{"references": ["Iec 61400-1 wind turbines - part 1 design requirements.", "J. Basurko, O. Salgado, I. Urresti, D. Tcherniak, S. Chauhan,\nA. Rodrguez-Tsouroukdissian, and C.E. Carcangiu. Test / model correlation\nin the alstom 3 megawatt wind turbine. In IMAC XXIX, Jacksonville-\nFL, USA, 2011.", "A. G'omez Gonz'alez, J Rodr'\u2500\u2592guez, X. Sagartzazu, A. Schumacher, and\nI. Isasa. Multiple coherence method in time domain for the analysis of\nthe transmission paths of noise and vibrations with non stationary signals.\nIn Proceedings of ISMA 2010, 2010.", "Mahadevan. S Jiang. X. Wavelet spectrum analysis approach to model\nvalidation of dynamic systems. Mechanical Systems and Signal Processing,\n25:575-590, 2011.", "F. Vanhollebeke B. Marrant W. Meeusen J.Peeters, S. Goris. A need for\nadvanced and validated multibody models as a basis for more accurate\ndynamic load prediction in multimegawatt wind turbine gearboxes. In\nProceedings of ISMA 2010, 2008.", "F. Oyague. Gearbox modeling and load simulation of a baseline 750-kw\nwind turbine using state-of-the-art simulation codes. Technical report,\nNREL/TP-500-41160, February 2009.", "Y.M. Cho S.H. Ahn C.K. Song Z. Hameed, Y.S. Hong. Condition\nmonitoring and fault detection of wind turbines and related algorithms:\nA review. Renewable & Sustainable Energy Reviews, 13:1-39, 2009."]}

Today modern simulations solutions in the wind turbine industry have achieved a high degree of complexity and detail in result. Limitations exist when it is time to validate model results against measurements. Regarding Model validation it is of special interest to identify mode frequencies and to differentiate them from the different excitations. A wind turbine is a complex device and measurements regarding any part of the assembly show a lot of noise. Input excitations are difficult or even impossible to measure due to the stochastic nature of the environment. Traditional techniques for frequency analysis or features extraction are widely used to analyze wind turbine sensor signals, but have several limitations specially attending to non stationary signals (Events). A new technique based on autoregresive analysis techniques is introduced here for a specific application, a comparison and examples related to different events in the wind turbine operations are presented.