Control chart monitoring of wind turbine generators using the statistical inertia of a wind farm average

• Published in: Renewable energy 2018-02, Vol.116, p.88-98
• Main Authors: Cambron, P., Masson, C., Tahan, A., Pelletier, F.
• Format: Article
• Language: English
• Subjects: Control chart; SCADA monitoring; Statistical inertia; Wind energy; Wind farm monitoring; Wind turbine monitoring
• ISSN: 0960-1481; 1879-0682
• DOI: 10.1016/j.renene.2016.09.029

A method for monitoring wind turbine generators (WTG) using data provided by the SCADA system is proposed. This method relies mainly upon comparing one WTG with the average of all remaining WTGs on a wind farm. Because environmental conditions on a wind farm are roughly the same over its entirety, the difference between each WTG and the average of the remaining WTGs on the wind farm is constant over time. The statistical inertia of averaged conditions for the entire farm provides a good yardstick for WTG monitoring. The results of monitoring four aspects of a WTG are presented here: these are electrical energy produced; tower vibration; nacelle yaw; and gearbox temperature. Control charts are used to detect abnormal behaviour. With regard to the electrical energy produced, one accidental activation of a curtailment algorithm was found. For tower vibration, we describe an application for the detection of rotor imbalance. For yaw, an example showing detection of nacelle drift is covered. Lastly, for gearbox temperature, the proposed methodology succeeded in detecting an issue two months prior to failure. We have included limitations as to the minimum wind farm size required in order to use the wind farm average. A centralized control chart is also proposed. •This paper presents a method for the monitoring of wind turbine generators in a wind farm.•This method uses the difference between a single wind turbine generator and the average of all the remaining ones.•Control charts are used for the detection of abnormal behaviour the wind turbine generator or of a component of the wind turbine generator.•Results of the application of the proposed methodology on an industrial wind farm on four aspect or components of a wind turbine generator is presented.•A discussion on the required size of the wind farm to use the proposed methodology is included and a centralized and normalized control chart able to monitor a number of aspect and components of the wind turbine generator is presented.