Loughborough University
Browse

File(s) under permanent embargo

Reason: This item is currently closed access.

Physics of failure approach to wind turbine condition based maintenance

journal contribution
posted on 2012-04-02, 13:11 authored by Christopher S. Gray, Simon Watson
Wind turbine condition monitoring systems provide an early indication of component damage, allowing the operator to plan system repair prior to complete failure. However, the resulting cost savings are limited because of the relatively low number of failures that may be detected and the high cost of installing the required measurement equipment. A new approach is proposed for continuous, online calculation of damage accumulation using standard turbine performance parameters and Physics of Failure methodology. The wind turbine system is assessed in order to identify the root cause of critical failure modes and theoretical damage models are developed to describe the relationship between the turbine operating environment, applied loads and the rate at which damage accumulates. Accurate estimates may then be made in real time concerning the probability of failure for specific failure modes and components. The methodology is illustrated for a specific failure mode using a case study of a large wind farm where a significant number of gearbox failures occurred within a short space of time. Such an approach may be implemented at relatively low cost and offers potential for significant improvements in overall wind turbine maintenance strategy.

History

School

  • Mechanical, Electrical and Manufacturing Engineering

Citation

GRAY, C.S. and WATSON, S.J., 2010. Physics of failure approach to wind turbine condition based maintenance. Wind Energy, 13 (5), pp. 395 - 405

Publisher

© John Wiley and Sons Ltd

Version

  • AM (Accepted Manuscript)

Publication date

2010

Notes

This article is closed access, it was published in the journal Wind Energy [© John Wiley & Sons Ltd]. The definitive version is available at: http://dx.doi.org/10.1002/we.360

ISSN

1095-4244

Language

  • en