earticle

영어

The national capacity around the world has focused to the development of renewable energy in order to respond to energy, environment, security and economic. Among them, the unit price of wind turbine was reduced by technical maturity. So, the comparative advantage against other renewable energies can be achieved because of proximity to the unit price of the existing fossil fuel. In other words, the wind turbine is the only alternative energy that can ensure that a competitive price is equal to the fossil fuels in the short term. The onshore wind turbine has difficulty to secure additional location by the depletion of good location and the increase in civil complaint. The korea also has factors of problem such as limitation of locational condition and noise. Therefore, it is essential to advance to the sea. However, the mechanical and electrical allowance that components of wind turbine must withstand was increase in a corresponding degree. Therefore, the possibility of failure was increased, and then the secondary damage by limited access caused additional costs by locating offshore. In this paper, in order to diagnose fault in advance and ensure the reliability of large wind turbine located in the sea, we took advantage of the condition monitoring system (CMS). In other words, we propose effective monitoring and control system by integrating the CMS and SCADA systems based on LabVIEW. First, the remote monitoring system based on PC using the ethernet gateway of wireless sensor network (WSN) should be constructed in order to overcome the environment of positional constraints. And then we collect measured signal data from distributed nodes of the installed WSN within wind turbine farms and extract feature information of the classified fault and normal signals pattern through Wavelet Analysis. The extracted feature information is used as the input of neural network learning. When the error signals have arisen in the wind turbine farms, it is possible that alarm is happen, and condition is controlled through the automatic fault diagnosis. In addition, simple faults as well as complex faults that occur over a long time can be diagnosed early.