Wind Turbine Blade

Ref.
Ares(2020)6198499 - 01/11/2020
Risk Based Technology for Blade Structural Assessment
Work Package Title WP4 – Quality Attainment

Deliverable Title D4.2 - White paper for ISO/CD 16079-1
Deliverable due date 31/08/2020
Actual submission date 30/10/2020
Deliverable type Report
Dissemination level Public
Lead Beneficiary TWI Ltd
Author Linghao Zhou, Jialin Tang
Reviewer Ryan Marks
Project's Coordinator: Mr. Benn Faulkner

Tel: +443306600363
E-mail: [email protected]
Page | 2
Document History
Version Issue Date Stage Changes Contributor(s)

1.0 26/10/2020 Final Final TWI Ltd, RA
Copyright ©, BladeSave Project Consortium

Page | 3
List of figures 4
List of tables 5
Introduction 6
Background 6
Relation to ISO 16079-1 and associated standards 6
Relation to ISO 16079-1 6
Relation to associated standards 7
The need for wind turbine blade monitoring 8
Description of the issue and its context 8
Industry trends albeit global lockdown 10
Impact on the global economy and clean energy market 10
Impact on global wind industry 11
Impact on policy and regulation 11
Conclusion 12
BladeSave system 12
Description of the system 13
How does the system work 14
System components 14
Specifications of hardware 16
User Interface demonstration 20
System qualifications 22
Case study – testing on wind turbine blades 23
Demonstration on EWT wind turbine 23
Validation on detecting blade cracks induced from bending tests 26
Other validation tests 29
Conclusion 29

Page | 4
List of figures
Figure 1 BladeSave System Concept & Approach 13

Figure 2 BladeSave system integration 15
Figure 3 BladeSave User Interface Login Page 19
Figure 4 BladeSave Wind Farm Page 19
Figure 5 BladeSave Wind Turbine Page 20
Figure 6 BladeSave Wind Turbine Logs Section 21
Figure 7 BladeSave system part 1 where (a): processing unit; (b): SmartScan system for vibration
monitoring; (c): SmartSonic system for AE monitoring and (d): FBG accelerometers 22
Figure 8 BladeSave system part 2 where (e): optic switch; (f): military level optic cables and (g) fibre
optic AE sensors 23
Figure 9 BladeSave recorded AE events 23
Figure 10 Wind turbine blade for bending tests 24
Figure 11 Sensors installation for debonding line monitoring 25
Figure 12 Test blade with debonding defect developed 25
Figure 13 BladeSave real-time monitoring of bending test 26

Page | 5
List of tables
Table 1: Relations between 16079-1 and subsequent standards, source from ISO 16079-1:2017 7
Table 2: Large final investment decisions taken for offshore wind January to June 2020 11
Table 3: List of BladeSave system components 14
Table 4: SmartPatch specification 16
Table 5: BladeSave AE sensor specifications 16
Table 6: BladeSave accelerometer specifications 17
Table 7: BladeSave SmartScan specifications 17
Table 8: BladeSave SmartSonic specifications 18

Page | 6
1 Introduction
1.1 Background
The BladeSave system is an innovative Structural Health Monitoring (SHM) system for wind
turbine blades based on Fibre Optic sensing techniques.
There is an increasing demand in the wind industry to reduce operation and maintenance
costs. Unexpected failures and incidents on the blades can cause significant downtime of a
wind turbine, thus reducing power generation capacity. Therefore, wind farm
owners/operators are keen to find a solution to closely monitor the health status of the wind
turbine blades, so that small damages can be fixed ahead of them becoming a more significant
issue thus reducing the downtime and saving costs.
In this context, the BladeSave system, as an SHM solution, can play a significant role. By
continuously monitoring the condition of wind turbine blades, the BladeSave system can
reduce wind turbine downtime by detecting small defects and monitor their growth before
they become a more serious issue. This allows maintenance operations to be effectively and
efficiently plan avoiding time-consuming and costly repairs of large damage. The project takes
forward the fibre optic sensing hardware solutions developed by Smart Fibres (UK), combining
the expertise from TWI Ltd (UK) and Renewable Advice (UK), in SHM and blade inspection and
maintenance respectively. In addition, ASSIST (RO) developed the system communication
interface based on cloud computing. The system has been tested and demonstrated on EWT
(NL) wind turbines.
1.2 Relation to ISO 16079-1 and associated standards
1.2.1 Relation to ISO 16079-1
As specified in the Grant Agreement, this white paper aims at making recommendations to
standard BS ISO 16079-1-2017: Condition monitoring and diagnostics of wind turbines – Part 1:
General Guidelines. The scope of the standard 16079-1-2017 states that it:
“Gives guidelines which provide the basis for choosing condition monitoring methods used for
failure mode detection, diagnostics and prognostics of wind power plant components” .
By closely investigating the standard, it is discovered that the document provides high level
instructions on implementing FMECA (Failure Modes, their Effects and Criticality Analysis)
without considering monitoring technologies. Therefore, this is not directly linked to the
condition monitoring techniques that the BladeSave system offers, including Acoustic Emission
(AE), vibration and strain.

Page | 7
In the meantime, ISO 16079-1 stresses that data reduction has always been “a big issue” for
general condition monitoring systems. BladeSave system addresses this issue by implementing
a processing unit, which receives and processes monitoring data collected by BladeSave
hardware in real-time, and communicates with a cloud database to securely store the analysis
results.
1.2.2 Relation to associated standards
ISO 16079-1 specifies its subsequent standard series, starting from ISO 16079-2 which gives
guidelines on monitoring the drivetrain of wind turbines. Instead of applying FMECA, it
advocates the use of failure mode and symptoms analysis (FMSA) to detect a certain failure
mode for the wind turbine drivetrain. The relationship between 16079-1 and the subsequent
standards are demonstrated in Table 1:
Table 1: Relations between 16079-1 and subsequent standards, source from ISO 16079-1:2017
Standards Methodology Scope

16079-1 FMECA To define the failure modes to be monitored
16079-2, 3 … FMSA Specify how to detect a certain failure mode
It is discovered that currently, no standard in the ISO 16079 series has been established on
monitoring wind turbine blades, which are critical components in a wind turbine. It is
beneficial to wind turbine manufacturers, wind farm owners and operators that an ISO
standard for wind turbine blades is embedded into ISO 16079 series.
Additionally, ISO 16079-1 calls on parent standard ISO 17359: Condition monitoring and
hich specifies guidelines for the general
diagnostics of machines — General guidelines, w
procedures to be considered when setting up a condition monitoring programme for all
machines. This standard gives two guidelines that BladeSave could help to improve:
● Taking measurements of different parameters wherever possible at the same time or

under the same operating conditions;
● Quality of measurements must be established.
For the first point, the operating conditions on a wind turbine are dynamic. Therefore,
BladeSave system captures data in configurable time intervals, and it can incorporate
parametric data such as the output power and wind speed to investigate the correlations

Page | 8
between the data and the conditions under which the data was captured. For the second
point, BladeSave system is built on the innovative fibre optic (FO) technology, so that all
sensors are passive and are immune to electromagnetic interference (EMI) noises. The signals
do not suffer significant loss from long optic cables compared with resistive cables for
traditional piezoelectric transducers.
To recapitulate, currently there are below issues discovered from existing ISO standards on
wind turbine blades condition monitoring:
● ISO 16079-1 is established but does not specify the usage of condition monitoring
techniques including vibration, AE and strain for wind turbines;
● ISO 16079 series is currently under development and no standards are established for
wind turbine blades;
● ISO 17359 which is a parent standard for ISO 16079-1, recommends monitoring
parameters only under non-dynamic conditions. The use of the FO system to improve
the quality of measurement could be considered.
The objectives of the white paper are therefore, to raise the awareness of the needs for wind
turbine blade condition monitoring, and to demonstrate the capability of fibre-optic based the
BladeSave system to fulfil the needs. To achieve the objectives, this white paper firstly
elaborates on the industry needs, with the Covid-19’s impact on the industry trends reviewed.
An introduction of the BladeSave system is then given and each component’s functionalities
are explained. Lists of the system specification, standards and regulations which the system
conforms to are also presented. The system has been demonstrated on a EWT wind turbine,
and tested during a lab-based blade bending test conducted by TWI Ltd.
This white paper fulfils the requirements of Deliverable D4.2 – White paper for ISO/CD
16079-1.
2 The need for wind turbine blade monitoring
2.1 Description of the issue and its context
While a wind turbine is in operation, the blades encounter very complex loading sequences.
The cyclic loading can cause large-deflection bending and can result in damage. In the past, the
heath conditions of a wind turbine blade have been estimated based only on strain

Page | 9
measurements. Damage characterisation was usually performed by visual inspection for

evaluating surface damages. This approach results in a high Operation and Maintenance
(O&M) costs, which constitute a sizable share of the total annual costs of a wind turbine’
maintenance as well as requiring a high-risk inspection process that is weather-dependent.
Therefore, manufactures attempt to lower these costs significantly by developing new turbine
designs or installing modern structural integrity systems that require fewer regular service
visits and less turbine downtime.
The BladeSave project aims to provide a reliable methodology for the assessment of wind
turbine blade structural integrity. Several structural integrity monitoring techniques including
strain monitoring, vibration analysis and AE monitoring have been applied successfully in wind
turbine blade damage detection during the course of this project. These three different
techniques, though with different testing principles, all use surface mounted sensors to detect
physical blade changes which are indicators of damage initiation and growth. Traditionally,
these measurement techniques are implemented using piezoelectric (PZT) sensors and
resistive sensors in the case of strain. These sensors have several advantageous characteristics
such as high sensitivity and good stability. However, PZT and resistive sensors are electrical
sensors and are inevitably susceptible to electromagnetic interference (EMI) and can only
operate in a limited temperature range. This means that these sensors can be affected
significantly (and even destroyed) during the operation of the turbine due to the EMI resulting
from the power generation and also from lightning strike. In addition, the shielded cable for
the PZT sensors is heavy, has a relatively large cross-section and are reasonably inflexible
which creates deployment and installation issues. In contrast, Fibre Optic (FO) sensors have
been demonstrated to be an excellent technology for real-time monitoring due to their
numerous advantages, such as full immunity to EMI, miniature size, lightweight, durability, and
the possibility to be integrated within the material. The above-mentioned advantages have
made FO sensors an excellent application in SHM systems for wind turbine blades.
The BladeSave system provides wind farm owners and operators a comprehensive monitoring
system for the SHM of the blades. The system can effectively reduce the risks related to blade
failures and contribute to the growth and competitiveness of wind energy in the world’s
renewable energy markets by commercializing an innovative solution.

Page | 10
2.2 Industry trends albeit global lockdown
Since the BladeSave proposal was written, the global coronavirus (Covid-19) pandemic has
sent economic shock waves through every aspect of the world economy. As lockdown
measures have eased from the end of the second and into the third quarter of 2020, more
data is becoming available on the impact of the pandemic, albeit with a lag in its collection,
analysis and publication.
This section presents a survey of public domain information on the impact of the pandemic on
the global wind power industry, policies and regulations.
2.2.1 Impact on the global economy and clean energy market
Due to the extraordinary nature of the COVID-19 pandemic, governments around the world
are faced with the unprecedented challenge of seeking to restart economic activity whilst
guarding against further outbreaks of the virus. Not knowing the likely progression of the virus
or when large scale vaccination programmes will start is creating huge uncertainty in industries
and financial markets. Forecasting when the global economic downturn will end and at what
speed industries will recover is extremely difficult.
Many central banks and international economic and monetary bodies are predicting a deeper
and more prolonged recession than was foreseen in March/April 2020. The International
Monetary Fund in its most recent forecast predicts a global economic contraction of 4.9%,
nearly 2% worse than their April 2020 update.
National Governments measures to contain the virus have had a major impact on all types of
energy consumption. Analysis by the International Energy Agency (IEA) shows that during full
lockdown, average national energy was reduced by 25% per week.
As lockdown measures are eased and economic activity resumes, demand for energy is
increasing once more, but with significant market volatility caused by continuing uncertainty
about the future course and impact of Covid-19. In the short term, investment in renewable
energy programmes is expected to fall along with fossil fuels as a direct but temporary effect
of the Covid-19 shock. The IEA predicts that investment in energy programmes (including clean
energy) globally will fall by approximately 20% this year.

Page | 11
Into 2021 and beyond, green energy will continue to grow its share of world energy
production. The maturity of offshore wind technology and proven commercial mechanisms
offers investors low risk opportunities. Furthermore, the twin imperatives of ensuring security
of energy supply as well as decarbonisation endure on a global basis.
2.2.2 Impact on global wind industry
Despite the global economic uncertainty threatening energy investment generally, early data
indicates a big increase in global investment in wind projects during the first half of 2020. This
is especially true for offshore wind projects, where according to Wind Power Monthly, over
$35 billion was committed from January to June, comfortably exceeding, in half the time, the
record level of $31.9 billion for the whole of 2019. Some very large projects have been
announced, as shown in Table 2.
Table 2: Large final investment decisions taken for offshore wind January to June 2020
Construction
Developer/project name Location Cost Power output
period
Vattenfall Hollandse Zuid Netherlands $3.9 billion 1.5GW 2021-23
SSE Seagreen UK $3.8 billion 1.1GW 2021-23
Changfang Xidao Taiwan $3.6 billion 600MW 2021-23
Guangdong Yudean
Yangjiang Yangxi China $1.8 billion 600MW 2021-23
Shapaat
Fecamp/St Brieuc France $5.4 billion 993MW 2021-23
In Europe, Windeurope.org reports that €14.3 billion has been raised for financing new wind
farms, combining €11 billion for offshore and €3.3 billion for onshore installation.
2.2.3 Impact on policy and regulation
The commitments made by governments to transition to clean energy and deliver net zero
carbon emissions by 2050 (in the Paris agreement) require concerted political will and its
practical manifestation in policy and investment. The dramatic drop in emissions and air
pollution seen in the Covid-19 crisis has triggered debate about whether this could be the
decisive moment that pivots economic policy in favour of sustainable development goals.

Page | 12
International Renewable Energy Agency (IRENA) estimates that increasing investment in clean
energy to $4.5 trillion per year would increase global economic growth by 1.3% and create
approximately 19 million new jobs by 2030. Governments around the world are bringing
forward post-crisis economic stimulus packages of a size unprecedented in peacetime, some of
which feature sustainability and clean energy measures.
More positively, the EC’s €750 billion stimulus package, named ‘Next Generation EU’ and
passed by the European Parliament in July, 2020, targets recovery funding at green measures
and is intended to accelerate the European Green Deal, a cornerstone of the EU’s growth
strategy (European Parliament News, 29 July 2020).
2.2.4 Conclusion
The global wind sector was one of the fastest growing sectors with huge demands prior to
global lockdown, now it may be the least badly affected industry during the pandemic. As
more data has been published through the year, initial caution and pessimistic forecasts have
given way to growth and optimism, as demonstrated by the massive increase in finance for
new projects seen in the first half of the year, especially for offshore wind.
With this information, it is concluded that with the growing wind sector, there is still a
booming market for a competent condition monitoring system for wind turbine blades albeit
global lockdown. And therefore, it is urged that condition monitoring standards related to
wind turbine components including blades, are established into ISO 16079 series.
3 BladeSave system
The BladeSave system has been developed to continuously monitor the health status of the
wind turbine blades. Based on innovative fibre optic sensing technology, the BladeSave
system is immune to EMI noise, light weight, scalable and easy to install inside the blades. The
system offers a fusion of three monitoring techniques, targeting different functionalities and
fault mechanisms:
● Strain: rotor imbalance, fatigue cycles, independent blade pitch control (IPC) and gross
blade damage;
● Vibration: ice accretion on blade surface and gross blade damage;

Page | 13
● AE: Incipient blade damage and cracks, such as fibre breakage, matrix crack and
delamination.
The BladeSave system also provides user friendly interface software, WindManagerTM, which
offers a web-based “traffic light” indication system for end users to be aware of the current
status of their turbine blades. More details about the system are presented in the following
sections.
3.1 Description of the system
As previously introduced, the BladeSave system consists of both hardware devices and
software applications. Hardware components include interrogators and FO sensors, which are
the data acquisition devices. Software applications include data acquisition control, data
processing, cloud data server and a web application which provides end users real-time blade
health status information.
The BladeSave system links the results to a Blade Management Software incorporating data
analysis algorithms to offer a comprehensive solution for wind turbine blade monitoring,
repairing and management, as demonstrated in Figure 1.

Page | 14
Figure 1 BladeSave System Concept & Approach
3.2 How does the system work
Each wind turbine will be installed with the BladeSave system which includes three different
types of sensors: AE sensors, vibration accelerometers and strain sensors. The system will be
installed in the wind turbine hub, so it rotates with the blades. All three types of sensors will be
mounted on the inside surface of the blades. The operators/wind farm owners can choose to
install the AE sensors on the areas where it’s most likely to have damages. The Vibration
accelerometers and Strain sensors are installed on the root area of the blade. It should be
made clear that this system is applicable to both new and currently in-service turbines.
Raw data acquired by the hardware components are transferred to the data pre-processing
software, where several key signal features are extracted from the raw data. The extracted
features are then transferred to the application programming interface (API) server where all
of the historical processed data is stored. The API server also pushes the signal features to web
applications for visualizing/managing the data. On the web application, historical data can be
viewed and alarms are raised when there are deviations from the normal operating
behaviours.
3.2.1 System components
A list of BladeSave system equipment is shown in Table 3.
Table 3: List of BladeSave system components
Category Component Functions

Taking measurement including strain, AE,
Fibre optic sensors
vibration
SmartScan Interrogator for strain and vibration
Hardware SmartSonic Interrogator for AE
Optical Switch Multiplex of sensor channels
Onsite processing and communication with cloud
Processing unit
server
SmartSoft Control software for SmartScan data acquisition
AEx Control software for SmartSonic data acquisition
SSAE Processor for data acquired by SmartSonic
SmfBlade Data manager for SmartSonic and SmartScan
Software Data transmission from SmfBlade to the cloud,
Native application
produces alerts when no data is produced
Storing and pushing data to web application and
API server
provides programmable access to the data
Visual interface using the API server for
Web application
visualising/managing the acquired data

Page | 15
These components are integrated to achieve data collection, onsite signal processing, wireless
communication with cloud server and data visualisation on web-based user interface software.
The processing unit hosts controlling software for SmartScan and SmartSonic interrogators,
which are connected to strain/vibration FBG sensors and AE FBG sensors respectively. The
sensors raw data waveforms which are sent back to the processing unit using USB3.0 for onsite
analysis. The processed results are transmitted by a native application residing in the
processing unit, which communicates with the cloud server through a wide area network
(WAN). BladeSave’s API server stores and pushes the transmitted results to WindManagerTM
software for post-analysis, and sends the final monitoring results to the front-end web
application to end users. This process is also illustrated in Figure 2.

Page | 16
Figure 2 BladeSave system integration
3.2.2 Specifications of hardware
Three types of sensors are used in the BladeSave system to monitor wind turbine blades,
namely strain, vibration and acoustic emission sensors. All sensors are manufactured with

Page | 17
Fibre Bragg Gratings (FBG) technique, which makes the sensors pure optical and are immune
to EMI noise.
Firstly, strain sensors are used to measure strain up to a very high level (5000 microstrain),
which makes them well suited to highly stressed composite constructions. The strain sensors
are installed in the root area of wind turbine blades. The measurements from the sensors are
converted to edgewise and flapwise bending moments. The bending moments measured are
able to provide information to assess the loading at the root of the blade at any instance in
time. To mitigate the interference from temperature changing, SmartPatch also records
temperature alongside strain. The specifications for the FBG strain sensors are presented in
Table 4:
Table 4: SmartPatch specification
Parameter Value
Patch Dimensions 100 x 15 mm

Gauge Length (approx.) 6 mm
Strain Range ± 5,000 µstrain
Strain Sensitivity 1.2 pm/µstrain
Strain Resolution 0.4 µstrain
Temperature Range -40 to +60 °C
Temperature Sensitivity 11 pm/°C
Temperature Resolution 0.05 °C
Fibre Type Single Mode SMF-28, 9/125 µm
AE sensors respond to elastic waves that are released during movement within the internal
structure of the material (known as an AE event). Examples of AE events are crack initiation
and propagation in the wind turbine blade. The specifications of BladeSave AE sensors are
presented in Table 5. It is worth noting that BladeSave’s AE sensors do not require
preamplifiers, which are common in electronics sensing systems.
Table 5: BladeSave AE sensor specifications
Parameter Value
Dimensions OD 38 x H 25 mm
AE Range 500 - 800 mm radius depending on substrate
AE Sensitivity typically 35 nm / Pencil Lead Break (PLB)

Page | 18
Frequency range DC to 300 kHz

Temperature Sensitivity 11 pm/°C
Weight < 50 gram
Fibre Type Single Mode SMF-28, 9/125 µm
BladeSave’s FBG accelerometers offer sufficient dynamic range and frequency coverage for
structural vibration monitoring, which is ± 1,000 ms-2 and up to 900 Hz, respectively.
Table 6: BladeSave accelerometer specifications
Parameter Value
Dimensions 16.5 x 19 x 29 mm
Typical Sensitivity at 159.2 Hz 2.0 to 2.5 pm/ms-2
Frequency Range (± 10 %
DC to 900 Hz
Amplitude)
Mounted Resonance Frequency > 3 kHz
Dynamic Range (Peak) ± 1,000 ms-2
Broadband Noise Estimated < 0.1 ms-2
Temperature Sensitivity ~10 pm/°C
Thermal Transient Sensitivity ~0.1 ms-2/°C
Mass 25 g
Two interrogators are used for strain/vibration and AE measurements. For strain and vibration
monitoring, SmartScan interrogator is used. Up to 4 channels can be directly connected to the
interrogator, which can be expanded to 16 with an optic switch. The hardware supports up to
25 kHz sampling rate for blade vibration monitoring. The specifications of SmartScan is in Table
7.
Table 7: BladeSave SmartScan specifications
Parameter Value
Wavelength Range 40 nm (1528 – 1568 nm)
Number of Optical Channels1 Up to 4
Maximum Number of Sensors/Channel 16
Scan Frequency (all sensors simultaneously) 250Hz

Page | 19
Maximum Scan Frequency (with reduced

25 kHz
wavelength range)
< 5 pm over operating temperature
Wavelength Stability
range, ± 20 pm over 25 years
Dynamic Range (laser launch power minus
37 dB
detection noise floor)
Dimensions 140 x 115 x 85 mm
Weight 0.9 kg
Operating Temperature -15 to +55 °C
SmartSonic interrogator is used for measuring AE events. The SmartSonic system supports 160
dB dynamic range, and a sampling rate of up to 600 kHz to be used with FBG AE sensors. The
hardware supports up to 64 FBG AE sensors to be connected with the use of optic switches.
The specifications for SmartSonic is in Table 8.
Table 8: BladeSave SmartSonic specifications
Parameter Value
Displacement Noise Down to 600 fε / √Hz
Dynamic Range 160 dB
Sampling Speed DC up to 600 kHz
Signal Acquiring True simultaneous
Operation Wavelength C & L band
Sensor Potential Up to 64
Data Output USB / LabVIEW
Fibre Type Single mode SMF-28, 9/125 µm
Mass 7 kg
Dimensions H x W x D 100 x 450 x 310
Operating Temperature Range -15 to +75 °C
Please note that, this section aims at giving an overview of hardware components for the
BladeSave system, however, the clients are not required to interact with the components as
they are installed, commissioned and configured by trained personnel from the consortium
partners during installation. The clients can interact with the user interface which is presented
below.

Page | 20
3.2.3 User Interface demonstration
From the BladeSave user interface, users can handle the aspects related to other users, wind
farms, turbines, logs, and alarms. The user interface is accessible from any browser, and from
any device (PC, tablet, mobile) that is connected to the internet.
- Login Page
This page contains a form that allows users to authenticate in the app. Users are also able
to recover their password in case they have forgotten it.
Figure 3 BladeSave User Interface Login Page
- Wind Farm Page
After login, the user is redirected to the main screen. In this page, all the turbines are displayed
in a table. For each turbine, the following information is displayed: wind farm name, turbine
name, status, ID, serial number and location. The user can filter them according to the wind
farm or status. The turbines with no logs in the last 24 hours are marked with orange.

Page | 21
Figure 4 BladeSave Wind Farm Page
- Wind Turbine Page

This page is divided into 3 main sections: Control panel, Logs section and Activity section.
In the Control panel section, the user can see information about the turbine (turbine name,
location, exact position on the map, type, model, serial number, status etc.) and send
commands to the server running inside the turbine (Stop, Reboot, Get logs). Also, from this
section the user can upload or download the wind turbine configuration.
The user can update or delete the turbine in a similar way to the one presented in the previous
step, Update/Delete farm.
Figure 5 BladeSave Wind Turbine Page

Page | 22
Logs section contains all logs for the current turbine. They are divided into 2 categories: List of
Defects (issues), Logs (all logs) and they are displayed in tabular form, in the order in which
they were issued (the most recent first), in sets of 25 logs per page. More details about each
log can be seen by clicking on it. From this section the user can mark the defects logs as Fixed
or ACK (acknowledged) and add manually new defects log.
Figure 6 BladeSave Wind Turbine Logs Section
3.3 System qualifications
The BladeSave system has also been tested against the safety standards and regulations as
listed below:
● Directive 2014/30/EU, EMC Directive

● Directive 2014/35/EU, Low-voltage Directive (LVD)
● Directive 2006/25/EC, Artificial Optical Radiation
● Directive 2011/65/EU, The Restriction of Hazardous Substances in Electrical and
Electronic Equipment (RoHS)
● EN 61000-6-4, Electromagnetic Emission
● EN 61000-6-2, Electromagnetic Immunity
● EN 61010-1, Safety (LVD)
● BS EN 60825-1, Safety of laser products - Part 1: Equipment Classification and
Requirements
● UNE-EN 60068-2-2:2008, Environmental testing. Part 2: Tests. Test B: Dry Heat.
● UNE-EN 60068-2-6:2008, Environmental testing. Part 2-6: Tests. Test Fc: Vibration
(sinusoidal)

Page | 23
● UNE-EN 60068-2-27:2011, Environmental testing. Part 2-27: Tests. Test Ea and

Guidance: Shock
● UNE-EN 60068-2-78:2002, Environmental Test. Part 2-78: Test Cab: Damp Heat, Steady
State
● BS EN 61326-1, Electrical equipment for measurement, control and laboratory use.
EMC requirements
The manufacture of BladeSave system, SmartFibres Ltd / Halliburton, has declared CE

conformity for BladeSave under EU directives specified in the list above. The test reports can
be provided upon requests from the Consortium.
3.4 Case study – testing on wind turbine blades
3.4.1 Demonstration on EWT wind turbine
The complete BladeSave system is designed to function autonomously for an extended time,
under the operational condition of a wind turbine. The system must cause no interference to
the operation of the wind turbine. To demonstrate that the system fulfils these requirements,
the BladeSave system has been demonstrated and tested in both real operational
environments as well as laboratories.
For the system demonstration in an operational environment, project partner EWT gave access
to a fault-free, newly installed wind turbine, on whose blades the BladeSave system was
installed for 4 months. This onsite trial aimed to demonstrate that the system was capable of
continuously operating for extended time in real conditions, without causing issues for the
wind turbine. The system for installation is illustrated in Figure 7 and Figure 8.

Page | 24
Figure 7 BladeSave system part 1 where (a): processing unit; (b): SmartScan system for vibration monitoring; (c):
SmartSonic system for AE monitoring and (d): FBG accelerometers
Figure 8 BladeSave system part 2 where (e): optic switch; (f): military level optic cables and (g) fibre optic AE
sensors
The BladeSave system was installed to monitor all three blades (labelled Blade 1, 2 and 3) on
the wind turbine, each with six AE sensors, four strain patches and an accelerometer. Thanks
to the optic switch, all AE sensors were managed by a single SmartSonic hardware. All strain
patches and accelerometers were managed by a single SmartScan hardware.

Page | 25
Figure 9 BladeSave recorded AE events
The demonstration started in March 2019 and completed in July 2019. During the four months
demonstration, AE events were monitored at all times for the blades, processing and cloud
server were tested. An illustration of the AE recordings during this time is shown in Figure 9,
where the AE energy was produced from raw AE waveform captured on Blade 1. Although it
was concluded that no mechanical faults were developed during the demonstration, the
system was shown to be able to robustly work autonomously on the operational wind turbine
environment. Positive feedback was obtained from the wind turbine owner and potential end
user for BladeSave, EWT:
“This statement concerns the customer feedback of EWT regarding the system demonstration
phase as defined in work package 3 of the BladeSave project.
… Therefore, EWT is content with the performance and results from the BladeSave system and
can provide valuable information about the health of the blades. This information will detect
blade problems at an early stage before the blade damage becomes large. Small damages are
much less costly to repair than large damages. This makes the BladeSave system an attractive
proposition to reduce the levelized cost of energy (LCOE) of wind turbines.”

Page | 26
More details for this demonstration are available from the deliverable D3.1 – System
demonstration with WFOs, which can be downloaded from project website
https://bladesave.eu/ once it is approved.
3.4.2 Validation on detecting blade cracks induced from bending tests
To validate AE’s capability in detecting cracks on the turbine blade, controlled experiments
were undertaken in project partner TWI's laboratory.
Figure 10 Wind turbine blade for bending tests
A 4-metre wind turbine blade was adopted with a 50 mm crack inserted, as shown in Figure
10. The test aimed to stimulate the growth of the crack by applying bending stress near the
crack area, or monitor any naturally occurring defects during this process.
A commissioning test was first undertaken, where load was gradually applied to force the
blade centre to bend downwards, increasing the strain near the cracking area. In the
commissioning test, the bending on the blade developed a 200 mm debonding line along the
blade's trailing edge at 0.8 kN. Afterwards, the sensors were rearranged to the debonding line
as shown in Figure 11.
It was shown in Figure 12 that the blade was crushed intentionally by the load machine, and
the debonding defect along the trailing edge of the blade was enlarged to 600 mm. The
process was continuously monitored and detected by the BladeSave system, which the
real-time monitoring results were illustrated in Figure 13. The BladeSave system continuously

Page | 27
monitored the AE events that occurred during the bending test, and produced AE parameters
including cumulative energy, AE counts, peak amplitude and frequency peak. These
parameters were also demonstrated in chronological order in Figure 13.
Figure 11 Sensors installation for debonding line monitoring
Figure 12 Test blade with debonding defect developed

Page | 28
Figure 13 BladeSave real-time monitoring of bending test
In the cumulative AE count and cumulative AE energy plots (upper left and upper right,
respectively), the results showed multiple small steps until 600 to 800 seconds after the test
started. These steps were caused by the gradually increased loading (displacement) as well as
developed micro damages along the debonding line. Similarly, the AE amplitude events
(bottom left) in Figure 13 demonstrated sparse AE events captured by the BladeSave system
before 14:00 pm. All three aforementioned parameters showed evident increase near 14:00
pm, when, upon post inspection, the debonding was fully developed. The frequency peak
parameter (bottom right) shows that before the major debonding crack occurred at around
14:00 pm, majority of the AE frequencies were below 50 kHz, which was likely to be related to
low frequency AE activities such as matrix delamination and cracking. By the time the
debonding was fully developed, significant AE activities were captured at 150 kHz, which can
be related to different fault mechanism during the formation of the debonding crack, such as
fibre breakage.
BladeSave system was able to monitor and detect the AE events caused by the development of
the debonding crack at the trailing edge of the blade. Detailed test discussions are available
from the deliverable D3.1 – System demonstration with WFOs, which can be downloaded from
project website https://bladesave.eu/ once it is approved.

Page | 29
3.4.3 Other validation tests
The BladeSave system has been tested on cyclic loading and simulated ice accretion tests,
which showcased the robustness and icing detection capabilities of the system using vibration.
To keep this document concise, more discussions are available from the deliverable D3.1 –
System demonstration with WFOs, which can be downloaded from project website
https://bladesave.eu/ once it is approved.
4 Conclusion
In this white paper, the booming market of wind energy and the enormous potentials for
condition monitoring implemented for wind turbine blades are explained. The BladeSave
system is specially designed to cope with the needs. As demonstrated in this paper, the system
brings many advantages to be deployed inside a wind turbine blade thanks to the fibre optic
technology. To be specific, the BladeSave system offers benefits in four categories:
● Structural monitoring capabilities: the BladeSave incorporates three different

condition monitoring techniques, including strain, vibration and AE. The multiple
sensing offers structural information that helps users to gain more awareness of their
wind turbines. The strain monitoring provides fatigue cycle evaluation and pitch
control, which have been commercially implemented on EWT turbines. Ice accretion
monitoring using vibration and blade cracks monitoring using AE are demonstrated in
lab on simulated conditions. BladeSave system is currently the only complete fibre
optic system on the market that offers multi-sensing, onsite real-time analysis and
integrated web-based user interface, which make the system very unique.
● System scalability: thanks to the non-electrical sensors and cables, the measurements
do not suffer from signal loss caused by long cables. The system can be scaled to
monitor larger wind turbine blades with modified longer optic cables, and
incorporating optic switch to connect up to 64 AE sensors and 16
strain/accelerometers into the system at the same time, without the need to expand
the interrogator hardware. These features offer easier installation, longer detection
range and better scalability compared to traditional electronics sensors.

Page | 30
● User friendliness: software is pre-installed on the processing unit and is configurable.

All hardware is transparent to the end users. Monitoring results are fed to the web
based graphical user interface (GUI) in real time and are accessible by registered users.
● Noise immunity: compared to electronics sensing technology, sensors manufactured
based on FO technology are completely passive and are immune to EMI and static
electricity, which is vital to guarantee the quality of measurements in wind turbine
operating environment. BladeSave AE sensors does not require any preamplifiers to
ramp up the signals. Additionally, another unique feature is that multiple sensing
elements can be embedded into a single fibre cable to further increase number of
sensors that a single channel can host.
Meanwhile, to respond to the issues identified in current ISO 16079-1 related standards, the
BladeSave system could contribute as described below:
● ISO 16079 series standards are established to give guidelines for implementing FMECA
and FMSA for wind turbine monitoring. ISO 16079-1 offers a non-exhausted list of
wind turbine failure modes, which does not include various blade failure mode that
BladeSave targets, such as blade shell crack, debonding, matrix delamination and glass
fibre breakages. ISO 16079 subsequent standards should consider establish guidelines
specifically for blade condition monitoring, which BladeSave system focuses on.
● ISO 17359 recommends monitoring parameters only under non-dynamic condition,
which is difficult to implement for wind turbine blade condition monitoring. Condition
monitoring system will likely to be unable to detect faults propagation under
non-demanding operating condition, if the system only collects data under such
condition. BladeSave system provides a full-on solution to track the progression of the
potential faults, as shown in Figure 9 and Figure 13.
In summary, as currently no standards have been established specifically for wind turbine
blade, it is suggested that the standard committee for ISO 16079 series to consider
incorporating the BladeSave system’s fibre optic enabled AE condition monitoring for wind
turbine blades, into future development of such standard .
This white paper will also be available to download from project website https://bladesave.eu/
once it is approved.

Page | 31

Wind Turbine Blade

Uploaded by

Copyright:

Available Formats

Wind Turbine Blade

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Wind Turbine Blade

Uploaded by

Copyright:

Available Formats

Ref.

Risk Based Technology for Blade Structural Assessment

Work Package Title WP4 – Quality Attainment

Project's Coordinator: ​Mr. Benn Faulkner

Version Issue Date Stage Changes Contributor(s)

Copyright ©, ​BladeSave Project Consortium

Copyright ©, ​BladeSave Project Consortium

Figure 1 BladeSave System Concept & Approach 13

Copyright ©, ​BladeSave Project Consortium

Copyright ©, ​BladeSave Project Consortium

1.2 Relation to ISO 16079-1 and associated standards

1.2.1 Relation to ISO 16079-1

Copyright ©, ​BladeSave Project Consortium

1.2.2 Relation to associated standards

Standards Methodology Scope

16079-2, 3 … FMSA Specify how to detect a certain failure mode

● Taking measurements of different parameters wherever possible at the same time or

Copyright ©, ​BladeSave Project Consortium

2 The need for wind turbine blade monitoring

2.1 Description of the issue and its context

Copyright ©, ​BladeSave Project Consortium

measurements. Damage characterisation was usually performed by visual inspection for

Copyright ©, ​BladeSave Project Consortium

2.2 Industry trends albeit global lockdown

2.2.1 Impact on the global economy and clean energy market

Copyright ©, ​BladeSave Project Consortium

2.2.2 Impact on global wind industry

Vattenfall Hollandse Zuid Netherlands $3.9 billion 1.5GW 2021-23

SSE Seagreen UK $3.8 billion 1.1GW 2021-23

Changfang Xidao Taiwan $3.6 billion 600MW 2021-23

Fecamp/St Brieuc France $5.4 billion 993MW 2021-23

2.2.3 Impact on policy and regulation

Copyright ©, ​BladeSave Project Consortium

Copyright ©, ​BladeSave Project Consortium

3.1 Description of the system

Copyright ©, ​BladeSave Project Consortium

Figure 1 BladeSave System Concept & Approach

3.2 How does the system work

3.2.1 System components

A list of BladeSave system equipment is shown in Table 3.

Table 3: List of BladeSave system components

Category Component Functions

Copyright ©, ​BladeSave Project Consortium

Copyright ©, ​BladeSave Project Consortium

Figure 2 BladeSave system integration

3.2.2 Specifications of hardware

Copyright ©, ​BladeSave Project Consortium

Table 4: SmartPatch specification

Patch Dimensions 100 x 15 mm

Table 5: BladeSave AE sensor specifications

Copyright ©, ​BladeSave Project Consortium

Frequency range DC to 300 kHz

Table 6: BladeSave accelerometer specifications

Table 7: BladeSave SmartScan specifications

Copyright ©, ​BladeSave Project Consortium

Maximum Scan Frequency (with reduced

Table 8: BladeSave SmartSonic specifications

Copyright ©, ​BladeSave Project Consortium

3.2.3 User Interface demonstration

Project's Coordinator: Mr. Benn Faulkner

Copyright ©, BladeSave Project Consortium

Copyright ©, BladeSave Project Consortium

Copyright ©, BladeSave Project Consortium

Copyright ©, BladeSave Project Consortium

Copyright ©, BladeSave Project Consortium

Copyright ©, BladeSave Project Consortium

Copyright ©, BladeSave Project Consortium

Copyright ©, BladeSave Project Consortium

Copyright ©, BladeSave Project Consortium

Copyright ©, BladeSave Project Consortium

Copyright ©, BladeSave Project Consortium

Copyright ©, BladeSave Project Consortium

Copyright ©, BladeSave Project Consortium

Copyright ©, BladeSave Project Consortium

Copyright ©, BladeSave Project Consortium

Copyright ©, BladeSave Project Consortium

Copyright ©, BladeSave Project Consortium

Copyright ©, BladeSave Project Consortium

Copyright ©, BladeSave Project Consortium

Copyright ©, BladeSave Project Consortium

Copyright ©, BladeSave Project Consortium

Copyright ©, BladeSave Project Consortium

Copyright ©, BladeSave Project Consortium

Copyright ©, BladeSave Project Consortium

Copyright ©, BladeSave Project Consortium

Copyright ©, BladeSave Project Consortium

Copyright ©, BladeSave Project Consortium

Copyright ©, BladeSave Project Consortium

Copyright ©, BladeSave Project Consortium

Copyright ©, BladeSave Project Consortium