Skip to main content
European Commission logo print header

Continuous health monitoring and non-destructive assessment of composites and composite repairs on surface transport applications

Final Report Summary - COMPAIR (Continuous health monitoring and non-destructive assessment of composites and composite repairs on surface transport applications)

Executive Summary:
Composites are used in a wide range of applications in surface transport and its use has increased steadily over the last two decades. An increasing number of safety critical parts are now being manufactured from these advanced composites and this trend is expected to continue. The use of fibre-reinforced composite materials has increased in both low and high technology engineering applications over recent years - apparent within the surface transport and automotive industry. Increasingly, advanced carbon-fibre reinforced plastics (CFRP) and glass-fibre reinforced plastics (GFRP) are providing an attractive and economically alternative to aluminium and steel due to the fact that they are light, robust and help to reduce C02 emissions. However, the expansion of composite use has been limited by the slower pace of developments in suitable NDT techniques for both manufacture and in-service inspection. Yet, if the initiated damage process goes undetected and reaches an advanced stage then significant outage is incurred: materials and time may be required to fix the panel. To address this, a process of inspection and monitoring was proposed.

Several European transport manufacturers and operators have identified a need for the development of NDT technologies suitable for the inspection of advanced composites in transport applications. A need has been identified for a wider range of swift and accurate NDT techniques to meet the current and future inspection requirements that underwrite structural integrity of composites and composites structures.

The ComPair project comprises 11 partners from 6 European countries. The aim of the consortium is to build a new and novel system for continuous health monitoring and non-destructive assessment of composites and composite repairs on surface transport applications. There are three categories of interest: road including high-end cars, trucks and buses; rail; and marine including top-end yachts. As an addition, guidelines on a certified procedure for proper damage assessment have been initiated.

The ComPair research has resulted in the development of a flexible, mobile, robotic scanner to provide quick and easy on-site inspection of composite panels. The scanner can easily be moved into position to inspect the panel of a train or truck at rest. The adaptable scanner head permits inspection of glass fibre reinforced plastic (GFRP) using Near Infra-red (NIR) testing with a Basler camera; and inspection of carbon fibre reinforced plastics (CFRP) using transient thermography techniques with a FLIR camera. The scanner head allows for easy exchange to satisfy the composite, or a mix of components under test. The ComPair software successfully interacts with the scanner control software to ensure smooth, consistent, and reliable system operation. The interactive software offers the operator an option for acceptance criteria prior to inspection. Once testing has been initiated, the robotic scanner captures images of the composite panel under test. The smart head ensures that image capture is consistently orthogonal with the panel surface. For curved panels under test, the software adjusts the scanner head angle to maintain the orthogonal set-up condition between the direction of image capture and the panel surface. Once the whole panel has been inspected, the software stitches the images together to produce an overall panel image to the operator. The software then performs ‘sentencing’ and provides a visual representation of defect detection.

A monitoring technique has been developed that is able to inspect composite structures in real time. The system comprises the combination of Acoustic Emission (AE) and Long Range Ultrasonic Guided wave techniques. The compatibility of both NDT techniques allows for a high-sensitivity to change in composite components whether under loading conditions or due to elastic activity within the panel under inspection. The combined ComPair AE/LRU inspection technique discriminates between stress activity due to sensor activity and external conditions. Signal conditioning using signal processing techniques permits defect detection and localisation. Spatial pattern recognition software indicates the location of the defect, where it occurred and when it exceeded set acceptance criteria limits.

The project has been split into three reporting periods with accompanying reports: Periodic Report 1, Periodic Report 2 and Periodic Report 3. The first report covered the following details:
• Typical composite materials used on surface transport applications
• Modelling of composites to ascertain the natural modes of propagation
• Thermal NDT techniques most appropriate for this application
• Specifications of the required scanner
• Disseminations and marketing opportunities

This was summarised in D1.1 D2.1 and D7.1. Work towards D3.1 was also presented within periodic Report 1.
Milestones completed within this period were:
M1: Project requirements and specifications – preparation of materials – coupons to be tested.
M2: NDT systems for manufacturing and assembly stages
The feedback for Periodic Report 1 was confined to the body of content within that document. As requested, the publishable summary was written with a concise narrative and in a ‘modular’ way, such that Section 1 may be ‘decoupled’ from the Periodic Report (if so desired) whilst remaining a ‘stand-alone’ document. The Deliverable table was reconfigured to comply with template guidelines. An explanation was provided as to the use of resources and all comments relating to finances addressed. All items listed in the correspondence of 12/10/10 were subsequently amended to the satisfaction of the EC Officer
Periodic Report 2 covers the second reporting period. This includes:
– Development, manufacture and optimisation of a robotic scanner that is able to easily accommodate the appropriate cameras for transient thermography and NIR NDT techniques (WP4)
– Flexible movement of the scanner for inspection of curved composite samples (WP4)
– Development of integration of scanner control software and ComPair inspection software for the purposes of image capture, processing, sentencing and visualisation (WP5)

– Integration of AE and LRU Guided wave techniques for corroborative in-situ composite inspection with appropriate signal discriminatory application. (WP3 & 5)

- Market research analysis with provision of a market strategy document and the final PUDK (WP7)

Deliverables completed within this period were D3.1. Work towards D4.1 D5.1 and D6.1 was also presented within Periodic Report 2.
Milestones completed within this period were:
M3: Real time monitoring of vehicles for maintenance.
Periodic Report 3 covers the third reporting period. This includes:
– Validation of the robotic scanner to easily accommodate the appropriate cameras for transient thermography and NIR NDT techniques (WP4)

– Full integration between the scanner control software and the ComPair inspection software for the purposes of image capture, processing, sentencing and visualisation (WP5)

- Perform laboratory and field trials to demonstrate the capability of the monitoring and inspection system (WP5)

– Demonstration of the AE and LRU Guided wave techniques for corroborative in-situ composite inspection with appropriate signal discriminatory application. (WP 5)

- Provision of an operator’s manual for best practice guidelines for operational use and integrated methodology (WP6)

- Conduct market research and strategy (WP7)

Deliverables completed within this period were D4.1 D5.1 D6.1.
Milestones completed within this period were:
M4: Robotic scanner and robot integration with thermal NDT (Month 30)
M5: Proving trials complete (Month 36)
M6: Guidelines, training and certification complete (Month 36)

The public website address for the project is www.compairproject.com this website will be kept as the main platform for any communications related to the ComPair project beyond the project term.

The ComPair project was successful and achieved all intended objectives. The prototype is an effective and reliable inspection device that can perform quick and simple inspection of composite components.

ComPair is collaboration between the following research organisations and SMEs from 6 different EU countries: TWI Ltd, VTT, CERETETH, ATOUTVEILLE, ENEA, KTU, G-Tronix, KCC, NTUA, HEXCEL, ENVIROCOUSTICS.

Project Context and Objectives:
The ComPair project has involved analysis of composite materials – Glass Fibre Reinforced Plastic (GFRP) and Carbon Fibre Reinforced Plastic (CFRP) – commonly found as panels on surface transport applications. The context of the project has been to develop a reliable and robust system of composite NDT in-service monitoring and in-situ inspection that is able to meet market demand and assist with the ‘roll-out’ of composite materials as part of rail, HGV and automotive structural design.
The ComPair project objective is to assist surface transport OEMs who - under pressure by government directives on CO2 emissions – have intensified their efforts to lighten vehicle weights for improved energy consumption. As a consequence, this will help allay the concerns of service providers and car owners around the world regarding fuel economy.
The main objective of the ComPair project is to integrate and validate the NDT techniques (transient thermography, NIR imaging) with the robotic scanning facility to provide a complete in-situ NDT Inspection system. The ComPair project comprises two techniques: in-Situ NDT Inspection; and NDT monitoring including the following:
– Identification of material types and geometries typically used in transport applications.
– NDT inspection techniques suitable for in-situ inspection
– Modelling and testing of composites using AE/LRU techniques
– Development, manufacture and optimisation of a robotic scanner that to easily accommodate the appropriate cameras for transient thermography and NIR NDT techniques (WP4)
– Flexible movement of the scanner for inspection of curved composite samples (WP4)
– Integration between the scanner control software and the ComPair inspection software for the purposes of image capture, processing, sentencing and visaulisation (WP5)
- Perform laboratory and field trials to demonstrate the capability of the monitoring and inspection system (WP5)

– Integration of AE and LRU Guided wave techniques for corroborative in-situ composite inspection with appropriate signal discriminatory application. (WP3 & 5)

- Provision of an operator’s manual for best practice guidelines for operational use and integrated methodology (WP6)
- Conduct market research and produce a market strategy document and the final PUDK (WP7)

Project Results:
The first 18 months of the ComPair project mainly concentrated on the three first work packages (WPs).

In the first six months of the project, the majority of the activity focused on WP 1 and WP 2.
The outcome of WP 1 was the identification of the composite materials typically used in European ground transport applications. This information was provided to Hexcel who subsequently developed various composite samples (with and without various defect types) for the purposes of experimental research for AE/LRU and thermal imaging testing techniques. The outcome of WP2 was the completion of composite modelling by KTU for the purposes of LRU guided waves to assess the effects of fibre orientation on the available modes of propagation. An investigation as to the applicable Guided wave techniques was completed by TWI and KTU, and an investigation into thermal NDT techniques suitable for inspection was completed by NTUA, G-Tronix, Cereteth and ENEA. At the 6 month meeting, WP 1 was completed and WP 2 was in good progress whilst at the same time plans for dissemination and exploitation were at a formative stage (WP 7). Periodic Report 1 (covering the first 18 months of the project progress) reported on the completed work of WP 1 and 2 with status updates on the progress of WP 3 and WP 4. Progress on the dissemination activities of WP7 was also included within Periodic Report 1.

The second reporting period commenced with the completion of WP 3 and the initiation of WP4 (starting promptly on month 6 of the project). WP3 completed the AE/LRU techniques - investigated by TWI, KTU and Envirocoustics – that would serve the requirements of the ComPair NDT monitoring system. Development of the robotic scanner was conducted by Cereteth as part of WP4 which also witnessed the completion of the integration of the thermal imaging techniques: NIR and transient thermography (NTUA and G-Tronix). Development of the ComPair interactive software with NDT visualisation was undertaken by KCC.

WP 4 continued from month 6 (as scheduled) to month 34. This was later than scheduled due to the temporary unavailability of equipment and software, however; whilst this did not impact on the overall schedule of the project, it did impose a slight delay on WP5 (which commenced on schedule on month 24) to allow for the implementation of system modifications arising from field trials. Subsequently, this delay imposed a 1 month delay on the completion of WP6 on guidelines and certification.

WP5 involved the integration of the NIR imaging and transient thermography techniques together with the robotic scanner and the interactive NDT software. The outcome of this work provided a ‘debug-list’ to be addressed prior to the demonstration. WP6 provided guidelines for the use of the robotic scanner and accompanying NIR and thermography techniques as well as the operation of the interactive software. Guidelines were also provided for the use of the ComPair NDT inspection system (AE/LRU).

The outcome of WP7 was the identification of relevant markets and how the project could be promoted. This included completion of a SWOT analysis, attendance at suitable exhibitions and conferences, and the provision of appropriate marketing material for promotional use (ENEA, Atoutveille).

At the 30 month stage of the project, the consortium agreed that an extension to the project term would be highly valuable in term of promoting and disseminating the good research of the project. A composites exhibition (JEC Show Paris 2011 – at which the ComPair consortium had a stand) was imminent and the project partners wanted to be able to exploit the feedback to inform the planned workshop in Rome and to establish contact with other consortia. The extension granted; another workshop was scheduled for November 2011 in Cambridge.

These work packages have been reported in Periodic Report 2 and Periodic Report 3. Additional details of the technical and scientific results for each work package are now presented.

WP1: Project specifications- Provision of work to be assessed

Mainly consisted of discussions with consortium partners to determine the performance requirements of an NDT system suitable for in-situ monitoring of surface transport vehicles. This involved research into the nature of the composite materials used, their geometry and the nature of their fitting. Subsequently, a literature review and a market research were undertaken by project partners (Atoutveille, ENEA) to establish and select the relevant project applications. The review concluded that the market needs demanded the project applications to be high and low speed trains, buses, heavy goods vehicles (HGV) and trams. A review on the applications of composites, in terms of materials used, manufacturing technologies and defects for each application type was also provided. Material types of interest included laminates with carbon or glass fibres, and sandwich structures with different skins and with either aluminium honeycomb or polymer foam core. It was established that the most common composite material in use were laminates comprising carbon-fibre reinforced plastics (CFRP) and glass-fibre reinforced plastic (GFRP).

Thereafter, three sets of GFRP and CFRP) samples with defects and no defects were produced and used in experiments to investigate the various NDT techniques. These samples were identified as Stage 1, Stage 2 and Stage 3 and the details are provided in Figure 1, Figure 2, and Figure 3 respectively. Whilst the Stage 1 and Stage 2 samples were suitable for the purposes of AE analysis, the Stage 3 samples were required in order that good data may be obtained when analysing long range ultrasonic (LRU) Guided waves. The sample dimensions and characteristics were agreed by TWI, Envirocoustics, KTU, G-Tronix, NTUA and Cereteth. The manufacture of the samples was undertaken by Hexcel.

WP2: Theoretical study and Modelling
The objective of WP 2 was to develop a theoretical model to understand the properties of Guided waves in composite materials as well as understanding the parameters that will determing defect presence using the selected thermal imaging techniques. Each techniques was optimised for the purpose it was due to serve. The aim of this analysis is to help improve the probability of defect detection in composite mataerials. Furthermore, the output from this work informs the inspection procedure described in WP6.

To achieve successful modelling for analysis of the LRU Guided wave in-service inspection technique the elastic properties of the material are required – and provided in Table 1. It is observed from Figure 4 that at low frequencies (up to 500kHz), there are a number of Lamb wave modes. The sample thickness modelled here is 8.4mm. To estimate the accuracy of the modelled data, experiments were conducted on the available samples at KTU. Figure 5 shows the experimental method and Figure illustrates good agreement between the model and the experimental data for a few of the detected modes. The data demonstrates that the lower frequency S0 mode is detected and may be used for the purposes of defect detection. This analysis was for the defect-free scenario.

The consortium agreed that the two thermal techniques best able to detect defects (on the composite panel sizes typically used on ground transport vehicles) were transient thermography and Near Infra-Red (NIR). Although vibrothermography is extremely fast in either lock-in or burst configuration, is more suitable for relatively small objects. For this reason, the consortium pursued NIR as an accompanying inspection method to transient thermography.

Thermal tests were performed on two different test samples: 300mmx300mm GFRP and CFRP. Two different long-range thermal cameras were investigated: ThermaCAM™ S65 and FLIR P640.

Three active methods were used to induce the temperature difference into the samples:
• pulsed thermography (Figure 6)
o flash method (xenon flash lamps)
o 2 * 4 kJ
o heating time 0.1 second
o monitoring 10 to 20 seconds, 50 Hz
• transient themography (Figure 7)
o IR-heating pulse
o heating time 30 seconds
o monitoring 10 to 30 minutes
• thermographic inspection for cooled sample (Figure 8)
o freezing in -20 °C
o freezing time min 3 hours
o monitoring 30 to 60 minutes

Figure 8 shows an individual thermal picture taken on the CFRP sample from a sequence for the FLIR P640 camera which featured a superior resolution when compared with the ThermaCAM S65 (twice that of the S65 in both x and y directions).

The NIR imaging technique can relatively easily penetrate GFRP materials using the lock-in NIR imaging technique. Two methods of NIR data collection were analysed
• a line scan across an object
• a video camera optimised for NIR operation

The line array detector of photo-diodes worked well and would be highly suitable for factory and fabrication environments where (for example) product continuously flows over such a line sensor placed under a conveyor belt. However, emphasis was generally focused on the camera approach since this technique provided the highest resolution of detail and throughput speeds with shortest processing times.

Figure 9 shows a 6.35mm countersink defect as being clearly visible using NIR camera imaging. Figure 10 illustrates the clear NIR image of a 5mmx5mm delamination defect. The front sided image result of the defect was clearly identified. The reverse sided image detects the same delamination; however, it was much larger in size.

NIR imaging is appropriate for GFRP materials since it exhibits some transparency and is thus susceptible to EM radiation (NIR wavelengths being between 800nm and 1,000 nm). However this method is not appropriate for CFRP materials since they are opaque and thus thermography imaging (3-5μm and 7.5-14μm wavelengths) are best suited for the purposes of producing thermograms, which are the result of thermal emissions from the specimen surface.

In summary, NIR was able to detect on GFRP:
• countersunk holes down to 12mm
• burned drilled holes down to 6mm
• impact damage (20J) down to 4.5mm diameter
• delamination down to 5mm

Thermography was able to detect on CFRP:
• countersunk holes down to 12mm
• burned drilled holes down to 6mm
• impact damage 60J) down to 6.5mm diameter
• delamination down to 10mm

WP3:Real Time Monitoring of Composite Components on Vehicles

The objective of work package 3 was to develop a suitable NDT technique for purposes of in-service monitoring of the structural integrity of composite materials.

The effectiveness of acoustic emission (AE); long range ultrasonic (LRU) Guided waves’ and their eventual integration was investigated.

An investigation as to AE performance and how it varies dependeing upon the sensors used, the accompanying hardware and the interacting software was investigated on the honeycomb sandwich specimen. Details are provided in Figure . Further to that, two different types of couplant were investigated. The following conclusions were determined:-

• The low-frequency R6/R15 sensor (250-500kHz) performed best under attenuation tests than the high-frequency or wideband sensors
• Material directionality had little impact on signal attenuation when tests were performed at horizontal, vertical or ±45° orientations using low-frequency sensors
• The two adhesive couplants analysed: Silicon grease (OKS 1110) and Adhesive silicon (Scotch 1201 Resin silicone) provided little difference in sensitivity for low-frequency sensors.
LRU Guided wave analysis was conducted using Macro Fibre Composite (MFC) sensors. M2814 Macro Fibre Composite (MFC) transducers were identified as an appropriate method for the detection of Guided waves within composite materials. The conformable, interdigitated, sensor predominantly excites in-plane wave modes which were identified as being most susceptible to change in the presence of a defect. Attenuation measurements, conducted to ascertain the sensitivity to elastic activity in GFRP, were recorded to be -55dBm-1, Figure .
Group velocity analysis, in relation to whether the sensor is aligned with the sample fibre orientation or non-aligned, revealed that the S0 (fastest in-plane mode) significantly changes wave velocity upon material orientation: ~3000m/s on fibre; ~2500m/s off-fibre, Figure .
Generated dispersion curves of LRU in a bidirectional GFRP copmposite (Figure) illustrates that for frequencies up to approximately 120KHz, the S0 mode does not suffer from the effects of dispersion – an important attribute since dispersion is often one sign of a change in the material properties. As is observed, the A0 mode exhibits dispersion for the full range of frequencies up to 150kHz.
An integrated laboratory test to assess integration of both AE and LRU techniques was performed to explore the level of AE/LRU sensor compatibility. A sparse array of AE sensors interleaved with a sparse array of LRU sensors were bonded to the Stage 3, 1.3mx1mx7.4mm 32-ply biaxial GFRP plate. Placement of both sensor types was completed on the same surface of the composite panel.
This aim of these laboratory trials was to achieve:-
• Individual performance criteria of AE array and LRU array on 1mx1.3m GFRP sample
• Ascertain level of discrimination reported by AE system of LRU sensors
• Integration of AE and LRU Guided wave sensors on a sample GFRP plate

The following technical actions were undertaken:
• A 3-point compressive loading test applied to the GFRP plate.
• Load applied (up to 7kN) by support rollers of 70mm, 75mm dia.
• Sample strategically affixed with ComPair LRU (M2814 interdigitated) sensors and AE sensors.
• Sample strategically affixed with PAC PICO and R15 AE sensors
• 15-point load cycle was applied over a period of 2.5 hours.

The conclusions of the experiment were:
• AE system and pattern recognition software successfully detected and accurately located the LRU Guided wave transmitters each time they pulsed.
• Demonstrated feasibility and compatibility of simultaneous application of the two techniques.
• No significant AE activity due to loading was observed and no significant planar-located events were present throughout the test.
• No significant LRU activity due to loading was observed.
• Maximum possible load applied was limited and not adequate to introduce further damage or existing defects’ growth.
The experiment successfully proved that the two sensor systems are compatible and was a demonstration of the feasibility of simultaneous application of the two NDT techniques. Hsu Neilson testing and velocity measurements indicated – using spatial pattern recognition software – suitable discrimination between AE and LRU activity.
The ComPair NDT monitoring system has the capability to successfully
• Detect the elastic activity at a distance of 1m (typically the size of surface transport panels)
• Discriminate between LRU and AE activity using pattern recognition software
• Form an LRU/AE sensor-array providing additional & corroborative information without degrading sensor performance
• Corroborated detection of impact excitation up to 800mm from the source

WP4: Development of robotic scanner and NDT integration

The research conducted in WP4 involved the research and development of a multi-axis generic robotic scanner for the purposes of in-service inspection of composite materials. This included the integration of the thermography inspection techniques (researched in WP 2) and the development of user interaction software able to control the robot and visualize the test inspection to the user.

The robotic scanner was developed by Cereteth. The high level requirements were defined as:
• Targeting of “in service” applications (rather than “in production”)
• Automatic inspection of composite panels
• Ability to scan samples of unknown geometry
• Flexibility of the scanning head during inspection
• Easy transfer of equipment for transient and NIR thermography techniques
• Inspection of flat and curved (in one plane) samples
• Adjustability of distance with sample during inspection
• Easy mobility of the robotic scanner

The robotic scanner features five subsystems:
1. The main base
− Supports the structure of the robotic scanner and offers stability during operation. It comprises tubular links (40mm x 40mm x 2mm) of carbon steel welded together to form a rigid mesh and supports the scanner weight. Connection plates are used to place the XYZ robotic system on the base to provide height adjustability for fine tuning during operation. Wheels provide mobility. During inspection, the scanner is based on four support pads for stability. A handle is provided for easy manoeuvrability of the scanner into the desired location.

2. The XYZ robotic system, (Figure 14)
− This system provides 4 DoF:
• Travel range in x-direction: 1.2m
• Travel range in y-direction: 1.0m
• Travel range in z-direction: 1.1m
• Rotational φ: -300 - +300
• Maximum speed: 150mm/s
• Maximum acceleration: 300mm/s2
• Encoding resolution: 4.4µm
• AC servo motor with 400Watts power for each stage driven by three 400 Watt AC servo motors using a 5:1 gearbox

The robotic system provides reduced moments on the carriages of the stages and is rigid enough to prevent oscillations during operation.

3. The smart head
− The smart head supports the inspection equipment for both inspection techniques (cameras and light sources for transient and near infrared thermography) and provides the 4th rotational degree of freedom for inspection of curved samples (up to an angle of ±30°). The rotational degree of freedom is driven by a 20 Watt DC motor integrated with a 83:1 gearhead and a 500 counts per revolution incremental encoder.

The robotic scanner's head carries two LED light sources - placed on the left and right of the camera - to prevent uneven illuminations. This configuration (suitable for both NIR and thermal imaging) was chosen in order that the Reflectance mode may be used since it is more realistic for in-service applications for a number of reasons:
− Access to both sides of the inspection specimen is not available in most cases (or space limitations may not permit synchronization of the camera and light source’s movement)
− Composite parts may be combined with layers of non-transparent materials, so NIR light will not be able to pass through the sample.

4. The electrical and electronics subsystem
The electrical and electronics subsystem comprises: the electric cabinet which houses the power supply; the sensors used for initialization, measuring and safety purposes; the hardware interfaces used for communication between the control pc and the motor drives; and the cable management.

5. The robotic scanner's control software
The control software provides low and higher level functionalities. The control software routines are: Initialization of the system; Operation of the scanner; Error handling; and Profile acquisition procedure

The research completed the integration of the thermography NDT system and the NIR system with the robotic scanner. For both cases an initialization procedure is run where the scanner moves to zero position and its axis are being set to zero value. The laser distance sensor on the head of the robotic scanner is used to check the parallelism of the robotic scanner with relation to the sample.

The operator may alternate between the equipment for transient thermography and NIR imaging, (Figure 15, Figure 16). The position on which the one of the two cameras is placed is the same, but the NIR camera is based on a connection plate with posts.
The provision of a Human Machine Interface (HMI) by KCC was developed and successfully integrated with the robotic scanner. In addition to system initialization & shut down; connection to scanner network (and scanner software) and toggling light sources; the ComPair software inspection system provides seven tab pages in the ComPair software inspection system:

• Panel Profile
Define panel surface profile;
• Define System
Select inspection system. Currently the software supports Basler Camera and Flir Camera
• Robot Arm Position
Define the spatial position of the Robot Arm: camera position to the panel surface can be identified, and the inspection position can be generated.
• Standard
Define the acceptance criteria.
• Inspection
Control the Robot to a particular position and take a picture, and then move to the next position until whole panel is sampled, Figure 17.
• Image Stitch
Stitch the inspect images to form a whole image, Figure 18.
• Sentence
Find the defects in the image

The software accommodates image sizes from 100 x 120 x 1240 x 1680 pixels. The maximum array is dependent upon hardware capability but a standard premium laptop with 4GB of RAM is capable of stitching an image array of 20 x 20 images.

WP5: Demonstration and Field Trials

The first integrated laboratory trials for the ComPair Inspection System were performed on 14th-18th March 2011. This involved the validation of the robotic scanner with both thermal imaging NDT approaches for maintenance inspection including: Cereteth’s involvement regarding development of a multi-axis scanner; implementation of transient thermography by NTUA and NIR thermography techniques by G-Tronix. The sample specifications upon which the scanner will be tested were agreed by Hexcel together with input from TWI, KTU and Envirocoustics. The samples provided a varied template upon which the thermography and AE/LRU techniques were verified.

This aim of the trials in Athens was to achieve (Figure 19):-

• Integration with Robot Arm
• Integration with FLIR camera
• Integration with NIR
• Inspection of GF (NIR)
• Inspection of CF (FLIR)
• Capture video

The demonstration of the compare in-situ Inspection System was carried out at NTUA’s premises in Greece on the 17th June 2011.

The final actions completed in time for this demonstration were:
• FLIR camera and software development kit (SDK) attained
• Integration with FLIR camera software
• Installation of the ComPair inspection software in NTUA’s computer to communicate with camera.
• Necessary software errors addressed
• Provision of both CFRP and GFRP curved samples in Athens

Problems encountered at the demonstration were:
• Software was re-installed in a Windows 7 computer. When connected to the NIR camera, it crashed again

The subsequent troubleshooting involved:
• KCC changed the Visual Studio 2010 C# compiler to the Visual Studio 2008 C# compiler. Thus software can work on most Windows systems.

Final outcome of the ComPair in-situ Inspection System demonstration was summarised as follows:
• End to End inspection process demonstrated (Figure 20)
• Visualisation by HMI software of image capture and image stitching (100 x 120 x 1240 x 1680 pixels) (Figure 21)
• Successful sentencing by HMI software with visual representation of defect detection within GFRP and CFRP samples
• Video captured for dissemination purposes

The demonstration of the ComPair in-service Monitoring System (contributing towards Task 5.3) was carried out at TWI’s premises (UK) on the 9-11th March 2011:

The aim of the trials in Cambridge was to:
• Test a GFRP sample to destruction to detect onset of loading damage using and integrated array of AE/LRU Guided wave sensors, Figure 22.

The final actions completed in time for this demonstration were:
• Manufacture of 800mmx70mmx4mm GFRP coupon samples
• ‘Wasting’ of the coupons (halfway along length) to increase likelihood of fracture at sample centre
• Set up of tensile loading machinery at TWI
• Adjustment of pulser/receiver software to permit automated excitation of LRU sensors every 1 µs

Outcome of ComPair Inspection System demonstration:
• Growing defects over a distance of 600mm
• Intrusive-damage down to a thickness of 2mm
• AE monitoring can be performed along with GW measurements without interference of GW to AE performance.
• AE can detect and localize damage during health monitoring. It can also discriminate the GW signals and locate the pulsing GW sensors, Figure 23
• LRU can record the presence and characteristics of the load present, Figure 23
• AE & Guided wave sensors can form a sensor-array providing additional & corroborative information without degrading sensor performance
• Identify the presence of loads down to 15kN
The demo was very successful with satisfaction from each of the consortium partners at the level of development achieved. Each of the technology aims were accomplished and it was concluded (by the consortium) that a viable system for in-situ inspection and in-service monitoring of CFRP and GFRP composite panels - to aid the surface transport sector for the purposes of reducing operational and maintenance costs - could be achieved.

WP6: Training – Validation - Certification

The work conducted in WP 6 produced procedures and specific guidelines for NDT of composite materials for surface transport applications. There were 2 guidelines produced during this project, a guideline for the use of the Compare NDT inspection technique and robot and a guideline for the Compare Monitoring system

The guidelines for the ComPair in-situ Inspection System contain information to operate the user software, hardware including NIR Basler Camera and transient thermography FLIR Camera, and Robot Arm. This includes:
• Moving the scanner
• Power on/off
• Communication software and installation
• Placement of inspection equipment
• Initialisation and operation
Use of ComPair software and data analysis including Panel Profile; System Definition; Robot Arm Positioning; Definition of acceptance criteria; Inspection; Image Stitching and Sentencing.
A video was produced to visually demonstrate to the user how:
• to power on/off the robotic scanner
• to manoeuvre the scanner
• to transfer the NDT technique from on thermal imaging camera to another
• to initiate the robotic scanner
• to initiate the ComPair interactive software
• select appropriate test method and acceptance criteria
• commence inspection testing
The guidelines for the ComPair Monitoring System contains information on how to use the ComPair technology to inspect and monitor composite panels. This includes:
• Ascertaining optimum sensor distribution for defect detection
• Software configuration of the pulser/receiver unit to establishing sensor excitation and reception. Including coupling criteria; cabling; communications to a PC; excitation signal set-up and configuration; suitable application of windowing functions
• Post-processing such as averaging, dispersion curves, time-of-arrival and wave-mode selection
• Set up of sensors and preamplifiers
• AE system acquisition settings
• System calibration via Hsu Neilson sources and velocity measurements
• Sensor set-up and attenuation measurements
• Application and inspection of AE pattern recognition software
The Guidelines and certification document is easy to implement and permits any operator to set-up and configure both the ComPair Monitoring System and ComPair Inspection System to inspect a CFRP or GFRP panel and achieve the results previously documented within this report (and accompanying Deliverable reports).

WP7: Dissemination and Exploitation

The process of promoting the research conducted within the ComPair project was initiated with a Plan for the Use and Dissemination of Knowledge (PUDK). The PUDK was compiled to define the project strategy and decision-making on allocating resources to exploit the experimental results and knowledge, and to promote the final ComPair product to the transport market.
The PUDK report identified three different potentially interested market areas:
1. Public: comprising key institutions such as environmental agencies, ministries, and legislation bodies; City Users Groups such as municipalities and local administrations; and end users such as road, rail and bus transport companies
2. Maintenance professional and/or NDT professional: including NDT System developers, s/w providers and robotic developers, health monitoring developers, and professional categories such as composite experts, R&D support experts, SMEs, engineers, consultants and researchers.
3. Manufacturing professional: including OEM’s (truck, Bus, Rolling stock and Composites manufacturers)

In addition, it was identified that Large Enterprises (public or private) will find the partial tools (NDT systems for manufacturing and assembly stages / real time monitoring of composites for maintenance) and the integrated automated scanning system very useful for the creation of specifications.

To exploit the marketability of the project, a SWOT analysis was produced to help identify the internal and external factors that may be favourable (or otherwise) for the promotion of the final ComPair product as a business venture - including details of the scientific knowledge developed within the project term. These details included:
• Regulation and technical direction for opportunities concerning the ComPair project
• Innovative elements of the ComPair project to better appeal to the customers within the composite field and make maintenance and operation easier thus contributing to cost reduction
• Surface transport market and the requirement to produce efficient, robust, lightweight vehicles that are environmentally friendly and provide free movement of people and transport of goods in Europe at local and regional levels
• Efficiency ‘drives’ to lower operational costs through automation processes
• Industry technology state-of-the-art (NDT techniques in composite industries), ComPair can develop European procedures for different NDT approaches
• Industry sector analysis: Aerospace, transport, energy
• NDT techniques and appropriate application with a view to composite panel inspection on surface transport applications
• Market analysis and different infrastructure capacities: including rail network, high-way road network and populations of UK, France, Italy, Lithuania, Finland and Greece
• Identification of the potential and qualitative market distribution of the six participating countries of the ComPair consortium
• Identification of the competition to the ComPair consortium
• Increasing public awareness on safety of composite constructed and maintained transport.
• Regulatory issues and conditions
• Marketing of project results: details on participant agreement (Who? Why? What?); marketing and exploitation; patentable qualitative and /or quantitative procedures foe assessment of various surface transport composites.

To initiate and maintain the momentum required to promote the ComPair project, a project website was launched and regularly updated. Currently, a video of the ComPair in-situ Inspection System is available for view, showing the robotic scanner actively inspecting a curved composite panel.
There has been regular attendance at conferences by ComPair partners throughout Europe for the duration of the project. Published papers are available on the project website.
The ComPair consortium had a stand at the JEC Show Paris 2011. Several communication tools were prepared for the promotion of the ComPair project to illustrate the technical advances to date. The promotional tools included:
• looped video footage on a large monitor of the scanning operation on a 1mx1.3m curved CFRP panel;
• a roll-up banner presenting the compare technical objectives;
• a roll-up banner presenting the ComPair consortium;
• a poster illustrating results of the in-situ inspection software;
• a poster illustrating results of the NDT inspection techniques;
• a description of the ComPair consortium - as exhibitor - in the official buyer’s show guide
• Technical documents, brochures and flyers.

The stand was attended by several partners on a rota basis to provide a constant presence of partners with marketing knowledge and technical knowledge. In this way, the consortium were consistently able to field questions, promote the technology and record details of all interested parties. Promotion of the then upcoming workshop in Rome (June 2011) was also conducted.
The ComPair stand generated a considerable amount of interest. The feedback forms (made available for the show) were filled by many delegates with contact details and their sector of interest. ENEA and Atoutveille subsequently compiled a report discussing the level of interest in the developed technology which concluded the following:
● There was tangible interest for NDT techniques: there were few stands at the JEC show concerning NDT topics (a distinct advantage for the ComPair stand), and this may suggest that there are market opportunities for new entrants on the NDT market for composites

● A series of relevant and detailed questions concerning different NDT techniques (thermography, LRU/AE) were asked. There was also interest concerning the potential of NDE combining Guided waves with AE for materials other than just composites

● Considerable interest for the ComPair scanning and control software: some companies appeared readily willing to test the robot on real parts:
− a French company which produces electrical insulators (10 m, Ω 1m)
− a Canadian company who have developed a prototype bus made with a composite body
− a Taiwanese company who produce composite bicycle frames
− Companies (including some in aeronautics or transportation) are also interested by the ability of the thermography technique to detect defects within the material, and not only on the surface

● A new French technical centre dedicated to composites and robotics (located in Basque country) is considering the opportunity to host the robot for a period of time to organize testing sessions with local industrial companies.
− Managers of public agencies from Bordeaux and West area of France have also shown interest in the robot

● More than 30 people agreed to give their contact information to receive the program of the workshop that was subsequently organized in Rome in June 2011

In the last six months of the project two workshops were held. The first workshop, Rome (hosted by ENEA in June 2011), attracted a largely continental European audience. Representatives from a number of high profile companies were in attendance including Lamborghini S.p.A Dantec Dynamics, Italian Society for Non Destructuve Testing Monitoring Diagnostics, Fondazione LaboratorioProve Materie Plastiche, Puglia Aerospace District. A ComPair brochure detailing the project technical achievements - and their benefits to the surface transport industry - was created specifically for this event and was distributed to attendees as part of a delegates’ package.

A news bulletin of the project at this workshop is available on the ComPair website.

A second workshop was held in Cambridge at the end of the project (hosted by TWI, November 2011). Although this workshop was less well attended, I did elicit some interesting contacts – one of which was the SMMT automobile association. As a result, the ComPair consortium have written and submitted a journal article for the SMMT January newsletter. Elsewhere, a separate article written by Atoutveille, has been submitted for inclusion in the January edition of the JEC magazine.

Potential Impact:
The ComPair project has been very successful. The field trials carried out on the purchased curved CFRP and GFRP panels confirmed that the in-situ NDT inspection development is an effective and reliable inspection device that has the potential to be of considerable interest to the surface transport sector. It is a unique system that can perform quick and simple inspection of composite components. In addition, the in-service NDT monitoring technique provides an effective system of defect detection. The ComPair project exhibits a level of flexibility that enables the two techniques to compliment one another, or equally be used in their own right.

The consortium had identified the need for such a system for utilisation in what is an increasingly demanding surface transport market. The development of the project has been widely available since project inception on the ComPair website (www.compairproject.com/) - as launched by TWI - for the facilitation of project dissemination. A Google search of the phrase ‘ComPair project’ provides details of the project as the first item (fifth item when searching the word ‘compair'). The website was maintained by the project coordinator with easy facility for updates by consortium partners. It remains the principal internet platform where information, contacts, news and developments about the ComPair project system were accessible to the public for the duration for the project - and will remain so beyond project completion. In addition, all partners were asked to add a link of the ComPair project website onto their respective websites. Whilst conducted by all partners, ENEA together with Atoutveille, have provided the momentum behind exploitation and dissemination activities.

The investment cost for such a system is considered to be high in the current status. This is mainly due to the robotic scanner, however; through a process of planning and suitable training (which report D6.1 provides), this cost can be reduced. At the final stage of the ComPair project, the price of the prototype was discussed but not defined. It was agreed that a drive to lower the costs should be taken seriously without the threat to system capability and integrity.

In terms of benefits, this project will benefit the consortium partners by promoting their expertise and technology, and deliver to the surface transport market a system to reduce long-term operational and maintenance costs. In addition, the process of enhancing the composite component life-cycle will promote confidence for their inclusion in transport design, increase the market share of composite manufacturers, and reduce CO2 emissions thereby meeting the directives set by national governments and supranational administrations.

There is currently no standard condition monitoring technique available that is used to routinely provide autonomous inspection of surface transport structural integrity. At present, the industry deploys hands-on-action as deemed necessary. The aim is to confine repairs to an ‘as and when required’ scenario: a call for inspection - by the ComPair NDT monitoring system - resulting with the ComPair system delivering a ‘positive sentence’. This will prevent costly and serious repairs that may require a period out-of-service by identifying defects at an incipient stage and halting defect advancement (system capability outlined in WP2, WP3 & WP5 of this document).

The benefits of applying an SHM system such as ComPair will become apparent to the transport industry sector when a way to calculate the point at which an NDT inspection system becomes cost-effective from the perspective of the owner-operator. There are a variety of reasons for the delay in standard implementation such a system. These include the less-than-simple geometry of some composite panels, and the non-standard composite manufacture contributing towards industry vehicles. This is where a monitoring system like that developed by the ComPair consortium may contribute. An additional strength of the ComPair project (as previously mentioned) is the combination of a monitoring system that exhibits NDT ‘duality’ (AE combined with LRU) to provide validation and considerably reduce the risk of false-positives. In short, the ComPair NDT inspection system and NDT monitoring system provide a cohesive and complimentary system of NDT assessment that has, to date, generated substantial interest within the industry sector.

The workshops in Rome and Cambridge helped the ComPair consortium establish contact with Lamborghini - an early adopter of composite materials in cars – who currently manufacture the Monocoque LB834 of the Lamborghini Aventador entirely from CFRP. The trend is now to use CFRP for the automotive mass-market sector. In 2010, BMW announced that the BMW i3, a new revolutionary electric automobile, scheduled for launch in 2013, will be the world’s first mass-produced vehicle with a passenger cell made from carbon fibre. This trend is also observed in the HGV and bus industry. For example, in South-Korea, cooperation between Hankuk Fiber Co. Ltd and Hyundai Heavy Industry Co. has resulted in the development of a non-polluting electric bus featuring composite bodywork and composite interior parts. Truck trailers made entirely from composite materials (Acrosoma) are designed specifically for heavy dynamic loads and are based on a unique and fully patented technology

Other related sectors where the ComPair consortium has established contact include: the Italian Association of NDT & Monitoring; leading Italian composites magazine, Assocompositi; Aerospace Technology District of Puglia; and the Society of Motor Manufacturers (SMMT). As a result of these contacts (and others), the consortium is currently looking for potential partnership with industrial surface transport manufacturers in order to continue this research and validate project efforts that may lead to future commercialization. The magazine articles that will feature in January’s edition of the JEC magazine and SMMT newsletter are but two examples where continued efforts are pursuing this end. As the project ends, it is apparent that the SMEs within the ComPair consortium are particularly well-placed to exploit this opportunity.

The ComPair consortium recognised and agreed the importance of understanding the strength, weakness of the developed technologies and to identify the opportunities and threats for the end results as a realization process. The SWOT (Strengths / Weaknesses / Opportunities / Threat) analysis carried out in the first reporting period was the first step of strategic planning. It helped all partners become aware of the different market aspects and guide the project to focus on key issues. Figure 24 shows the SWOT analysis carried out by ComPair project partners.

Threats for the system are not overtly critical and depend upon the way the information resulting from the ComPair project will be disseminated and exploited. However, additional promotional effort regarding financial benefits may have to be made to mitigate the effects of the current economic climate within Europe. In a similar vein, the price of the ComPair system has to be addressed to make it a viable option in the transport industry sector.

The SWOT analysis shows that there is no shortage of opportunities that will benefit the industry from employment contribution, reduced CO2 emissions, standardised procedures, and increased competition. The strengths of the ComPair project are the contribution to transport technology both in manufacturing and in-service terms; long term financial benefits on the operational economics; and the applicability of the project research outcomes to other industry sectors (such as aerospace and renewable energy). The most significant weakness of the project has been the absence of an end-user within the consortium to help inform partners. However, the level of dissemination and information gathering performed by ENEA and Atoutveille has helped neutralise the impact of such a weakness through their constant dissemination of their research unto the technical partners within the consortium.

A number of dissemination activities were performed during the course of the project to expose the research results and make developments visible to the industry. Conferences and seminars related to composite materials and transport were attended by the project partners and presentations were given.

Flyers and brochures were the main marketing tools for the project and were made available at conferences, workshops and exhibitions by the project partners. Dissemination activities after completion of the project were considered critical for the successful implementation and commercialisation of the product.