基于温度参数的风电机组异常识别

文章针对风电机组运行过程中机组早期的异常状态识别问题,提出一种考虑有功功率的基于机组温度参数变化特性的风电机组异常识别模型。首先,分析风电机组各系统与温度相关的参数。然后,利用相关性理论,确定了与有功功率相关的温度参数:齿轮箱高速轴轴承前端温度、齿轮箱高速轴轴承后端温度、齿轮箱油温、发电机驱动端轴承温度、发电机非驱动端轴承温度、发电机定子绕组温度,形成了异常检测的参数体系。再次,以正常状态下机组温度参数的偏度和峰度的最大区间作为阈值,建立风电机组异常识别模型。最后,采用滑动窗口对机组运行状态进行在线监测。通过实例研究发现,当机组发生异常状态时,温度参数的偏度或者峰度超过了阈值,比警报提前了15 d。该识别模型为风电机组的早期故障预警提供了参考。     

基于工况辨识的风电机组齿轮箱状态监测

针对大型风电机组复杂多变的运行工况,采用单一固定阈值来评价风电机组运行状态误报警率高的问题,采用标准模糊C均值聚类进行运行工况识别;在各个子工况空间下,利用非线性状态估计方法建立齿轮箱正常工作状态下温度模型进行轴承温度估计;采用滑动窗口残差统计方法对残差进行分析,建立残差均值报警阈值。最后以某风电机组SCADA(数据采集与监视控制系统)数据进行应用研究。结果表明:基于工况辨识的模型可以准确估计齿轮箱温度并能够降低误报警率,可以实时在线监测齿轮箱运行状态。     

一种基于非线性偏最小二乘的风电机组齿轮箱状态监测方法北大核心CSCD

风电机组状态监测是提升机组运行水平和经济效益的重要手段。文章提出了一种基于非线性偏最小二乘(PLS)的风电机组齿轮箱状态监测方法,利用数据采集与监控系统(SCADA)数据对齿轮箱油温进行建模和监测。首先,基于无监督聚类对SCADA数据进行预处理,利用相关性分析选取与齿轮箱油温相关的输入变量;然后,构建用于表征非线性关系的输入变量,建立正常运行工况下齿轮箱油温的非线性PLS模型;最后,根据模型输出结果与齿轮箱油温的残差分布,设置合理阈值,用于齿轮箱状态监测。应用该模型对某大型风电机组齿轮箱进行状态监测。监测结果表明,相比于BP神经网络模型,该模型具有更高的拟合优度和预测精度。     

基于SCADA数据的风电机组齿轮箱状态监测方法

为解决故障劣化渐变过程的长时间序列对齿轮箱状态监测模型的影响问题,提升其决策精度,提出一种基于数据采集与监控(SCADA)数据的组合建模方法。首先,采用主成分分析法(PCA)选取与齿轮箱温度密切相关的输入观测向量,并应用长短期记忆(LSTM)神经网络分别对齿轮箱正常工况和异常工况独立建立温度模型;其次,结合模型输出结果与SCADA数据提取残差分布特征向量,建立随机森林残差分布模型对机组齿轮箱运行状态进行监测;最后,对某大型风电场机组进行模型建立和仿真研究。结果表明,基于LSTM神经网络结合随机森林算法对风电机组齿轮箱状态监测有较强的实用性和较高的准确率,为后续开展齿轮箱健康度评价提供了新的方法和思路。     

基于长短时记忆网络融合SCADA数据的风电齿轮箱状态监测

针对不具有时间记忆能力的机器学习方法融合风电机组数据采集与监控系统(SCADA)的时序数据而导致风电齿轮箱状态预测精度不高的问题,提出基于长短时记忆(LSTM)网络融合SCADA数据的风电齿轮箱状态预测模型。选择能表征风电齿轮箱运行状态的某个监测量作为模型的输出量,基于灰色关联度选择与该监测量关联密切的SCADA参数作为预测模型的输入量;使用正常状态下的SCADA数据训练LSTM预测模型,得出预测值和残差,通过3σ准则计算出上下预警阈值,用于风电齿轮箱状态监测和故障预警。某风电场风电齿轮箱的SCADA数据验证表明所提出的方法能有效预警风电齿轮箱故障。     

XGBoost算法在风电机组发电机故障监测预警中的应用研究

发电机作为整个风电机组的核心部件之一,其能否正常运行将严重影响风电机组的持续发电.利用集成机器学习算法中的梯度提升算法XGBoost对风电机组发电机故障监测预警进行研究.首先,提取数据采集与监控(SCADA)系统数据库中风电机组在并网状态下的正常运行大数据,对缺失、异常数据进行预处理后,结合运维专家经验利用XGBoos...     

基于迁移成分分析的多风电机组运行状态识别方法

风电机组运行状态识别对风电机组发电性能评估和风电场精细化管理具有重要意义,然而不同风电机组的数据采集与监视控制系统(supervisory control and data acquisition,SCADA)数据分布差异明显,如果将已训练好的单台风电机组正常行为模型直接应用于多风电机组运行状态辨识,辨识精度较低.为了...     

基于温度信号的风电机组发电机实时可靠性监测新方法

实时可靠性监测对降低风电机组的维护成本,提高其运行水平具有重要的意义。文章提出了一种基于温度信号监测的风电机组发电机实时可靠性监测方法。该方法采用正常运行状态下的发电机温度数据,利用线性回归技术,得出了消除环境温度和输出功率影响后的修正温度;然后基于性能可靠性理论,对修正温度进行标准正态变换,建立了发电机运行温度的实时可靠性监测模型。以某机组实际运行数据为例,验证所提方法,结果表明,该方法能利用温度信号监测机组运行的实时可靠度来发现其运行异常状态,达到实时可靠性监测目的。     

小型风力发电机电磁设计特点

文章分析了小型风电机组用的永磁同步发电机的电磁设计特点,并与其他普通永磁电机区别,详细介绍了永磁同步发电机各个电磁参数的特点以及对发电机性能的影响。重点介绍永磁同步发电机定子、转子设计的特点,对不同的极数与槽数的配合进行对比分析,总结降低齿槽转矩的规律。     

永磁直驱风力发电机瞬态温度特性研究

采用有限元法对大功率永磁直驱风力发电机定子三维瞬态温度场进行计算,得到了电机定子温度场。对发电机进行了型式试验,测试结果与计算的稳态结果较为接近,验证了计算的准确性。结果表明,电机在额定工况下,温升稳定后端部绕组温度高于中间绕组,最高温度点在背风侧的端部绕组,最高温度与平均温度相差16℃左右。将实际风场的风资源数据作为外界条件,对永磁直驱发电机一定时间跨度的瞬态温度进行计算,结果与SCADA系统采集的风力发电机运行数据较为接近,为绝缘系统设计和电机的优化提供了参考。     

半直驱永磁风力发电机散热性能影响因素研究

目的  随着风力发电技术的发展,永磁风力发电机单机容量不断增加,发热功率也随之增大,发电机散热正面临前所未有的挑战,如何有效地解决发电机温升高、散热难的问题一直是我们研究的重点。 方法  文章基于STAR-CCM+软件平台建立了发电机散热-冷却一体化数值模拟方法,通过与实验值对比验证了数值计算方法的可靠性。在此基础上对大功率永磁风力发电机内部温度分布规律以及冷却系统散热性能影响因素进行了研究。 结果  在额定功率下,绕组、定子铁芯的最高温度出现在中层区域,工艺条件允许的情况下,冷却水管位置应尽可能靠近绕组,以带走更多的热量;增大发电机总进风量,可显著降低定子铁芯温度和绕组温度;进风温度与绕组、定子铁芯最大温度呈线性关系,进风温度每降低5℃,最高温度降低约1.4℃,降低进风温度可一定程度上改善发电机散热性能;进水温度与绕组、定子铁芯最大温度呈线性关系,进水温度每降低5℃,最高温度降低约3.3℃,降低进水温度可大幅提升发电机散热性能。此外,冷却水管与定子铁芯的间隙极大地阻碍了冷却水管的换热,在间隙中填充导热性能好的材料可有效改善发电机散热性能。 结论  文章所得结论可有效指导永磁风力发电机的散热设计,能够保障风电机组在正常工作中安全运行。     

基于机器学习算法与SCADA系统的风电机组变桨系统变频器的故障预警方法研究

变桨系统是风电机组的关键设备,但由于风电机组长期处于复杂的工作环境,导致变桨系统故障成为风电机组故障中最常见的故障之一,而变桨系统变频器故障在变桨系统故障中的占比很高.基于此,提出了一种变桨系统变频器的故障预警方案,分析SCADA系统数据,将机器学习算法应用于故障预警,并将模型温度残差作为故障预警的指标;然后,针对随机...     

基于改进劣化度模型的风电机组日常运行状态评估

为准确地掌握整台风电机组的运行状态,提出一种基于改进劣化度模型的评估方法。首先求得风电机组层次结构中各状态参数的劣化度;其次通过组合赋权法确定各参数的权重,利用岭型分布隶属度函数确定各参数的隶属度矩阵;最后利用模糊综合评判法和彩色图谱对机组的运行状态进行评估和展示。案例机组的日常运行状态评估表明：所提方法可较SCADA系统提前发现机组发电机非驱动端轴承的异常,且误报次数较少。     

Wind shear estimation model using load measurement of wind turbine tower and surrogate model

A wind shear estimation method based on fore‐aft moment is proposed to estimate wind shear strength without a Doppler lidar. We construct wind shear estimation models (WSEMs) using surrogate models whose input is the time‐averaged fore‐aft moment and various supervisory control and data acquisition (SCADA) system data. Learning data for the WSEMs are generated by numerical simulation or field measurement of a real turbine using SCADA, strain gauges, and Doppler lidar. By using simulation data, we construct 20 WSEMs with various input combinations and surrogate methods to select a model with the highest accuracy. The best WSEM is constructed with the universal Kriging surrogate model and uses the fore‐aft moment and wind speed as its input. Subsequently, the best WSEM is applied to a real turbine to validate its accuracy in real wind conditions, and we confirm that the WSEM has reasonable accuracy. However, the estimation error in the real wind condition is about twice as high as that in the simulation due to the real wind shear not completely corresponding to the assumed wind profile and a large yaw error. Further improvement in wind shear estimation accuracy will be achieved by adding yaw error and turbulence intensity to the input variables and applying the WSEM to wind farms on simple terrain or offshore wind farms where wind profile error decreases.     

风力机叶片覆冰状态监测基准值与分级诊断标准研究

以2 MW风力机为研究对象,基于实际风力机状态(SCADA)系统大数据,选取叶片正常状态和覆冰状态下的风速、功率、桨距角和偏航角数据,采用核密度-均值数据处理方法,得到叶片覆冰状态监测基准值及其定量表达式。同时,根据叶片不同覆冰时期桨距角和功率值随风速的变化情况,提出叶片覆冰状态分级诊断标准。应用结果表明,根据桨距角随风速的变化情况可判断在叶片覆冰过程中机组最大功率追踪情况以及气动性能损失情况,根据风速-功率值分布情况可较准确地判别叶片的覆冰状态。     

Modelling and control of variable‐speed multi‐pole permanent magnet synchronous generator wind turbine

Emphasis of this article is on variable‐speed pitch‐controlled wind turbines with multi‐pole permanent magnet synchronous generator (PMSG) and on their extremely soft drive‐train shafts. A model and a control strategy for a full back‐to‐back converter wind turbine with multi‐pole PMSG are described. The model comprises submodels of the aerodynamic rotor, the drive‐train by a two‐mass model, the permanent magnet generator and the full‐scale converter system. The control strategy, which embraces both the wind turbine control itself and the control of the full‐scale converter, has tasks to control independently the active and reactive powers, to assist the power system and to ensure a stable normal operation of the wind turbine itself. A multi‐pole PMSG connected to the grid through a full‐scale converter has no inherent damping, and therefore, such configuration can become practically unstable, if no damping by means of external measures is applied. In this work, the frequency converter is designed to damp actively the drive‐train oscillations, thus ensuring stable operation. The dynamic performance of the presented model and control strategy is assessed and emphasized in normal operation conditions by means of simulations in the power system simulation tool DIgSILENT. Copyright © 2008 John Wiley & Sons, Ltd.     

Modelling and control of synchronous generators for wide‐range variable‐speed wind turbines

The modelling and control of a wide‐range variable speed wind turbine based on a synchronous generator are presented. Two different methods to control the operation of the synchronous generator are investigated, i.e. load angle control and instantaneous vector control. The dynamic performance characteristics of these control strategies are evaluated and compared using three model representations of the generator: a non‐reduced order model including both stator and rotor transients, a reduced order model with stator transients neglected, and a steady‐state model that neglects generator electrical dynamics. Assessment on the performance of grid‐side controller is shown during network fault and frequency variation. A simplified wind turbine model representation is also developed and proposed for large‐scale power system studies. Simulation results in Matlab/Simulink are presented and discussed. Copyright © 2007 John Wiley & Sons, Ltd.     

Analysis and estimation of transient stability for a grid-connected wind turbine with induction generator

Increasing levels of wind energy in modern electrical power system is initiating a need for accurate analysis and estimation of transient stability of wind turbine generation systems. This paper investigates the transient behaviors and possible direct methods for transient stability evaluation of a grid-connected wind turbine with squirrel cage induction generator (SCIG). Firstly, by using an equivalent lump mass method, a three-mass wind turbine equivalent model is proposed considering both the blades and the shaft flexibility of the wind turbine drive train system. Combined with the detailed electromagnetic transient models of a SCIG, the transient behaviors of the wind turbine generation system during a three-phase fault are simulated and compared with the traditional models. Secondly, in order to quickly estimate the transient stability limit of the wind turbine generation system, a direct method based on normal form theory is proposed. The transient models of the wind turbine generation system including the flexible drive train model are derived based on the direct transient stability estimation method. A method of critical clearing time (CCT) calculation is developed for the transient stability estimation of the wind turbine generation system. Finally, the CCT at various initial mechanical torques for different dynamical models are calculated and compared with the trial and error method by simulation, when the SCIG stator terminal is subjected to a three-phase short-circuit fault. The results have shown the proposed method and models are correct and valid.     

基于运行数据的风电机组转矩控制性能评估

风电机组的性能评估方法具有多样性及复杂性的特点,基于风电场SCADA系统中采集的大量风电机组运行数据,对风电机组转矩控制的性能评估方法进行了研究。在深入分析风电机组中发电机转速与发电机转矩关系的基础上,提出了风电机组在最佳风能利用系数C₍p(max))跟踪区内的转矩优化控制的性能评估方法。通过筛选有效数据,拟合计算出风电机组的实际运行转矩增益系数;再通过与理论最优转矩增益系数进行对比,找出风能捕获能力较弱的风电机组,进而采取措施提高其发电量。通过软件仿真及案例分析表明,该方法在不增加设备及成本的情况下,可有效识别因转矩控制的性能差而影响发电量的风电机组,以便及时进行控制策略调校,维护风电场的利益。