Wind Turbine Online Vibration Monitoring Analysis And Visualization Research

Wind turbine has complex structure, coupled with its harsh operating environment, causing high maintenance costs and reducing the economy and the availability of wind farm operation. Wind turbine condition monitoring is essential to improving this situation. Vibration monitoring, which has high test efficiency,accuracy and real-time, based fault monitoring methods has become one of the most efficient technology for machinery and equipment early fault detection and prediction.In this paper, we study and analyze the various vibration data processing and analysis methods, including time-domain characteristics analysis, wavelet analysis, Hilbert-Huang transform, and cepstrum analysis. We propose the spectrum analysis by Sparse Fast Fourier Transform(SFFT) algorithm. First, SFFT algorithm leverages window function to filter vibration signal. Then, after the Fourier coefficients being hashed, the k largest(in magnitude) coefficients of the Fourier Transform of vibration signal can be estimated by location and estimation methods. SFFT algorithm with sublinear runtime in the signal size n has a simple structure and is suitable for online spectrum analysis of vibration signals. Besides, a feature extraction method based on variational mode decomposition is proposed. The independent feature component is obtained by variational mode decomposition. Selecting IMF component including the fault information reconstructs the signal based on kurtosis index. Using the singular value decomposition technique reduce noise of reconstructed signal. And then extracts the fault feature frequency by envelope spectrum analysis.We study the overall design concept and framework of wind turbines online vibration monitoring, using C# programming language and GDI+ technology to design and develop of the wind turbine line vibration monitoring visualization system. Part of the data processing methods mentioned above is embedded in the vibration monitoring system. The system has some main functions about wind turbine vibration signal processing include time-domain monitoring, spectrum monitoring, waterfall monitoring, time-domain features monitoring, historical data analysis and alarm management, which provide the basis for fault detection on wind turbine's major machinery components.