Comparison of novel SCADA Data Cleaning Technique for Wind Turbine Electric Pitch System

Abstract Wind turbines typically do not operate in the ideal operating conditions, leading to abnormal behaviour that is reflected their power curves. This can affect performance of condition monitoring processes, as it may mask faulty behaviour. By cleaning other data, such curtailment, models learn normal turbines. paper presents a novel technique utilises combination data binning and Mahalanobis distance. removes between 5 6% without great loss data. When compared against techniques, one presented this produces more curve. could improve data-based techniques.