基于优化经验小波变换与改进支持向量机的齿轮箱故障检测策略

针对强噪声背景下风力机齿轮箱振动信号易被掩盖、难以提取的难题,基于频域谱负熵（Frequency-domain Spectral Negentropy,FSN）改进经验小波变换（Empirical Wavelet Transform,EWT）提出优化经验小波变换方法（Improved Empirical Wavelet Transform,IEWT）,并采用改进灰狼算法（Improved Grey Wolf Optimization,IGWO）优化支持向量机（Support Vector Machine,SVM）惩罚系数α及核参数σ。基于NREL GRC风力机齿轮箱数据验证所提方法的有效性。结果表明：IEWT-IGWO-SVM可有效提取故障信息并进行故障识别,分类准确率高达99.66%。     

Monitoring wind turbine gearboxes

Concerns amongst wind turbine (WT) operators about gearbox reliability arise from complex repair procedures, high replacement costs and long downtimes leading to revenue losses. Therefore, reliable monitoring for the detection, diagnosis and prediction of such faults are of great concerns to the wind industry. Monitoring of WT gearboxes has gained importance as WTs become larger and move to more inaccessible locations. This paper summarizes typical WT gearbox failure modes and reviews supervisory control and data acquisition (SCADA) and condition monitoring system (CMS) approaches for monitoring them. It then presents two up‐to‐date monitoring case studies, from different manufacturers and types of WT, using SCADA and CMS signals. The first case study, applied to SCADA data, starts from basic laws of physics applied to the gearbox to derive robust relationships between temperature, efficiency, rotational speed and power output. The case study then applies an analysis, based on these simple principles, to working WTs using SCADA oil temperature rises to predict gearbox failure. The second case study focuses on CMS data and derives diagnostic information from gearbox vibration amplitudes and oil debris particle counts against energy production from working WTs. The results from the two case studies show how detection, diagnosis and prediction of incipient gearbox failures can be carried out using SCADA and CMS signals for monitoring although each technique has its particular strengths. It is proposed that in the future, the wind industry should consider integrating WT SCADA and CMS data to detect, diagnose and predict gearbox failures.Copyright © 2012 John Wiley & Sons, Ltd.     

Using transfer path analysis to assess the influence of bearings on structural vibrations of a wind turbine gearbox

Noise and vibration issues can be dealt with using several approaches. Using the source–transfer path–receiver approach, a vibration issue could be solved by attenuating the source, modifying the transfer path or by influencing the receiver. Applying this approach on a wind turbine gearbox would respectively correspond with lowering the gear excitation levels, modifying the gearbox housing or by trying to isolate the gearbox from the rest of the wind turbine. This paper uses a combination of multi‐body modelling and typical transfer path analysis (TPA) to investigate the impact of bearings on the total transfer path and the resulting vibration levels. Structural vibrations are calculated using a flexible multi‐body model of a three‐stage wind turbine gearbox. Because the high‐speed mesh is often the main source of vibrations, focus is put on the four bearings of this gear stage. The TPA method using structural vibration simulation results shows which bearing position is responsible for transmitting the highest excitation levels from the gears to the gearbox housing structure. Influences of bearing stiffness values and bearing damping values on the resulting vibration levels are investigated by means of a parameter sensitivity study and are confirmed with the results from the TPA. Because both the TPA and the parameter sensitivity analysis revealed a big influence on radial stiffness for a certain bearing, this was investigated in more detail and showed the big importance of correct axial bearing position. The main conclusions of this paper are that the total vibration behaviour of a wind turbine gearbox can be altered significantly by changing both bearing properties such as stiffness, damping and position, and bearing support stiffness. Copyright © 2014 John Wiley & Sons, Ltd.     

Multi‐body modelling and analysis of a planet carrier in a wind turbine gearbox

There have been some recent efforts to numerically model and analyse the wind turbine gearbox. To date, much of the focus has been on increasing model refinement and demonstrating its added value. This paper takes a step back and examines in detail the modelling and analysis of an important wind turbine gearbox component, the planet carrier, in a multi‐body setting. The planet carrier studied in this work comes from the 750 kW wind turbine gearbox used in the National Renewable Energy Laboratory's Gearbox Reliability Collaborative project. The study is performed in two parts. First, the influence of subcomponents mated to the planet carrier in the gearbox assembly is investigated in detail. These components consist of the planet pins, bearings and the main shaft. In the second part of the study, the flexible body modelling of the planet carrier for use in multi‐body simulations is examined through the use of condensed finite element and multi‐body simulation models. Both eigenvalue analyses and time domain simulations are performed. Comparisons are made regarding the eigenfrequencies, categorized mode shapes and the maximum and minimum planet carrier rim deflections from the time domain simulations. The mode shapes are categorized into seven distinct deformation patterns. An actual load case from the dynamometer tests, a 100% rated torque loading, is used in the time domain simulations. The results from this comprehensive study provide an insight into the proper modelling of a wind turbine planet carrier in a multi‐body setting. Copyright © 2012 John Wiley & Sons, Ltd.     

Improving wind turbine drivetrain designs to minimize the impacts of non‐torque loads

Non‐torque loads induced by the wind turbine rotor overhang weight and aerodynamic forces can greatly affect drivetrain loads and responses. If not addressed properly, these loads can result in a decrease in gearbox component life. This work uses analytical modeling, computational modeling and experimental approaches to evaluate two distinct drivetrain designs that minimize the effects of non‐torque loads on gearbox reliability: a modified three‐point suspension drivetrain studied by the National Renewable Energy Laboratory (NREL) Gearbox Reliability Collaborative (GRC) and the Pure Torque® drivetrain developed by Alstom. In the original GRC drivetrain, the unequal planetary load distribution and sharing were present and they can lead to gear tooth pitting and reduce the lives of the planet bearings. The NREL GRC team modified the original design of its drivetrain by changing the rolling element bearings in the planetary gear stage. In this modified design, gearbox bearings in the planetary gear stage are anticipated to transmit non‐torque loads directly to the gearbox housing rather than the gears. Alstom's Pure Torque drivetrain has a hub support configuration that transmits non‐torque loads directly into the tower rather than through the gearbox as in other design approaches. An analytical model of Alstom's Pure Torque drivetrain provides insight into the relationships among turbine component weights, aerodynamic forces and the resulting drivetrain loads. In Alstom's Pure Torque drivetrain, main shaft bending loads are orders of magnitude lower than the rated torque and hardly affected by wind speed, gusts or turbine operations. Copyright © 2014 John Wiley & Sons, Ltd.     

An engineering condition indicator for condition monitoring of wind turbine bearings

Condition monitoring (CM) of wind turbine becomes significantly important part of wind farms in order to cut down operation and maintenance costs. The large amount of CM system vibration data collected from wind turbines are posing challenges to operators in signal processing. It is crucial to design sensitive and reliable condition indicator (CI) in wind turbine CM system. Bearing plays an important role in wind turbine because of its high impact on downtime and component replacement. CIs for wind turbine bearing monitoring are reviewed in the paper, and the advantages and disadvantages of these indicators are discussed in detail. A new engineering CI (ECI), which combined the energy and kurtosis representation of the vibration signal, is proposed to meet the requirement of easy applicability and early detection in wind turbine bearing monitoring. The quantitative threshold setting method of the ECI is provided for wind turbine CM practice. The bearing run‐to‐failure experiment data analysis demonstrates that ECI can evaluate the overall condition and is sensitive to incipient fault of bearing. The effectiveness in engineering of ECI is validated though a certain amount of real‐world wind turbine generator and gearbox bearing vibration data.     

基于油液光谱监测的风电机组齿轮箱磨损状态的研究

风电机组齿轮箱的磨损微粒主要是铁颗粒,铁颗粒含量的增长趋势能直接反映出风电机组齿轮箱的磨损状态.以Spectro油液光谱分析仪监测风电机组齿轮箱在用齿轮油中的铁元素含量,通过一段时间内铁元素的增加量和风电机组可利用小时数,可计算得到单位可利用小时数下的铁元素增加量ΔQFe;引入可靠性理论研究了ΔQFe的分布规律,并以风...     

Wind turbine gearbox failure and remaining useful life prediction using machine learning techniques

This research investigates the prediction of failure and remaining useful life (RUL) of gearboxes for modern multi‐megawatt wind turbines. Failure and RUL are predicted through the use of machine learning techniques and large amounts of labelled wind turbine supervisory control and data acquisition (SCADA) and vibration data. The novelty of this work stems from unprecedented access to one of the world's largest wind turbine operational and reliability databases, containing thousands of turbine gearbox failure examples and complete SCADA and vibration data in the build up to those failures. Through access to that data, this paper is unique in having enough failure examples and data to draw the conclusions detailed in the remainder of this abstract. This paper shows that artificial neural networks provide the most accurate failure and RUL prediction out of three machine learning techniques trialled. This work also demonstrates that SCADA data can be used to predict failure up to a month before it occurs, and high frequency vibration data can be used to extend that accurate prediction capability to 5 to 6 months before failure. This paper demonstrates that two class neural networks can correctly predict gearbox failures between 72.5% and 75% of the time depending on the failure mode when trained with SCADA data and 100% of the time when trained with vibration data. Data trends in the build up to failure and weighting of the SCADA data inputs are also provided. Lastly, this work shows how multi‐class neural networks demonstrate more potential in predicting gearbox failure when trained with vibration data as opposed to training with SCADA data.     

750 kW风力机组齿轮箱动力学仿真分析

风力发电机组系统运行时产生剧烈的振动,对齿轮箱的运行精度和齿轮寿命的影响非常大。针对这一情况,文章对齿轮箱进行了重新设计、建模。基于多体系统动力学方法、模态振动、冲击-接触理论,以750 kW型风机齿轮箱为研究对象,通过对齿轮箱的仿真分析,得出齿轮啮合和碰撞力以及动能随时间的变化曲线。文章还对高速齿轮轴进行了模态分析,得到弯曲模态振型图,并将高速齿轮轴、行星轮、行星架和齿轮Z1变为柔体进行应力分析,得到齿轮的应力分布图,为齿轮箱总体动力学特性分析及齿轮箱优化设计奠定基础。     

Effects of floating sun gear in a wind turbine's planetary gearbox with geometrical imperfections

This paper addresses the effect of gear geometrical errors in wind turbine planetary gearboxes with a floating sun gear. Numerical simulations and experiments are employed throughout the study. A National Renewable Energy Laboratory 750 kW gearbox is modelled in a multibody environment and verified using the experimental data obtained from a dynamometer test. The gear geometrical errors, which are both assembly dependent and assembly independent, are described, and planet‐pin misalignment and eccentricity are selected as the two most influential and key errors for case studies. Various load cases involving errors in the floating and non‐floating sun gear designs are simulated, and the planet‐bearing reactions, gear vibrations, gear mesh loads and bearing fatigue lives are compared. All tests and simulations are performed at the rated wind speed. For errorless gears, the non‐floating sun gear design performs better in terms of gear load variation, whereas the upwind planet bearing has more damage. In the floating sun gear scenario, the planet misalignment is neutralized by changing the sun motion pattern and the planet gear's elastic deformation. The effects of gear profile modifications are also evaluated, revealing that profile modifications such as crowning improve the effects of misalignment. Copyright © 2014 John Wiley & Sons, Ltd.     

风力机风轮模态局部化的动力学机理与影响因素分析

该文对风力机叶片损伤导致风轮模态局部化的动力学机理与影响因素进行研究。首先从代数特征值角度,揭示模态局部化的动力学机理,发现结构产生模态局部化的主要原因是存在密集模态。其次,建立NREL 5 MW风轮结构的有限元模型,分析叶片失谐度、模态阶数和失谐位置对风力机风轮结构模态局部化的影响。结果表明：叶片损伤失谐会造成叶片的振型发生显著变化,产生模态局部化现象;同时在某些模态下,系统振动能量集中于损伤叶片,会加速叶片损伤,致使其产生疲劳破坏。因此在风力机结构设计时,需考虑模态局部化对风力机结构的动力学特性影响。     

风电机组传动链典型故障特征提取方法

为解决风电机组传动链易发生故障的问题,文章阐述了风电机组齿轮箱特征频率的计算方法和基于振动信号分析的故障特征提取方法。结合实际情况,以行星级齿轮磨损、中间轴小齿轮崩齿、高速轴齿轮崩齿和发电机轴承电腐蚀等典型故障为例,通过齿轮箱特征频率和传动链典型故障振动信号基本特征分析,可较好地完成故障识别。结果表明,采用经典信号处理方法能对上述典型故障进行特征提取,验证了经典方法对单一、明显故障特征提取的有效性,为深入开展传动链故障特征提取方法研究奠定了基础,为风电机组故障检修维护提供了技术支撑。     

Finite element simulation of wind turbine aerodynamics: validation study using NREL Phase VI experiment

A validation study using the National Renewable Energy Laboratory (NREL) Phase VI wind turbine is presented. The aerodynamics simulations are performed using the finite element arbitrary Lagrangian–Eulerian–variational multiscale formulation augmented with weakly enforced essential boundary conditions. In all cases, the rotor is assumed to be rigid and its rotation is prescribed. The rotor‐only simulations are performed for a wide range of wind conditions, and the computational results compare favorably with the experimental findings in all cases. The sliding‐interface method is adopted for the simulation of the full wind turbine configuration. The full‐wind‐turbine simulations capture the blade–tower interaction effect, and the results of these simulations are also in good agreement with the experimental data. Copyright © 2013 John Wiley & Sons, Ltd.     

大型风力机齿轮传动系统机电耦合动态特性研究

针对目前风力机柔性齿轮箱动力学研究时简化电气系统的问题,以某8 MW永磁同步风力机为对象,建立包含详细电气系统和柔性传动链的风力机模型。基于该模型探究电气系统效应对风力机齿轮箱啮合刚度、动态接触应力、振动加速度等动力学特性影响。结果表明：电气系统效应使系统转速、时变啮合刚度、接触应力相位滞后且波动减小;电气系统效应抑制各级齿轮角加速度及箱体振动加速度高频成分并减小传动链振动;风速突变时,电气系统效应可减小高速级齿轮峰值啮合力,增强风力机传动链抵御冲击能力。     

Balancing fatigue damage and turbine performance through innovative pitch control algorithm

Improvements to current pitch control strategies are explored by analysing the addition of a dynamic peak‐shaving algorithm, called the load limiting algorithm (LLA). The goal of this dynamic peak‐shaving algorithm is to reduce blade fatigue damage caused by wind gusts without sacrificing energy capture. This paper introduces a multivariate procedure based on the Taguchi method to systematically test different controller configurations and evaluate the algorithm's effectiveness. The LLA was tested through numerical analysis and field experiments. Numerical studies were performed on the Controls Advanced Research Turbine (CART) and the National Renewable Energy Laboratory (NREL) 5 MW model. Field testing was conducted on the CART. The primary metrics of LLA effectiveness were blade fatigue damage, measured in 20 year damage equivalent loads (DEL) for root flap bending and annual energy production (AEP). Numerical results indicate a reduction in weighted flap bending DEL (Flap DEL) for both turbine platforms. The CART demonstrates a reduction of 3.2% with a half‐percent loss in AEP. The LLA is markedly more effective on the NREL 5 MW, demonstrating a reduction of 5–9% in Flap DEL with a drop of 1–3% in AEP. Secondary benefits such as DEL reductions for other components, operating extreme blade loads, and pitch duty cycle were also explored.Copyright © 2011 John Wiley & Sons, Ltd.     

Vibration response of a 2.3 MW wind turbine to yaw motion and shut down events

Structural health monitoring (SHM) is a process of implementing a damage detection strategy for a mechanical system. Wind turbine machinery stands to benefit from SHM significantly as the ability to detect early stages of damage before significant malfunction or structural failure occurs would reduce costs of wind power projects by reducing maintenance costs. Vibration analysis of dynamic structural response is an approach to SHM that has been successfully applied to mechanical and civil systems and shows some promise for wind turbine application. Traditionally, a setback to turbine vibration‐based SHM techniques has been the unavailability of turbine vibration response data. This study begins to address this issue by presenting vibration response for a commercial 2.3 MW turbine to a limited number of operating conditions. A database of acquired vibration response signals detailing turbine response to yaw motion, start‐up, operation and shutdown has been assembled. A Daubechies sixth‐order wavelet was used to perform an eight‐level discrete wavelet decomposition such that general trends and patterns within the signals could be identified. With further development, the presented analysis of vibration response may be integrated into routines to reduce downtime and failure frequency of utility scale wind turbines. Copyright © 2011 John Wiley & Sons, Ltd.     

Experimental identification of modal parameters of an industrial 2‐MW wind turbine

This contribution presents modal testing of a 2‐MW wind turbine on a 100‐m tubular tower with a 93‐m rotor developed by W2E Wind to Energy GmbH. This research is part of the DYNAWIND project of the University of Rostock and W2E. Beside classical modal analysis schemes, this contribution mainly focusses on the application of operational modal analysis techniques to a wind turbine. Specific problems are addressed, and hints for modal testing on wind turbines are given. Furthermore, an effective measurement setup is proposed for identification of the modal parameters of a wind turbine. The measurement campaign is divided in two parts. First, a measurement campaign using 8 sensor positions on a rotor blade was done while the rotor is lying on ground. Second, a detailed measurement campaign was done on the entire wind turbine with the rotor locked in Y position using 61 sensor positions on the tower, the mainframe, the gearbox, the generator, and the low‐voltage unit. While the rotor blade was tested by classical and operational modal analysis techniques, the entire wind turbine was tested by operational modal analysis techniques only. The mode shapes and eigenfrequencies of the wind turbine identified within the measurement campaigns are within the expected range of the design values of the wind turbine. But in contrast, the damping ratios differ strongly from those given in guidelines and literature. Furthermore, a strong influence of aerodynamic damping compared to structural damping is observed for the first tower mode even for a parked wind turbine.     

基于一维密集连接卷积网络的风电齿轮箱智能故障分类

人工智能技术的飞速发展为现代能源装备的精益化故障诊断与健康管理提供了可能。风电齿轮箱由多个齿轮、轴承组成,且长期在变速、变载荷工况下运行,依靠传统的故障特征提取结合机器学习方法进行故障诊断存在精度低、缺乏智能性等缺点。文章提出了基于一维密集连接卷积网络的风电齿轮箱故障分类方法:将原始振动信号直接送入网络模型,经过密集连接、合成连接与卷积运算,匹配对应的故障类型,迭代训练故障分类模型;振动信号输入模型后的分类结果决定所属故障类别。文章所提出的风电齿轮箱故障分类方法具有诊断流程简单、故障识别率高等特点,多工况试验台故障数据验证了该方法的有效性。     

基于改进小波包的风电机组齿轮箱复合故障特征提取研究

针对小波包分解振动信号时会产生频谱混叠从而导致齿轮箱复合故障特征能量谱提取困难的问题,提出基于旁路滤波改进小波包的方法对双馈风电机组齿轮箱复合故障振动信号进行研究,并以风电场的大量齿轮箱振动信号为基础,运用传统小波包及旁路滤波改进小波包分别对齿轮箱振动信号提取特征能量谱。实验结果表明：运用旁路滤波改进小波包对双馈风电机组齿轮箱复合故障振动信号进行分析,可有效避免传统小波包分析振动信号的频谱混叠现象,准确提取每种故障状态的特征能量谱。