind farms in southern China is analyzed from multiple perspectives; a segmentation scheme of wind speed intervals for yaw control was determined from the results. The Non-dominated Sorting Genetic Algorithm-II (NSGA-II) is used to optimize the control parameters for each sub-interval with the goals of minimizing yaw operation time and maximizing energy generation. A set of Pareto solutions is obtained and evaluated with the Technique for Order of Preference by Similarity to an Ideal Solution (TOPSIS) method to produce the optimal solution. The simulation results indicated that after the WT adopts the optimized yaw system restart strategy in the wind speed region below the rated wind speed, the number of yaw motions is decreased by 82.9%, with only a slight decrease in energy generation. The optimization method proposed in this study can significantly reduce yaw operation time with minimal loss of energy generation, which provides a new approach to balancing the relationship between reducing the WT failure rate and ensuring power generation efficiency and is expected to significantly improve the economic benefits to wind farms.

A Bayesian reliability analysis exploring the effect of scheduled maintenance on wind turbine time to failure

AbstractThis article presents a Bayesian reliability modelling approach for wind turbines that incorporates the effect of time‐dependent variables. Namely, the technique is used to explore the effect of annual services on wind turbine failure intensity through time for turbines within a currently operational wind farm. In the operator's experience, turbines seemed to fail more frequently after scheduled maintenance was performed; however, this is an unexplored effect in the literature. Additionally, the effects of seasonality, year of operation and position in the array on failure intensity are explored. These features were included in a Cox‐like model formulation which allows for time‐dependent covariates. Inference was performed via Bayes rule. Results show a spike in failure intensity reaching 1.57 times the baseline in the six days directly proceeding annual servicing, after which failure intensity is reduced compared to baseline. Also observed is a significant year‐on‐year reduction of failure intensity since the introduction of the site's data management system in 2018, a clear preference for modelling time to failure via a Weibull distribution and a dependence on location in the array with respect to the prominent wind direction. Results also show the benefit of employing a Bayesian regime, which provides easily interpretable uncertainty quantification.

Multiaxial Fatigue Analysis of Jacket-Type Offshore Wind Turbine Based on Multi-Scale Finite Element Model.

The fatigue damage of a local joint is the key factor accounting for the structural failure of a jacket-type offshore wind turbine. Meanwhile, the structure experiences a complex multiaxial stress state under wind and wave random loading. This paper aims to develop a multi-scale modeling method for a jacket-type offshore wind turbine, in which local joints of the jacket are modeled in a detail by using solid elements, and other components are modeled via the common beam element. Considering the multiaxial stress state of the local joint, multi-axial fatigue damage analysis based on the multiaxial S-N curve is performed using equivalent Mises and Lemaitre methods. The uniaxial fatigue damage data of the jacket model calculated using the multi-scale finite element model are compared with those of the conventional beam model. The results show that the tubular joint of jacket leg and brace connections can be modeled using the multi-scale method, since the uniaxial fatigue damage degree can reach a 15% difference. The comparison of uniaxial and multiaxial fatigue results obtained using the multi-scale finite element model shows that the difference can be about 15% larger. It is suggested that the multi-scale finite element model should be used for better accuracy in the multiaxial fatigue analysis of the jacket-type offshore wind turbine under wind and wave random loading.

Research on Wind Turbine Blade Damage Fault Diagnosis Based on GH Bladed

With the increasing installed capacity of wind turbines, ensuring the safe operation of wind turbines is of great significance. However, the failure of wind turbines is still a severe problem, especially as blade damage can cause serious harm. To detect blade damage in time and prevent the accumulation of microdamage of blades evolving into severe injury, a damage dataset based on GH Bladed simulation of blade damage is proposed. Then, based on the wavelet packet analysis theory method, the MATLAB software can automatically analyze and extract the energy characteristics of the signal to identify the damage. Finally, the GH Bladed simulation software and MATLAB software are combined for fault diagnosis analysis. The results show that the proposed method based on GH Bladed to simulate blade damage and wavelet packet analysis can extract damage characteristics and identify single-unit damage, multiple-unit damage, and different degrees of damage. This method can quickly and effectively judge the damage to wind turbine blades; it provides a basis for further research on wind turbine blade damage fault diagnosis.

Analysis of Wind Turbine Equipment Failure and Intelligent Operation and Maintenance Research

Power generation from wind farms is growing rapidly around the world. In the past decade, wind energy has played an important role in contributing to sustainable development. However, wind turbines are extremely susceptible to component damage under complex environments and over long-term operational cycles, which directly affects their maintenance, reliability, and operating costs. It is crucial to realize efficient early warning of wind turbine failure to avoid equipment breakdown, to prolong the service life of wind turbines, and to maximize the revenue and efficiency of wind power projects. For this purpose, wind turbines are used as the research object. Firstly, this paper outlines the main components and failure mechanisms of wind turbines and analyzes the causes of equipment failure. Secondly, a brief analysis of the cost of wind power projects based on equipment failure is presented. Thirdly, the current key technologies for intelligent operation and maintenance (O&M) in the wind power industry are discussed, and the key research on decision support systems, fault diagnosis models, and life-cycle costs is presented. Finally, current challenges and future development directions are summarized.

A Novel Adaptive STFT-SFA Based Fault Detection Method for Nonstationary Processes

In the field of fault detection, the nonstationary characteristics caused by external disturbances of wind turbines and other reasons can mask the fault signals, while the inconsistent data distribution between training data and test data due to equipment loss and other reasons can lead to model mismatch problems, both of which can lead to the degradation of fault detection performance. In order to solve the above problems, a novel adaptive fault detection framework is proposed in this work. First, the stationary features of nonstationary variables are extracted based on short-time Fourier analysis, after which the features are combined with the stationary variables. Second, isolation-based anomaly detection using nearest-neighbor ensembles is introduced as monitoring metrics for designing the statistic for the slow feature analysis method. Then, the differences between online normal data and training data are calculated, the model update factor and update strategy are designed, and an adaptive fault detection framework based on short-time Fourier transform-slow feature analysis (STFT-SFA) is proposed. Finally, the effectiveness of the proposed fault detection framework is verified using the Tennessee-Eastman process and actual wind turbine data. The results show that the proposed STFT-SFA method has a 94.0% correct monitoring rate (CMR) for Tennessee Eastman process failures, and the proposed adaptive STFT-SFA method has a 95.6% CMR for wind turbine failures, which are better than other comparative fault detection methods.

Abnormality Detection Method for Wind Turbine Bearings Based on CNN-LSTM

Wind turbine energy generators operate in a variety of environments and often under harsh operational conditions, which can result in the mechanical failure of wind turbines. In order to ensure the efficient operation of wind turbines, the detection of any abnormality in the mechanics is particularly important. In this paper, a method for detecting abnormalities in the bearings of wind turbine energy generators, based on the cascade deep learning model, is proposed. First, data on the mechanics of wind turbine generators were collected, and the correlation between the data was studied in order to select the parameters related to the bearing temperature. Then, the logical relationship between the observation parameters and the target parameters was established based on a one-dimensional convolutional neural network (CNN) and a long short-term memory (LSTM) network, and the difference between the predicted temperature and the actual temperature was assessed using the root mean square error evaluation model. Finally, a numerical example was used to verify the operational data from a wind farm unit in northwest China. The results show that the CNN-LSTM model proposed in this paper can detect abnormalities earlier in the state of the main bearing than the LSTM model, and the CNN-LSTM model can detect abnormalities in the main bearing that the LSTM network cannot find.

A novel method for increasing wind power output considering wind turbine failure

This work develops a complete process for power capture and power distribution. In our proposed approach, a power allocation model that considers wind turbine (WT) failure is presented using a new maximum power capture structure for wind farms (WFs). Specifically, the proposed approach adopts a hierarchical control strategy to maximize the wind power output from the WF. In the first level, a control strategy is employed between a centralized controller located in the substation and local controllers installed on each WT to maximize wind energy capture. Apart from this, we exploit a new competitive swarm optimizer with a local search (LSCSO) to implement the optimization process of maximum power capture at this level. In the second level, a new power allocation model that considers WT failure is presented, and the power distribution is optimized using a distributed split Bregman method (SBM). Finally, the proposed approach integrates the LSCSO-SBM method and yields improved wind-farm power optimization compared to traditional methods. Furthermore, this scheme has been applied to the Roy Billinton Test System to evaluate the reliability of the generating system, and the results demonstrate that the proposed approach can effectively reduce the impact of the wake effect and total electrical losses in WFs, especially the significant benefit of increasing the energy production of WFs and generating system reliability.

Active learning-based wind turbine failure prediction with consideration of data from different wind farms

The SCADA system installed in each wind farm provides real-time and historical data required for fault diagnosis. However, labeling the data requires time and effort. Meanwhile, the performance of failure prediction and diagnostic relies on the volume of labeled failure data. Therefore, a fault diagnosis and prediction scheme for wind turbines based on active learning is proposed. Firstly, historical data from a selected wind turbine is screened out by residual analysis to set up the initial training data sets. Then a classifier is pre-trained by an improved active learning method. The active learning model based on the committee selection (QBC) sampling strategy and random forest (RF) learner is combined to realize the data labeling and, therefore, wind turbine fault detection. Afterward, the pre-trained classifier is applied to different wind turbines in the same/different wind farms to decide their operational status by labeling their operational data. Finally, the labeled data from different resources are fed back to retrain the classifier to improve its performance. Data from the SCADA system of wind turbines of different manufacturers in different wind farms are used as test cases, and simulation results show the effectiveness and feasibility of the proposed scheme.

Wind Turbine Fault Detection With Multimodule Feature Extraction Network and Adaptive Strategy

Unexpected failures in wind turbines (WTs) lead to tremendous economic losses and safety hazards to wind farms. Intelligent condition monitoring and fault detection can prevent unscheduled downtime and improve reliability and efficiency. However, the variational external environments and operation states result in time-varying correlations and fluctuations of variables, which pose challenges to precise modeling and detection for WTs. This article proposes a dual-stage temperature monitoring and fault detection method, which can be applied to multiple key components in WTs. First, a novel multimodule feature extraction network (MMFENet) is designed to model the normal behavior of temperature data in the monitoring component. The proposed model decomposes the correlations into distinct patterns and devises different modules for quantification. Specifically, the static module aims to characterize the stationary correlations resulting from the inherent structure to provide reference expectations of fluctuations in the dynamic module and accelerate the model training. The dynamic module is devoted to depicting time-varying interdependence caused by variational operation states to pursue precise modeling. In addition, the proposed network also uses a heterogeneous data fusion module to use external information. Second, an adaptive detection strategy is devised for converting the prediction errors into detection results. Based on the evaluation results in the training procedure, parameter ranges during online detection can be adaptively determined to contain multiple rational values, which can enhance robustness and scalability. Experimental results using field data from real wind farms demonstrate the effectiveness and superiority of the proposed method.

Field‐data‐based reliability analysis of power converters in wind turbines: Assessing the effect of explanatory variables

AbstractFrequent failures of power converters affect the availability of wind turbines and cause considerable maintenance costs. To enhance the reliability of power converters in wind turbines, the prevailing causes and modes of failures have to be identified. This publication contributes to root‐cause analysis of the power‐converter failures in wind turbines from a statistical point of view. For this purpose, the failure behavior of power‐converters is modeled via lifetime models as well as repairable‐system models. By means of regression models, covariates are incorporated, including both design‐related and site‐specific covariates. The analysis is based on a worldwide extensive field‐data collection covering more than 9000 turbines, including different turbine designs, sites, and ages. The results obtained by means of the applied regression models indicate that the location of the power converter within the turbine, the cooling system, the converter rated power, the DC‐link voltage, the IGBT‐module manufacturer, and the commissioning date of the turbine as design‐related covariates have a significant effect on the phase‐module failure behavior and with that on converter reliability. Among the site‐specific covariates, the analysis results confirm humidity as a likely significant driver of failures.

Detection of Wind Turbine Failures through Cross-Information between Neighbouring Turbines

In this paper, the time variation of signals from several SCADA systems of geographically closed turbines are analysed and compared. When operating correctly, they show a clear pattern of joint variation. However, the presence of a failure in one of the turbines causes the signals from the faulty turbine to decouple from the pattern. From this information, SCADA data is used to determine, firstly, how to derive reference signals describing this pattern and, secondly, to compare the evolution of different turbines with respect to this joint variation. This makes it possible to determine whether the behaviour of the assembly is correct, because they maintain the well-functioning patterns, or whether they are decoupled. The presented strategy is very effective and can provide important support for decision making in turbine maintenance and, in the near future, to improve the classification of signals for training supervised normality models. In addition to being a very effective system, it is a low computational cost strategy, which can add great value to the SCADA data systems present in wind farms.

Review of Tribological Failure Analysis and Lubrication Technology Research of Wind Power Bearings.

Wind power, being a recyclable and renewable resource, makes for a sizable portion of the new energy generation sector. Nonetheless, the wind energy industry is experiencing early failure of important components of wind turbines, with the majority of these issues also involving wind power bearings. Bearing dependability is directly tied to the transmission efficiency and work performance of wind turbines as one of its major components. The majority of wind turbine failures are due to bearings, and the vast majority of bearing failures are due to lubrication. The topic of improving the accuracy and life of wind power bearing motion is becoming increasingly essential as the wind power industry develops rapidly. This study examines the various constructions and types of wind turbines, as well as their bearings. We also examined the most typical causes of friction and lubrication failure. Furthermore, contemporary research on wind turbine bearings has been compiled, which mostly comprises the study and development of lubrication technology and other areas. Finally, a conclusion and outlook on current challenges, as well as future research directions, are offered.

Wind Turbine Gearbox Failure Detection Through Cumulative Sum of Multivariate Time Series Data

The wind energy industry is continuously improving their operational and maintenance practice for reducing the levelized costs of energy. Anticipating failures in wind turbines enables early warnings and timely intervention, so that the costly corrective maintenance can be prevented to the largest extent possible. It also avoids production loss owing to prolonged unavailability. One critical element allowing early warning is the ability to accumulate small-magnitude symptoms resulting from the gradual degradation of wind turbine systems. Inspired by the cumulative sum control chart method, this study reports the development of a wind turbine failure detection method with such early warning capability. Specifically, the following key questions are addressed: what fault signals to accumulate, how long to accumulate, what offset to use, and how to set the alarm-triggering control limit. We apply the proposed approach to 2 years’ worth of Supervisory Control and Data Acquisition data recorded from five wind turbines. We focus our analysis on gearbox failure detection, in which the proposed approach demonstrates its ability to anticipate failure events with a good lead time.

One-Shot Fault Diagnosis of Wind Turbines Based on Meta-Analogical Momentum Contrast Learning

The rapid development of artificial intelligence offers more opportunities for intelligent mechanical diagnosis. Fault diagnosis of wind turbines is beneficial to improve the reliability of wind turbines. Due to various reasons, such as difficulty in obtaining fault data, random changes in operating conditions, or compound faults, many deep learning algorithms show poor performance. When fault samples are small, ordinary deep learning will fall into overfitting. Few-shot learning can effectively solve the problem of overfitting caused by fewer fault samples. A novel method based on meta-analogical momentum contrast learning (MA-MOCO) is proposed in this paper to solve the problem of the very few samples of wind turbine failures, especially one-shot. By improving the momentum contrast learning (MOCO) and using the training idea of meta-learning, the one-shot fault diagnosis of wind turbine drivetrain is analyzed. The proposed model shows a higher accuracy than other common models (e.g., model-agnostic meta-learning and Siamese net) in one-shot learning. The feature embedding is visualized by t-distributed stochastic neighbor embedding (t-SNE) in order to test the effectiveness of the proposed model.

A novel cyclic-correntropy based indicator for gear wear monitoring

Gearbox is a vulnerable component of a turbine’s drivetrain and plays a vital role in the power transmission in wind turbines. Wind turbines usually operate under harsh working environments, such as in deserts, oceans, and on hills. The adverse operating conditions (such as inevitable fluctuating wind loads and speeds) make the gearbox transmission prone to reliability degradation and premature failure. Gear wear is a common and unavoidable surface degradation phenomenon during the lifespan of the gear transmission system. The gear wear propagation can result in severe failures, such as gear surface spalling, gear root crack, and gear tooth breakage, all of which could lead to the failure of the drivetrain system of wind turbines and bring unexpected economic loss, even serious accidents. Thus, it is crucial to monitor the gear wear propagation progression in order to enable reliable and safe operation. To this end, this paper develops a novel vibration-based health indicator to monitor the gear surface degradation induced by gear wear progression. With the help of the novel indicator developed, the health status of the gearbox can be well evaluated and thus predictive maintenance-based decisions can be made to reduce maintenance costs and minimize gearbox failures in wind turbines. A series of endurance tests under different lubrication conditions and operational conditions are carried out to verify the effectiveness of the gear wear monitoring indicator.

Missing wind speed data reconstruction with improved context encoder network

Missing wind speed data are mainly caused by harsh weather, wind turbine failures, and data transmission errors, which have adverse effects on the performance of wind power forecasting, power curve modeling, and energy assessment. Inspired by context encoders (CE), this paper proposes an improved context encoder network (ICE) for missing wind speed data reconstruction. An auto-encoder architecture with multiple one-dimensional convolutional layers is established for data generation. During network training, a joint loss function that includes reconstruction loss and adversarial loss is presented to obtain the stable and near-real reconstructed wind speed data. We add an Inception layer to the generator network to automatically select the appropriate convolutional filters and then recalibrate the channel relationship between feature maps via the squeeze-and-excitation network. At last, this paper uses wind speed data collected from an on-shore wind farm to verify the effectiveness of the proposed network. The results show that the mean absolute error (MAE) and root mean square error (RMSE) of the ICE network in different data missing rates are 0.019–0.021 and 0.021–0.025, respectively. It has the lowest reconstruction errors compared with six typical data reconstruction methods.

Incorporating Machine Learning into Vibration Detection for Wind Turbines

With machine learning techniques, wind turbine components can be detected and diagnosed in advance, so degeneration can be prevented. Automatic and autonomous learning is used to predict, detect, and diagnose electrical and mechanical failures in wind turbines. Based on the implementation of machine learning algorithms adapted to the different components and faults of wind turbines, this study evaluates different methodologies for monitoring, supervision, and fault diagnosis.

An abnormal wind turbine data cleaning algorithm based on color space conversion and image feature detection

Wind power curve (WPC) is established through data collected from the Supervisory Control and Data Acquisition (SCADA) system of each wind turbine, which can be used to analyze the operation status. However, numerous outliers are contained in SCADA data caused by wind turbine failures, shutdown maintenance or other extreme conditions to deform the wind power curve. This paper proposes a data cleaning algorithm for wind turbine abnormal data based on wind power curve image by color space conversion and image feature detection. Considering wind speed, wind power and data frequency, a three-dimensional (3D) WPC image is constructed. The scattered outliers are cleared by their statistical characteristics. The Canny edge detection and Hough transform are introduced to extract image features of stacked outliers and locate them accurately. The proposed algorithm is compared with three common outlier detection algorithms, including two data-based algorithms and an image-based algorithm. Extensive experiments conducted on the data of 22 wind turbines from two different wind farms in China indicate the efficiency, stability and reliability of the proposed algorithm.