风力发电机组振动状态监测与故障诊断系统研究

发布时间：2018-01-26 13:01

本文关键词： 风力发电机组传动系统状态监测故障诊断专家系统　出处：《华北电力大学》2014年硕士论文　论文类型：学位论文

【摘要】：风力发电机组传动系统自身结构复杂，在故障检测和维修中都会占用很长时间，对电力生产造成的影响也很大，因此对风力发电机组传动系统的故障进行预警和诊断具有重要的实际价值。论文依据信号分析和人工智能技术，，对风力发电机组传动系统振动监测与故障诊断系统展开研究。论文首先以某风场某型号1.5兆瓦风力发电机组齿轮箱和轴承为研究对象，计算了齿轮箱各级齿轮和轴承的故障特征频率，分析了齿轮箱高速轴测点的振动信号，提取了3种故障信号的特征，特征分析结果与开箱检查的故障相符合，为专家系统的故障特征知识库的建立奠定了基础。然后论文构建了专家系统的知识库和推理机，结合传动系统的信号处理，分析了正常和故障零件的峭度指标值、频谱能量分布值、信息熵值的差异。确定了峭度指标作为监测特征参数，以频谱能量分布值与信息熵作为故障特征参数。以训练后BP神经网络作为专家系统故障诊断的推理机，将提取到的齿轮箱振动信号的故障特征参数作为输入样本进行仿真测试，测试结果显示推理机能够实现故障诊断的目标。最后，基于LabVIEW软件的开发环境，设计了风力发电机组传动系统振动监测与故障诊断系统，系统能实现的功能主要包括信号的时域波形显示、频域分析、降噪处理、故障预警、故障诊断、数据管理等。
[Abstract]:The structure of the wind turbine transmission system is complex, which will take a long time in the fault detection and maintenance, and also has a great impact on the power production. Therefore, the early warning and diagnosis of wind turbine transmission system has important practical value. The paper is based on signal analysis and artificial intelligence technology. The vibration monitoring and fault diagnosis system of wind turbine transmission system is studied. Firstly, the gearbox and bearing of a 1.5-megawatt wind turbine in a certain wind field are studied, and the fault characteristic frequency of gear box and bearing is calculated. The vibration signals of gearbox high speed shaft measuring points are analyzed, and the characteristics of three kinds of fault signals are extracted. The results of characteristic analysis are consistent with those of open box inspection. It lays a foundation for the establishment of fault feature knowledge base of expert system. Then the knowledge base and inference machine of expert system are constructed, and the kurtosis index value and spectrum energy distribution value of normal and fault parts are analyzed in combination with the signal processing of transmission system. The kurtosis index is determined as the monitoring characteristic parameter, the spectrum energy distribution value and the information entropy as the fault characteristic parameter, and the trained BP neural network as the inference machine of expert system fault diagnosis. The fault characteristic parameters of the vibration signal of the gearbox are used as input samples for simulation test. The test results show that the inference machine can achieve the goal of fault diagnosis. Finally, based on the development environment of LabVIEW software, the vibration monitoring and fault diagnosis system of wind turbine transmission system is designed. The main functions of the system include signal display in time domain. Frequency domain analysis, noise reduction, fault warning, fault diagnosis, data management, etc.
【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】：TP277;TM315

【参考文献】