ESR 1

The aim of the project is to develop an accurate methodology for wind turbine power curve modelling and outlier detection, based on high frequency SCADA data. In parallel, SCADA alarms will be classified and prioritised. As an outcome, methodologies for wind turbine prognosis will be developed to improve wind turbine performance and O&M strategies, mainly by moving from corrective/preventive maintenance towards predictive maintenance.

The wind turbine power curve shows the relationship between the wind speed and the power output. It is therefore the official performance indicator of the wind turbine that has to be guaranteed by the turbine manufacturer. In case of underperformance or failure, the power output deviates from the normal power curve.

A theoretical power curve is always provided by the manufacturer. However, this curve is neither site-specific nor does take into account the wear and tear of the wind turbine. This can be overcome by obtaining an operational power curve, derived from measured data from an operating wind turbine. This need for modelling site-specific wind turbine power curve has gained great significance over the past years. The main objective for modelling the power curve is twofold: monitoring the global condition of the wind turbine and predictive control and optimisation of the performance, through fault diagnosis. Indeed, the presence of outliers and abnormal values in the power curve might be due to several reasons: environmental issues, faulty anemometers, shut-down due to maintenance or power curtailment, control system issues, blade damage, etc.

From this standpoint, the objective of this work aims to produce the following key outcomes:

• An accurate procedure to obtain a reference power curve, to be used for performance monitoring;

• An effective method for outlier detection;

• A methodology for wind turbine prognosis.

Industrial

Enel Green Power (EGP)
Tentative schedule: Spring – Summer 2016

Academic

University of Strathclyde (USTRATH)
Tentative schedule: Spring – Summer 2017

• Gonzalez, E, Tautz-Weinert, J, Melero, JJ and Watson, SJ (2018) Statistical Evaluation of SCADA data for Wind Turbine Condition Monitoring and Farm Assessment, Journal of Physics: Conference Series, 1037, 032038, DOI: 10.1088/1742-6596/1037/3/032038, Full text: http://iopscience.iop.org/article/10.1088/1742-6596/1037/3/032038/

• Gonzalez, E, Reder, M and Melero, JJ (2016) SCADA alarms processing for wind turbine component failure detection, Journal of Physics: Conference Series, 753(072019), DOI: 10.1088/1742-6596/753/7/072019. Full text: http://iopscience.iop.org/article/10.1088/1742-6596/753/7/072019

• Reder, M, Gonzalez, E and Melero, JJ (2016) Wind Turbine Failures – Tackling current Problems in Failure Data Analysis, Journal of Physics: Conference Series, 753(072027), DOI: 10.1088/1742-6596/753/7/072027. Full text: http://iopscience.iop.org/article/10.1088/1742-6596/753/7/072027

• Gonzalez, E, Reder, M and Melero, JJ (2016) SCADA alarms processing for wind turbine component failure detection, The Science of Making Torque from Wind (TORQUE 2016), Munich. Full text: https://www.researchgate.net/publication/309040863_SCADA_alarms_processing_for_wind_turbine_component_failure_detection

• Reder, M, Gonzalez, E and Melero, JJ (2016) Wind Turbine Failures – Tackling current Problems in Failure Data Analysis, The Science of Making Torque from Wind (TORQUE 2016), Munich. Full text: https://www.researchgate.net/publication/308901537_Wind_Turbine_Failures_-Tackling_current_Problems_in_Failure_Data_Analysis