远程风电齿轮箱状态监测诊断系统研究

发布时间：2018-11-20 19:26

【摘要】：为了保证风电机组当中最为脆弱的齿轮箱部分运转安全性，早期发现故障，防止故障发展为重大事故，对齿轮箱的运行状态的监测与故障诊断日益紧迫。并且风力发电机组工作环境一般都十分特殊，都是地处偏远，发生故障以后再进行数据采集时也是非常困难。伴随着网络快速大面积覆盖以及网络通信质量的提高，基于Internet的设备远程状态检测与诊断已经逐步的可以实现，这为大型设备的安全监测与诊断提供了全新的途径。使得以往受限于距离的设备的安全运行问题的到解决，可以随时为对象进行工况查询与故障定位，并且可以充分共享远程资源与现场数据。为此本文对机组齿轮箱的在线的监测与快速诊断做了以下研究工作：（1）本文所采用的状态检测与诊断途径是基于模型的故障诊断思路，为了更接近真实建立，轴系的非线性动力学模型。其后又讨论了各种常见故障在模型中的实现。（2）故障的早期发现与监测的快速性密切相关。为此利用以上搭建的动力学模型，将现代控制理论当中的观测器方法应用到故障诊断当中，将二次滤波与自适应观测器结合起来。利用该观测器对采集的实际数据进行了监测验证。将其监测到的故障工况用重分配小波普来进行定位。为了验证以上结论的正确性，在第一部分所建模型中进行了故障仿真的验证，并在此基础上分析了系统特性。验证了诊断结果。（3）为了解决风电机组的远程在线监测问题。本文主要搭建了一种基于arm11的远程数据采集系统结构。下位机采用linux操作系统，采用多线程的并发的数据采集与传输结构设计，提高了系统运行的速度为实时的故障诊断提供了基础；采用AD7821作为模数转换采集器，，为此开发了AD7821与IDT7206的下位机驱动程序；应用LABVIEW设计了一款及时数据接收与仿真分析监测平台。从而实现了远程设备的无缝试在线监测与故障诊断。
[Abstract]:In order to ensure the safety of the most vulnerable part of the gearbox in wind turbine, to detect the fault early and prevent the fault from developing into a major accident, it is increasingly urgent to monitor and diagnose the running state of the gearbox. And the working environment of wind turbine is very special, it is very difficult to collect data after failure. With the rapid coverage of the network and the improvement of network communication quality, the remote status detection and diagnosis of equipment based on Internet has been gradually realized, which provides a new way for the safety monitoring and diagnosis of large equipment. It can solve the problem of safe operation of equipment limited by distance in the past, can query and locate the work condition for the object at any time, and can fully share the remote resources and the field data. In this paper, the on-line monitoring and rapid diagnosis of gear box are studied as follows: (1) the method of condition detection and diagnosis is model-based fault diagnosis, in order to be closer to the real establishment. The nonlinear dynamic model of shafting. Then the realization of various common faults in the model is discussed. (2) the early detection of faults is closely related to the rapidity of monitoring. Based on the above dynamic model, the observer method in modern control theory is applied to fault diagnosis, and the quadratic filter and adaptive observer are combined. The observer is used to monitor and verify the collected data. The fault condition is located by redistributing wavelet. In order to verify the correctness of the above conclusions, fault simulation is carried out in the first part of the model, and the characteristics of the system are analyzed. The diagnostic results are verified. (3) in order to solve the problem of remote on-line monitoring of wind turbine. In this paper, a remote data acquisition system based on arm11 is built. The lower computer adopts linux operating system and multithread concurrent data acquisition and transmission structure design, which improves the speed of system running and provides the foundation for real-time fault diagnosis. Using AD7821 as A / D conversion collector, the driver of AD7821 and IDT7206 is developed, and a timely data receiving and simulation monitoring platform is designed by LABVIEW. Thus, the remote equipment seamless trial online monitoring and fault diagnosis is realized.
【学位授予单位】：新疆大学
【学位级别】：硕士
【学位授予年份】：2013
【分类号】：TP277;TH132.41

【相似文献】