Title: Phase I analysis of hidden operating status for wind turbine

Abstract: Data-driven methods based on Supervisory Control and Data Acquisition (SCADA) become a recent trend for wind turbine condition monitoring. However, SCADA data are known to be of low quality due to low sampling frequency and complex turbine working dynamics. In this work, we focus on the phase I analysis of SCADA data to better understand turbines' operating status. As one of the most important characterization, the power curve is used as a benchmark to represent normal performance. A powerful distribution-free control chart is applied after the power generation is adjusted by an accurate power curve model, which explicitly takes into account the known factors that can affect turbines' performance. Informative out-of-control segments have been revealed in real field case studies. This phase I analysis can help improve wind turbine's monitoring, reliability, and maintenance for a smarter wind energy system.