JEC COMPOSITES MAGAZINE - Issue #96 - April 2015 - 53

Fig. 1: Sensors and damage locations in the second prototype (all units in mm)

intense close to a damage, but smooth out very quickly. So trying
to get information about damage occurrence from strain measurements is a difficult task, as the detected strain changes may be
very small, and may be masked by external factors. This drives the
need to include a large sensors array into the structure, which is
not difficult when using optical fibre sensors, but then the data
processing has to be done in a fully automated approach, and
algorithms are needed to compare and extract information from
the multiple strain measurements.

each one having six FBGs. In total, 24 sensors were used in each
prototype. The sensors were located with a longitudinal separation of 2 m between them. Two optical fibres were located in the
blade intrados and two in the extrados. The fibres were located at
20% and 80% of the chord from the leading edge, along the span
of the blade. Additionally, for the second prototype, four plain
fibre optics were embedded into the blade next to the FBGs with
the purpose of performing distributed sensing by means of an
OBR (see Figure 1).

Two different techniques for data condensation and features extraction are proposed: principal component analysis (PCA) and
hierarchical non-linear PCA (h-NLPCA).

Prototype manufacturing

In order to test the developed methodology in a real composite
structure, two 13.5-metre-long wind turbine blade prototypes
were used. The prototypes were designed by CENER as part of
the Spanish National project "NeWind". The UPM (Universidad
Politécncia de Madrid) and INDRA worked together developing,
manufacturing and testing a system to measure strains in wind
turbine blades in real time. The blades were made of glass/vinylester doped with carbon nanofibres.
The whole acquisition system was designed and built: sensors
selection and fabrication, design of the sensor network, design
of the embedding process used during the blade manufacturing,
design of wiring, design of a miniaturized interrogation system,
design of the interrogation sequence, implementation of damage
detection algorithms, etc.
The first prototype served as a technology demonstrator to probe
the manufacturing concept and improve it. The sensitivity of the
damage detection technique was also tested. The second prototype was manufactured to validate the blade design by means of a
certification test.
Four optical fibres were used for the damage detection system,

An experimental technique was used to manufacture the blades.
The technique is similar to RTM (resin transfer moulding) but
the pressure is much lower and the process is vacuum assisted.
The main objective of the technique, called RTM light, was to
avoid the issues caused by the use of adhesives and to reduce the
weight of the blades.
Due to the use of RTM light, it was necessary to develop a methodology that allowed instrumenting the blade by directly embedding the sensors during the manufacturing process, avoiding all
the issues associated with the use of a closed mould.
The first prototype served, as mentioned before, to probe the
manufacturing concept and to improve it. The final purpose
of this prototype was to enable the study of test specimens for
impregnation verification and micrographs, among other destructive tests. For this reason, UPM and INDRA decided to perform
a test to probe the sensitivity of changes in the strain field as a
consequence of typical damage occurrence, before the prototype
was destroyed.
A simple three-point bending test was performed. Although this
load case does not correspond to any real operational condition
of a wind turbine blade, it served to load the blade and test how
sensitive the technique is to detect defects in a real-scale wind turbine blade. Additionally, this test also made it possible to probe
No96 April 2015 /

jec composites magazine 53

JEC COMPOSITES MAGAZINE - Issue #96 - April 2015

Table of Contents for the Digital Edition of JEC COMPOSITES MAGAZINE - Issue #96 - April 2015