A novel baseline model-based technique for condition monitoring of wind turbine components

Author: Hills A F   Lang Z Q   Soua S   Gan T-H   Perera A   van Lieshout P   Grainger B  

Publisher: The British Institute of Non-Destructive Testing

ISSN: 1354-2575

Source: Insight - Non-Destructive Testing and Condition Monitoring, Vol.53, Iss.8, 2011-08, pp. : 434-438

Disclaimer: Any content in publications that violate the sovereignty, the constitution or regulations of the PRC is not accepted or approved by CNPIEC.

Previous Menu Next

Abstract

The operating conditions of wind turbine components have been monitored by fitting acoustic emission (AE) and vibration sensors inside wind turbines and analysing sensor measurements using appropriate signal processing techniques. However, there is still no systematic approach which can take the operating parameters of wind turbines into account when analysing the sensor measurements for condition monitoring purposes. The operating parameters of wind turbines, such as, for example, wind speed, wind direction and turbine power output etc, often have a significant effect on the measured AE and vibration signals. Therefore, there is a need to develop an effective technique which can systematically consider the effect of these wind turbine operating parameters on the features of measured AE and/or vibration signals. Consequently, any changes in the signal features which are literally produced by an abnormality of the turbine components and/or system can be correctly identified.Motivated by this objective, in the present study we propose a novel baseline model-based technique for condition monitoring of wind turbine components. The idea is to establish a baseline model for an operating wind turbine which represents the relationship between the wind speed, turbine power output, turbine operating parameters and the appropriate features of correspondingly measured vibration and/or AE signals. Given a specific working condition, the model output can be regarded as the baseline features of measured AE and/or vibration signals. Therefore, any significant deviation of practically measured AE and vibration signal features from the baseline can be explained to be due to some kind of abnormality inside the turbine components and/or system. Consequently, the information can be effectively used to achieve the objective of condition monitoring and even fault diagnosis.The application of the proposed method to the real AE, vibration and turbine operating parameter (wind speed and power output) data collected from an operating wind turbine is described in the paper. The results verify the effectiveness of the new approach and demonstrate its potential to address complicated offshore wind turbine condition monitoring problems.

Related content