A Study Of Fault Diagnosis Technology For Wind Turbine

Renewable energy sources is attracted increasing attention due to the scarcity of non-renewable energy sources. More and more countries focus on wind power technology, which is most-promising and has increasing market share. Because the production efficiency has been constrainted by the fault diagnosis of wind turbines, many countries in the world are conducting research on the technology of fault diagnosis of wind turbines, expecting to detect faults in time, reduce downtime and improve productivity.This thesis focuses on fault diagnosis technology of wind turbines, which involves some key technologies such as data transmission, data processing and fault diagnosis. There are vary kinds of faults in the wind turbines. In the paper, based on the study of vibration characteristics of the faults, an Internet-based remote fault diagnosis platform has been built in Windows platform. Through this platform, remote communications, data collection and fault diagnosis with the industrial PC can be achieved. Unlike traditional methods, this thesis proposes a method combined wavelet packet transformation and correlation dimension to troubleshoot, to analyze the vibration signals of faults from time domain, frequency domain and spatial domain, which is the major feature of this research.According to the framework and fuctional requirements of the system, this thesis develops a wind turbine vibration monitoring box and wind turbine fault diagnosis software. Hardware and software has been implemented and the fault diagnosis method is verified, combined with the test data of wind turbine. The result shows that, the diagnosis method promoted in the paper is accurate, reliable, high efficiency.