The full-text may be used and/or reproduced, and given to third parties in any format or medium, without prior permission or charge, for personal research or study, educational, or not-for-pro t purposes provided that:• a full bibliographic reference is made to the original source • a link is made to the metadata record in DRO • the full-text is not changed in any way The full-text must not be sold in any format or medium without the formal permission of the copyright holders.Please consult the full DRO policy for further details.
AbstractCurrent wind turbine (WT) studies focus on improving their reliability and reducing the cost of energy, particularly when they are operated offshore. WT Supervisory Control and Data Acquisition (SCADA) systems contain alarm signals providing significant important information. Pattern recognition embodies a set of promising techniques for intelligently processing WT SCADA alarms. This paper presents the feasibility study of SCADA alarm processing and diagnosis method using an artificial neural network (ANN). The back-propagation network (BPN) algorithm was used to supervise a three layers network to identify a WT pitch system fault, known to be of high importance, from pitch system alarm. The trained ANN was then applied on another 4 WTs to find similar pitch system faults. Based on this study, we have found the general mapping capability of the ANN help to identify those most likely WT faults from SCADA alarm signals, but a wide range of representative alarm patterns are necessary for supervisory training.