基于定子电流的风力发电机关键部件的故障监测与诊断

发布时间：2018-05-30 09:38

本文选题：风力发电机 + 故障诊断　；参考：《燕山大学》2015年硕士论文

【摘要】：随着风力发电产业的迅猛发展以及对风力发电机系统的稳定性、易维护性等方面要求,风力发电机状态监测与故障诊断技术引起了学术界和工程界的广泛关注。齿轮箱和转子分别作为风力发电机组的关键部件,其运行状态的实时监测和准确分析,对整个风力发电机的故障诊断和运行维护均具有重要的意义。当前,振动信号分析法是风力机组故障监测与诊断的主要方式,但该方法具有设备成本高、传感器安装不便以及受外界干扰较大等弊端。相比而言,定子电流信号分析法具有不易受外界干扰、信号易采集、信噪比高以及可实现在线监测等优势,已被国内外专家学者作为风力机组故障诊断的重要手段。因此,本文采用定子电流信号分析法,对风力发电机的齿轮和转子进行故障诊断研究。主要围绕以下几方面内容展开:(1)详细介绍了风力发电机的系统构成和常见故障,重点对发电机转子断条故障机理和风力发电机的齿轮故障振动特性及定子电流检测原理进行分析,为下文进行风力发电机的转子断条故障检测和齿轮点蚀故障检测提供依据。(2)根据发电机转子断条故障机理,提出将谱减法引入定子电流频谱分析中,并与解析小波变换相结合进行转子断条故障检测,实现在负荷突变情况下转子断条故障的特征提取和故障检测。通过数值仿真信号和模型仿真信号来验证所提方法的有效性。进一步引入故障程度因子来量化转子断条故障程度。(3)根据风力发电机齿轮故障定子电流信号特点,本文将基于经验模态分解(EMD)和独立分量分析(Fast ICA)的故障特征提取方法与样本熵算法相结合用于齿轮点蚀故障检测,有效量化齿轮故障特征向量。通过仿真信号分别验证基于EMD和Fast ICA故障特征提取算法的有效性和样本熵用于量化时间序列复杂度的有效性。通过搭建风力发电机齿轮点蚀故障诊断平台,在不同的转速条件下,验证本文所提方法的有效性。
[Abstract]:With the rapid development of wind power generation industry and the requirements of stability and maintainability of wind turbine system, wind turbine condition monitoring and fault diagnosis technology has attracted extensive attention from academia and engineering circles. Gear box and rotor are the key components of wind turbine. The real-time monitoring and accurate analysis of the running state of the gearbox and rotor are of great significance to the fault diagnosis and operation maintenance of the whole wind turbine. At present, vibration signal analysis is the main method of wind turbine fault monitoring and diagnosis, but this method has the disadvantages of high equipment cost, inconvenient installation of sensors and large external interference. In contrast, the stator current signal analysis method has the advantages of easy external interference, easy signal acquisition, high signal-to-noise ratio and on-line monitoring, and has been used as an important means of wind turbine fault diagnosis by experts and scholars at home and abroad. Therefore, the stator current signal analysis method is used to study the fault diagnosis of the gear and rotor of the wind turbine. Focusing on the following aspects: 1) the system structure and common faults of the wind turbine are introduced in detail. The mechanism of broken bar fault of generator rotor, the vibration characteristics of gear fault and the detection principle of stator current of wind turbine are analyzed. This paper provides a basis for rotor broken bar fault detection and gear pitting fault detection of wind turbine. According to the fault mechanism of generator rotor bar breakage, the spectral subtraction method is introduced into stator current spectrum analysis. Combined with the analytic wavelet transform, the fault detection of rotor bar break is carried out, and the feature extraction and fault detection of rotor broken bar fault are realized in the case of sudden change of load. The validity of the proposed method is verified by numerical simulation signal and model simulation signal. Furthermore, the fault degree factor is introduced to quantify the fault degree of rotor broken bar. (3) according to the characteristics of stator current signal of gear fault of wind turbine, In this paper, the method of fault feature extraction based on empirical mode decomposition (EMD) and independent component analysis (ICA) is combined with the sample entropy algorithm for pitting fault detection of gears. The validity of fault feature extraction algorithm based on EMD and Fast ICA and the validity of sample entropy used to quantize the complexity of time series are verified by simulation signals. By setting up a fault diagnosis platform for pitting corrosion of wind turbine gear, the effectiveness of the proposed method is verified under different rotational speed conditions.
【学位授予单位】：燕山大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：TM315

【参考文献】