风电齿轮早期故障预警与诊断的研究

发布时间：2019-06-14 15:28

【摘要】：风电作为较为发展成熟的可再生能源已经得到广泛利用,但是由于工作环境恶劣,风电机组故障发生率较高。齿轮箱作为风电机组的主要组成部分,起着传递动力的作用,齿轮箱的运行状况如何直接关系到风电机组运行的好坏。加强齿轮箱的状态监测,实现齿轮箱早期故障的预警和诊断对于提高机组运行效率,降低维修费用都有着重要意义。本文的研究对象时华锐SL1500机组的齿轮箱,从振动参数和SCADA参数两方面着手,对风电机组齿轮箱的齿轮磨损和点蚀故障进行了深入研究和分析,具体包括如下几个方面。首先,对齿轮箱的结构进行了研究和设备划分,根据齿轮箱的详细参数,对所需频率进行了计算。从齿轮的振动机理出发,建立齿轮振动模型,研究其振动机理并简化振动数学模型。对于磨损和点蚀故障进行深入研究,结合齿轮的FTA分析建立了故障原因库、故障影响库、故障征兆库、故障措施库,建立了完整的故障知识库,为下一步的预警和诊断工作的开展奠定了理论基础。其次,针对SCADA参数,采取了选定相关参数,进行历史系数矩阵和实时系数矩阵计算,并计算两者之间的偏差,通过偏差来实现齿轮箱一段时间内的状态预警。针对振动信号,本文采取了计算迅速的简单差值方法,将经过相同时间采样的时域信号转变成了经过相同角度采样的角域信号,实现了信号的整周期重采样。对所得到的角域信号进行无量纲特征提取,通过仿真分别找出与磨损、点蚀故障最为匹配的无量纲特征值,对其采用K邻近度异常检测方法,通过求取相关距离来进行异常点判定,通过计算异常点百分比实现磨损和点蚀的早期故障预警。另一方面,引入HHT变换和阶比分析,实现了频率特征值阶比能量谱的提取,并将其作为早期故障诊断的故障特征值。通过研究阶比能量谱的变化来实现齿轮磨损和点蚀早期故障诊断。预警和诊断相辅相成,在具体的齿轮磨损和点蚀故障的预警和诊断过程中,结合机理分析,提出了基于知识的诊断方法,通过特征匹配来实现特定故障模式的预警诊断。最后,将早期故障预警和诊断的方法运用的工程应用当中去。确定了风电机组传动系统的振动测点布置图以及传感器选型。根据华锐SL1500齿轮箱的详细参数,结合故障机理对点蚀故障进行了故障识别工作,并形成相应故障诊断报告,给维修工作做出指导。
[Abstract]:Wind power has been widely used as a more mature and mature renewable energy source, but due to the bad working environment, the fault rate of the wind power unit is high. As the main part of the wind power unit, the gear box acts as a transmission power, and the operation condition of the gear box is directly related to the operation of the wind turbine unit. It is of great significance to strengthen the state monitoring of the gear box, to realize the early warning and diagnosis of the early fault of the gear box, and to improve the operation efficiency of the unit and to reduce the maintenance cost. In this paper, the gear box, the vibration parameters and the SCADA parameters of the Huarui SL1500 unit are studied, and the gear wear and pitting faults of the gear box of the wind turbine generator set are studied and analyzed in detail, including the following aspects. First, the structure of the gear box is studied and the equipment is divided, and the required frequency is calculated according to the detailed parameters of the gear box. On the basis of the vibration mechanism of the gear, a gear vibration model is set up, its vibration mechanism is studied and the vibration mathematical model is simplified. In this paper, the fault cause library, the fault-affected library, the fault symptom library and the fault-measure library have been set up with the FTA analysis of the gear, and the complete fault knowledge base is established, and the theoretical basis for the development of the next-step early-warning and diagnosis work is established. secondly, according to the SCADA parameters, the selected relevant parameters are adopted, the historical coefficient matrix and the real-time coefficient matrix are calculated, the deviation between the two is calculated, In this paper, a simple difference method is adopted to calculate the vibration signal. The time domain signal sampled by the same time is converted into the angular domain signal sampled by the same angle, and the whole period of the signal is realized. and carrying out non-dimensional feature extraction on the obtained angular domain signal, Point determination that the early result of wear and pitting is achieved by calculating the percentage of abnormal points On the other hand, the HHT transformation and the order ratio analysis are introduced, the extraction of the frequency characteristic order energy spectrum is realized, and it is used as the fault of the early fault diagnosis The characteristics of gear wear and pitting are realized by studying the change of the energy spectrum of the order ratio. The early warning and diagnosis are mutually reinforcing. In the process of early warning and diagnosis of specific gear wear and pitting failure, based on the analysis of the mechanism, a knowledge-based diagnosis method is put forward, and the pre-warning of the specific failure mode is realized by the feature matching. The engineering of the method of early warning and diagnosis of early faults. In the application, the layout of the vibration measuring points of the transmission system of the wind turbine generator set and the layout of the vibration measuring points are Type selection of the sensor. According to the detailed parameters of the gear box of the Huarui SL1500, the fault identification of the pitting fault is carried out in combination with the fault mechanism, and the corresponding fault diagnosis report shall be formed for the maintenance man.
【学位授予单位】：华北电力大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TM315

【参考文献】