Publications Repository - Gdańsk University of Technology

It is important from the economic point of view to detect damage early in the wind turbines before failures occur. For this purpose, a monitoring device was built that analyzes both acoustic signals acquired from the built-in non-contact acoustic intensity probe, as well as from the accelerometers, mounted on the internal devices in the nacelle. The signals collected in this way are used for long-term training of the autoencoder neural network. The appropriately trained network automatically detects deviations, signaling them to technical service. The applied methods of analysis of vibroacoustic signals and neural network training are the subject of the presented paper. In addition, the motion magnification of video is used for extracting information on vibrations of the whole wind turbine construction. Finally, spectral analysis is applied for detecting unnatural components presence meaning defects in both: visual and vibroacoustic representations. The process of reduction and construction of a wind turbine model is discussed with a particular emphasis on application to perform extensive tests of the developed methods and algorithms. [The research was subsidized by the Polish National Centre for Research and Development within the project “STEO—System for Technical and Economic Optimization of Distributed Renewable Energy Sources”, No. POIR.01.02.00-00-0357/16.]

Authors

• prof. dr hab. inż. Andrzej Czyżewski link open in new tab