风电机组状态监测与故障诊断系统的设计与实现

发布时间：2018-04-13 11:58

本文选题：状态监测 + 无线网络　；参考：《华南理工大学》2014年硕士论文

【摘要】：风能作为一种清洁、绿色的能源在注重低碳和环保的当今社会得到越来越多的发展和重视，风力发电的经济效益是风力发电企业主要考虑的指标和因素，在风电场的运行和维护过程中，风机日常维护和故障维修费用占发电企业支出的比例越来越高。风电机组是由多种机械和电气设备构成的复杂整体，一旦某个部位发生故障将直接影响整机的运行，目前国内风电场主要采用定期维护和事后检修的方法，由于维护部件的不确定性往往造成人力和物力的较大浪费，而状态监测和故障诊断系统在新型风机上有所应用，但对于生产时间较早且已进入故障高发期的风电机组来说无法通用。本文设计了一套集数据采集、无线传输和诊断预测为一体的系统，对风力机易发故障部位进行状态监测和故障诊断，提供实时可视的工作状态数据，对已经发生的故障能迅速定位故障部位，并能预测未来一段时间内部件工作状态的趋势。系统通过压电式加速度传感器和霍尔电流传感器，采集发电机和齿轮箱在工作状态时的振动信号和电流信号，通过电荷放大器与采集卡进入工控机，在工控机上进行数据库的读取和存储，利用多天线技术、提高发射和接收增益等方法建立覆盖范围较大的Wi-Fi网络，，实现风力机与主控端之间的无线通信，采集到的数据传输至主控终端后进行分析和诊断，利用小波包分析和傅里叶变换相结合对信号处理，提取信号中隐藏的故障特征向量，最后通过BP神经网络对获取的特征向量进行分析，获取部件的故障部位和故障程度，同时利用神经网络对部件未来的工作状态进行预测，为风电场运行维护人员提供指导性的维修意见，提高维修维护的针对性和实效性。本套系统在汕尾市红海湾风电场进行现场安装和试验，结果表明该系统能有效的对风电机组当前的工作状态进行监测，能准确诊断出发电机和齿轮箱的常见故障。
[Abstract]:As a kind of clean and green energy, wind energy is developing and paying more and more attention to the low-carbon and environment-friendly society. The economic benefit of wind power generation is the main index and factor to be considered by wind power enterprises.In the process of operation and maintenance of wind farm, the expenses of fan daily maintenance and fault maintenance account for more and more expenses of power generation enterprises.Wind turbine is a complex whole composed of a variety of mechanical and electrical equipment. Once a fault occurs in a certain part, it will directly affect the operation of the whole machine. At present, the domestic wind farm mainly adopts the methods of regular maintenance and after-repair.Because the uncertainty of maintenance components often leads to a great waste of manpower and material resources, the condition monitoring and fault diagnosis system is applied to the new fan.But for the wind turbine unit with earlier production time and has entered the period of high failure, it is impossible to generalize.In this paper, a system which integrates data acquisition, wireless transmission and diagnosis and prediction is designed to monitor and diagnose the status of wind turbine fault prone parts, and to provide real-time and visual working state data.The fault location can be quickly located and the working state of the components can be predicted for some time to come.The system collects vibration and current signals of generator and gearbox by piezoelectric accelerometer and Hall current sensor, and enters industrial control computer through charge amplifier and data acquisition card.In order to realize wireless communication between wind turbine and main control terminal, the database is read and stored on industrial control computer, and multi-antenna technology is used to improve the transmitting and receiving gain to set up a Wi-Fi network with a large coverage range, so as to realize the wireless communication between the wind turbine and the main control terminal.After the collected data is transmitted to the main control terminal for analysis and diagnosis, wavelet packet analysis and Fourier transform are used to process the signal and extract the hidden fault feature vector of the signal.Finally, the feature vectors obtained are analyzed by BP neural network, and the fault location and fault degree of components are obtained. At the same time, the neural network is used to predict the working state of components in the future.To provide guidance for wind farm maintenance personnel, improve the pertinence and effectiveness of maintenance.The system has been installed and tested in Hongwan wind farm in Shanwei City. The results show that the system can effectively monitor the current working state of wind turbine and accurately diagnose the common faults of generator and gearbox.
【学位授予单位】：华南理工大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TM614

【参考文献】