procedures would result in significant O&M costs and downtimes.
CM techniques provide useful information that support operational
efficiency and contribute to the improvement of new turbine
designs.
Some components fail earlier than intended by their design and
cause unscheduled downtimes which reduce the productivity of
the wind farm. Condition Monitoring Systems (CMS) can contribute
to the improved operational control of the critical components [5]
[6], and [7]. CM techniques, such as vibration and oil analysis,
acoustic emission, temperature measurement, etc., together with
advanced signal processing methods and data trending, provide
continuous information regarding the status of the component
being monitored [8] and [9]. CM techniques are used to collect the
main functional parameters of critical components, such as the
gearbox, generator, main bearings, blades, tower, etc. [10]. This
paper presents a novel approach for determining the critical components
of any WT in different conditions based on a real case
study. The results reported herewith support the optimisation of
CM design and investment. For this purpose a method based on
fault tree analysis (FTA) that allows qualitative analysis is presented.
Quantitative Fault Tree Analysis (FTA) is performed by
employing Binary Decision Diagrams (BDDs). In Section 2 are presented
the FTAs, BDDs, the conversion from FTA to BDD and some
experiments to test and verify the approach. In Section 3, importance
measures for the Fault Tree (FT) have been presented and
tested in order to identify the events that are more important for
the fault of the top event. Finally, in Section 4, a case study of an FT
for aWT has been developed considering large research studies and
analysed qualitatively and quantitatively, where the main results
are presented in Section 5. The main components ofWTs and their
relationship have been set taking into account the comments of
industrial experts involved in the European Projects NIMO [11] and
OPTIMUS [12]. The critical components have been set according to
different scenarios. This study will be a useful reference for those
involved in the optimisation of the design of the CMS and therefore
the investment required.