SEVENTH FRAMEWORK PROGRAMME THEME SST.2010.5.2.1 ... · Automated and cost effective railway infrastructure maintenance Automated and Cost Effective Maintenance for Railway Contract:

ACEM-Rail −−−− 265954 D2.1 Inspection and Monitoring

[2011/05/31] Page 1 of 35

SEVENTH FRAMEWORK PROGRAMME

THEME SST.2010.5.2.1. Automated and cost effective railway infrastructure maintenance

Automated and Cost Effective Maintenance for Railway

Contract: 265954

D2.1 Report on railway inspection and monitoring techniques – Analysis of different approaches

Deliverable number D2.1

Work Package number WP2: Technologies and processes for the analysis of infrastructure condition

Task Task 2.1

Revision 0

Due date 2011/05/31 Submission date 2011/05/31

Distribution Security PU Deliverable type R

Authors A. Barragan, P. Cembrero, N. Caceres, F. Schubert

Partners CEMOSA, US, FRAUNHOFER

Verification F. G. Benitez (US)

Approval (coord.) N. Jiménez-Redondo (CEMOSA)


[2011/05/31] Page 2 of 35


[2011/05/31] Page 3 of 35

Table of Contents

1 INTRODUCTION......................................................................................................................6

2 TRACK MEASUREMENTS....................................................................................................7

2.1. MEASUREMENTS AND EVALUATION OF TRACK DEFECTS ............................................................ 7

2.1.1 Superstructure................................................................................................................................................ 7 2.1.2. Substructure .................................................................................................................................................. 8

2.2. MEASUREMENTS AND EVALUATION OF DEFECTS IN TRACK GEOMETRY .................................. 22

2.3. NON-DESTRUCTIVE EVALUATION METHODS FOR RAILW AY COMPONENTS .............................. 22

2.3.1. Ultrasonic inspection .............................................................................................................................. 22 2.3.2. Acoustic Inspection Techniques............................................................................................................. 22 2.3.3. Electromagnetic Inspection.................................................................................................................... 22 2.3.4. Thermographic inspection ..................................................................................................................... 22 2.3.5. Radiographic inspection......................................................................................................................... 23 2.3.6. Inspection using visual cameras............................................................................................................. 23 2.3.7. Distributed Optical Fibres ..................................................................................................................... 23

2.4. APPLICATION MATRIX .................................................................................................................. 23

ANNEX A −−−− GLOSSARY OF NON-DESTRUCTIVE TESTING TECHNIQUES .... ..............27

ANNEX B −−−− ACRONYMS..............................................................................................................31

REFERENCES AND BIBLIOGRAPHY.......................................................................................32


[2011/05/31] Page 4 of 35

List of Figures Figure 1. Sliding micrometer installation for settlement measurement in embankment .....................9 Figure 2. Sliding micrometres equipment............................................................................................9 Figure 3. Settlement cells performance for settlement measurement embankment. .........................10 Figure 4. Rod extesometer mounted scheme .....................................................................................11 Figure 5. Displacement transducer.....................................................................................................11 Figure 6. Displacement transducer mounted in head of rod extensometers. .....................................11 Figure 7. Vibrating wire piezometers mounted scheme....................................................................13 Figure 8. Vibrating wire piezometers sensor .....................................................................................13 Figure 9. Vibrating wire piezometers multi level sensor ...................................................................13 Figure 10. Pulley to assist depth control ............................................................................................14 Figure 11. Calculation lateral movement...........................................................................................14 Figure 12. Inclinometer equipment....................................................................................................14 Figure 13. Crank or joint meter 3D mounted scheme........................................................................15 Figure 14. Insulating flexible backing strain gauge...........................................................................16 Figure 15. Weldable vibrating wired strain gauge.............................................................................16 Figure 16. Embedment vibrating wire strain gauge...........................................................................17 Figure 17. Shear Load cell .................................................................................................................17 Figure 18. Tension load cell...............................................................................................................18 Figure 19. Center hole load cell .........................................................................................................18 Figure 20. Digital Tap extensometer..................................................................................................18 Figure 21. Tilt meter and tilt plate .....................................................................................................19 Figure 22. High precision laser sensor until 150 meters....................................................................20 Figure 23. Convergence measurement equipment with extensometer and tiltmeter. ........................21 Figure 24. Convergence measurement equipment with tilt meter. ....................................................21


[2011/05/31] Page 5 of 35

List of Tables Table 1. Defects and instrumentation ...............................................................................................26


[2011/05/31] Page 6 of 35

1 Introduction

Inspection and monitoring techniques represent one of the key issues of the project. Only reliable data from these techniques will lead to an accurate input for the maintenance planning and thus, to an efficient and economic maintenance process. In order to estimate the ability of the different techniques for the identification of the relevant kinds of defects it is important to describe the latter in detail. This is done in section 2.5 of deliverable D1.1 (State of practice of railway infrastructure maintenance). In section 2 the state-of-the-art of the general track measurement concept, the measurement of the track geometry, and the substructure measurement are described. This section reference in some section to section 3 and 5 of deliverable D1.1. This section ends with an application matrix giving a quick overview of the ability of a specific inspection technique for finding a special kind of defect. It is planned to update this matrix regularly during execution of the project in order to take new developments and experience from the field measurements into account. Therefore, in this sense, this deliverable will be an alive document.


[2011/05/31] Page 7 of 35

2. Track Measurements

This section focuses on actual measurement techniques to evaluate the track condition. If measures are within the threshold limits, then train operation is safe and complies with an adequate level of comfort. Otherwise, maintenance tasks should be carried out. This section describes:

• Techniques used to evaluate defects in the different elements of track infrastructure within the scope of the ACEM-Rail project.

• Techniques used to analyse the track geometry. • Techniques used to evaluate elements included in the substructure, such as, cutting,

embankment, bridges or tunnels This section describes (or reference) measurement and inspection techniques widely used in the railway sector and new inspection techniques that are still under development and need to be further optimized in order to be embarked into commercial or test trains.

2.1. Measurements and evaluation of track defects This section describes several measurements techniques used to evaluate track defects. Measurements techniques are presented following the same structure than section 2 in deliverable D1.1. First, those techniques used in the superstructure are described and then those appropriate for substructure inspections. 2.1.1 Superstructure

2.1.1.1. Track In the paragraph below, a short description of measurement techniques are included. All of them were described with detail in section 3 of deliverable D1.1 i) Rail Measurements techniques widely used to evaluate rail profile are: Ultrasonic and Laser. ii) Sleepers The inspection of sleepers is currently a human driven operation. iii) Fastenings Fastenings can be inspected by visual cameras. The main goal for using images of the track is to eliminate, or reduce as much as possible, the visual inspection carried out by workers walking along the track to detect faults, missing components, etc. iv) Switches and crossing Switches are the most complicated and delicate component of the track. Inspections in switches and crossings have been purely manual until now because no instrument able to deliver the large number of measurements required. The situation is now changing, thanks to improvements in cameras, lasers, embedded computers and inertial packs.


[2011/05/31] Page 8 of 35

2.1.1.2. Track bed The track bed inspection is nowadays partially automatic with artificial vision systems which can measure the profile of the ballast curb and surrounding track area. More details are included in section 3 of deliverable D1.1. 2.1.2. Substructure 2.1.2.1. Subgrade

Subgrade layers are on earth. Usually it is necessary either to dig in the earth (cutting) or to provide earth to adapt the natural profile to the track line profile as it was explain in section 2.4.2.4 of deliverable 1.1. These soil works cause stress in inner soil layers and settlements which may affect to track geometry.

The subgrade inspection is currently a human operation. Maintenance in the subgrade is so expensive that it is only performed in case of failure. In this case, a geotechnical study of failure is required.

In new constructions, geotechnical instrumentation is placed in certain areas of the subgrade (like large cutting or embankments). Below some systems used for the inspection and monitoring of the subgrade are described.

2.1.2.1.1. Sliding micrometers.

The Sliding micrometre is used to its full advantage in cases where the complete distribution of strains and axial displacements along a straight line has to be recorded with great accuracy as in embankments, concrete dams, tunnels, piles and diaphragm walls.

The sliding micrometre is a tool with high-precision, in the measuring line, axial displacement measurements for boreholes and measuring lines in rock, concrete or soil in any arbitrary direction.

The measuring principle is Linewise Displacement Measurement.

Linewise measurement of displacement vectors along measuring lines gives information on the behaviour of a monitored site in rock and soil or of the structure as well as of the interaction between the structure and the ground. With the Sliding Micrometer and the Sliding Deformeter, displacement and deformation profiles can be measured very accurately in several geotechnical structures.

The mobile and modular measuring system consists of probe, cable, rod, readout unit, data-processing unit and calibration device (Figure 1). The modular structure of the system allows an optimal combination of all components.

The probe uses the ball-and-cone positioning principle in the measuring marks of the measuring tube, and thus with high precision sensors and regular calibration before and after each series of measurements, a very high accuracy of measurement and long-term stability is achieved.

The ball-and-cone positioning principle is based on the spherically shaped heads of the probe and the circular cone shaped measuring marks which ensure a precise positioning of the one metre long probe during measurement.


[2011/05/31] Page 9 of 35

The Sliding Micrometer measures the displacements ∆z along the axes of the measuring lines that can have any arbitrary direction. The accuracy of measurement of the Sliding Micrometer is better than ± 0.002 mm/m.

Figure 1. Sliding micrometer installation for settlement measurement in embankment

Figure 2. Sliding micrometres equipment

2.1.2.1.2. Settlements cells or settlements gauges

A settlement cell is designed to monitor settlements in embankments, fills, and foundation soil. It is required in construction areas which are inaccessible to standard optical survey techniques. It is especially useful in measuring large changes in settlement under earth dams, landfills, and soft soils.


[2011/05/31] Page 10 of 35

aIt provides a single point measurement of settlement or heave.

The settlement cell consists of three components: a liquid filled tube, a pressure transducer, and a reservoir of liquid. One end of the tubing is connected to the pressure transducer, which is embedded in the soil. The other end of the tubing is connected to the reservoir, which is located at a higher elevation on stable ground, away from construction activity.

The transducer measures the pressure created by the column of liquid in the tubing. The height of the column is equal to the difference in elevation between the transducer and the reservoir. As the transducer settles with the surrounding soil, the height of the column increases and the transducer measures a higher pressure.

Settlement is calculated by converting changes in pressure to millimetres or inches of liquid head.

Figure 3 illustrates the performance of settlement cells.

Figure 3. Settlement cells performance for settlement measurement embankment.

2.1.2.1.3. Rod extensometers

Rod extensometer are installed in boreholes to monitor settlements in foundations, subsidence above tunnels, displacements of retaining structures, and deformations in underground openings.

The main components of a rod extensometer are anchors, rods inside protective pipe, and a reference head.

The anchors are installed downhole with rods attached. The rods span the distance from the downhole anchors to the reference head at the surface. The protective plastic pipe prevents bonding between rods and grout backfill.

Readings are obtained at the reference head by measuring the distance between the top (near end) of the rod and a reference surface. A change in this distance indicates that movement has occurred.

Movements are referenced to a stable elevation, typically a downhole anchor. The resulting data can be used to determine the zone, rate, and acceleration of movements, and to calculate strain.


[2011/05/31] Page 11 of 35

In one borehole, several extensometer could be installed, as soon the figure, in order to determinate the settlement in different layers of soil. Displacement transducers are installed in head of the extensometer as the last step of the installation process. One transducer is used for each rod. The transducer screws into the end of the rod, so that it can monitor both extension and compression. The sensor is then connected to a readout and adjusted to its initial position.

Figure 5. Displacement transducer

Figure 4. Rod extesometer mounted scheme

Figure 6. Displacement transducer mounted in head of rod extensometers.


[2011/05/31] Page 12 of 35

2.1.2.1.4. Vibrating wire piezometers

Vibrating wire piezometer is sealed in boreholes and embedded in fills to measure pore-water pressures. It can also be placed in standpipes and wells to measure water levels. Typical applications include:

� Monitoring dewatering schemes for excavations and underground openings.

� Monitoring ground improvement techniques such as vertical drains, sand drains, and dynamic compaction.

� Monitoring pore pressures to determine safe rates of fill or excavation.

� Investigating the stability of natural and cut slopes.

� Monitoring the performance of earthfill dams and embankments.

� Monitoring seepage and ground water movement in embankments, land fill dikes, and dams.

� Monitoring water levels in wells, standpipes, lakes, reservoirs, and rivers.

Vibrating wire piezometer converts water pressure to a frequency signal via a diaphragm and a tensioned steel wire. The piezometer is designed so that a change in pressure on the diaphragm causes a change in tension of the wire.

When excited by a magnetic coil, the wire vibrates at its natural frequency. The vibration of the wire in the proximity of the magnetic coil generates a frequency signal that is transmitted to the readout device. The readout device processes the signal and displays a reading.

Calibration factors, which establish a relationship between pressure applied to the diaphragm and the frequency signal returned to the readout device, are used to convert Hz readings to engineering units.


[2011/05/31] Page 13 of 35

Figure 8. Vibrating wire piezometers sensor

Figure 7. Vibrating wire piezometers mounted scheme

Figure 9. Vibrating wire piezometers multi level sensor

2.1.2.1.5. Inclinometer

Inclinometers are used to monitor lateral movements in embankments and landslide areas, deflections of retaining walls and piles, and deformations of excavation walls, tunnels, and shafts. Inclinometer is typically installed in a near vertical borehole that passes through suspected zones of movement into stable ground.

During a survey, the probe is drawn upwards from the bottom of the casing to the top, halted in its travel at half-meter intervals for tilt measurements. The first survey establishes the initial profile of the casing. Subsequent surveys reveal changes in the profile if ground movement occurs.

The inclination of the probe body is measured by two force-balanced servo-accelerometers. One accelerometer measures tilt in the plane of the inclinometer wheels, which track the longitudinal grooves of the casing. The other accelerometer measures tilt in the plane perpendicular to the wheels. Inclination measurements are converted to lateral deviations.

Changes in lateral deviation, determined by comparing data from current and initial surveys, indicate ground movements. Plotting the cumulative changes at each measurement interval yields a high resolution displacement profile. Displacement profiles are useful for determining the magnitude, depth, direction, and rate of ground movement.


[2011/05/31] Page 14 of 35

Figure 10. Pulley to assist depth control

Figure 11. Calculation lateral movement

Figure 12. Inclinometer equipment

2.1.2.2. Structures: bridges

The structure inspection is mainly a human driven operation. In many countries an annual basic inspection and 5-year periodic main inspection for bridges and tunnels is mandatory. There are different inspection techniques for specific structures. The type of structure and their characteristic defines the inspection techniques and the corresponding analysis.

In this analysis may be necessary an instrumentation to determine the condition of the infrastructure.

In the market exist several technical to monitories this measure as it will be describe in this section.

Below several instruments suitable for the inspection of engineering structure are presented.


[2011/05/31] Page 15 of 35

2.1.2.2.1. Crank and joint meters

The crank or joint meter is used to monitor movement at joints and cracks. The crank and joint meter is suitable for applications such as:

� Monitoring movement at sub-merged construction joints in concrete-face dams.

� Monitoring joints or cracks in tunnels and tanks.

The jointmeter consists in displacement transducers, which in 1D, 2D or 3D mounting system, determine the movement between two parts.

The groutable anchors are installed on opposite sides of the joint. The mounting system is secured to the anchors, and the displacement transducers are positioned to provide the required range. Signal cables from the sensors are routed to a readout station.

An indicator or data logger is used to read the transducers. The initial reading serves as a reference datum. Subsequent readings are compared to the initial to calculate the magnitude and rate of changes.

Figure 13. Crank or joint meter 3D mounted scheme

2.1.2.2.2. Strain gauges

A strain gauge is a device used to measure the strain of an object. The most common type of strain gauge consists of an insulating flexible backing which supports a metallic foil pattern. The gauge is attached to the object by a suitable adhesive. As the object is deformed, the foil is deformed, causing its electrical resistance to change. This resistance change, usually measured using a Wheatstone bridge, is related to the strain by the quantity known as the gauge factor.

Another kind of strains gauges is the vibration wire strain gauge. It is a steel wire held in tension inside a tube. The tube is mounted on a steel flange or embedment in concrete. Strain in the structural member is transferred through the flange to the tube and the wire inside. An increase in tensile strain increases tension in the wire, and a decrease in tensile strain decreases tension in the wire.


[2011/05/31] Page 16 of 35

A sensor that is placed atop the wire is used to pluck the wire, causing it to vibrate at a frequency relative to its tension. The vibration of the wire within the magnetic field of the coil induces a frequency signal which is transmitted to the readout device.

The readout device processes the frequency signal using calibration factors that relate frequency to strain in the wire, and then displays a number which can be a frequency, period or microstrain.

There are different kinds of strain gauges used in civil engineering.

For stress measurements in steel, weldable strain gauges measure are used. Typical applications include:

� Monitoring stresses in structural members of buildings, bridges, tunnel linings and supports during and after construction.

� Monitoring the performance of wall anchors and other post-tensioned support systems.

� Monitoring loads in strutting systems for deep excavations.

� Measuring strain in tunnel linings and supports.

� Monitoring areas of concentrated stress in pipelines.

� Monitoring distribution of load in pile tests.

For stress measurements in concrete: Embedment strain gauges measures are used. Typical applications include:

� Measuring strains in reinforced concrete and mass concrete.

� Measuring curing strains.

� Monitoring for changes in load.

� Measuring strain in tunnel linings and supports.

Figure 15. Weldable vibrating wired strain gauge

Figure 14. Insulating flexible backing strain gauge.


[2011/05/31] Page 17 of 35

Figure 16. Embedment vibrating wire strain gauge

2.1.2.2.3. Load cells

A load cell is a transducer that is used to convert a force into electrical signal. Through a mechanical arrangement, the force being sensed deforms a strain gauge. The strain gauge measures the deformation (strain) as an electrical signal, because the strain changes the effective electrical resistance of the wire.

A load cell usually consists of four strain gauges in a Wheatstone bridge configuration. Load cells of one strain gauge (quarter bridge) or two strain gauges (half bridge) are also available. The electrical signal output is typically in the order of a few millivolts and requires amplification by an instrumentation amplifier before it can be used. The output of the transducer is plugged into an algorithm to calculate the force applied to the transducer.

Although strain gauge load cells are the most common, there are other types of load cells as well. In industrial applications, hydraulic (or hydrostatic) is probably the second most common, and these are utilized to eliminate some problems with strain gauge load cell devices.

In civil infrastructure a typical load cells is center-hole load cells designed to measure loads in tiebacks, rock bolts, and cables. Applications for these load cells include:

� Proof testing and long-term performance monitoring of tiebacks, rock bolts, and other anchor systems.

� Monitoring loading of vertical supports in underground openings.

The load-bearing element of the load cell is a spool of heat-treated steel alloy. Four or more strain gauge rosettes are bonded to the spool. Each rosette consists of two strain gauges, one oriented to measure axial strain; the other oriented to measure tangential strain. The rosettes are spaced evenly around the periphery of the spool and are wired together to provide a single output. The strain gauge rosettes are protected from moisture and impact damage by a strong aluminium housing filled with a high-density resin.

For best results, the load cell is centered on the bar and bearing plates are placed above and below the cell. Bearing plates must be able to dis-tribute the load without bending or yielding.

Figure 17. Shear Load cell


[2011/05/31] Page 18 of 35

Figure 18. Tension load cell

Figure 19. Center hole load cell

2.1.2.2.4. Digital extensometer The digital extensometer is usually a tap extensometer with a digital measurement. It is used to detect and monitor changes in the distance between two reference points. Typical applications include:

� Monitoring displacement of retaining structures, bridge supports, and other structures. � Monitoring convergence of tunnel walls. � Monitoring deformations in underground openings.

Reference points are permanently installed at measurement stations along the tunnel or structure. To obtain a measurement, the operator stretches the tape between two reference points, hooking the free end of the tape to one point and the instrument body to the other. The operator tensions the tape by turning a knurled collar until two index marks are aligned and then notes the reading from the tape and the digital display. The sum of these readings is the distance between the two reference points. By comparing the current reading to the initial reading, the operator can calculate the change in distance between the two reference points.

Figure 20. Digital Tap extensometer.

2.1.2.2.5. Tiltmeter

Tiltmeter is used to monitor changes in the inclination of a structure. Tiltmeter data can provide an accurate history of movement of a structure and early warning of potential structural damage. Typical applications include:


[2011/05/31] Page 19 of 35

•Monitoring rotation caused by mining, tunneling, soil compaction, or excavation.

•Monitoring rotation of concrete dams and retaining walls.

The tiltmeter system has a number of tilt plates, the portable tiltmeter, and a readout unit. The portable tiltmeter uses a force-balanced servo-accelerometer to measure inclination. The accelerometer is housed in a rugged frame with machined surfaces that facilitate accurate positioning on the tilt plate. The bottom surface is used with horizontally-mounted tilt plates and the side surfaces are used with vertically-mounted tilt plates.

Tilt plates are mounted on the structure in specified locations. They are typically bonded to the structure, but may also be screwed to the surface.

To obtain tilt readings, the operator connects the tiltmeter to the readout unit, positions the tiltmeter on the tilt plate, and notes the displayed reading. The operator then rotates the tiltmeter 180 degrees and obtains a second reading. Later, the two readings are averaged to cancel sensor offset. Changes in tilt are found by comparing the current reading to the initial reading.

Figure 21. Tilt meter and tilt plate

2.1.2.2.6. Laser sensor

There are a lot of instrument in the market for displacement measure, but in civil engineering the dust environment and other climate condition affect to technologies application.

Laser sensor provide high accurate in large measurement, which it is implemented quickly.

Some traditional measurements with extensometer are being replaced by laser measurement because laser has enough precision in large dimensions (from meter to two hundred meters), for fixed external installation this technology are limited by climate condition or dust environment.


[2011/05/31] Page 20 of 35

Figure 22. High precision laser sensor until 150 meters

2.1.2.3. Tunnels

The most common technique to measure the tunnels clearance is a rotating range finder. It actually measures a helical curve. This technique was included in section 3 of deliverable D1.1.

Other techniques could be applied for the geothechnical evaluation of tunnels.

Below several techniques to monitor tunnels are described.

2.1.2.2.7. Convergence

The convergence measurement is the most important measurement in tunnels. Several equipment can be used to measurement the convergence. Some of this equipment is:

� Laser displacement sensor.

� Topography equipment as total station.

� Digital tap extensometer.

� Vibrating wire extensometer.

� Tilt sensor.

Reference points are permanently installed at measurement stations along the tunnel. Depending on the characteristics of the tunnels and the type of available sensors, different systems can be used to monitor the convergence of the tunnel.

Figure 23 shows an example of the convergence measurement with extensometer and tilmeter.


[2011/05/31] Page 21 of 35

Figure 23. Convergence measurement equipment with extensometer and tiltmeter.

Usually extensometer is combined with a tiltmeter is used in order to provide a profile of tunnel and its evolution in the time. This allow convergence measurement.

Example of application with tiltmeter can be seen in the below figure.

Figure 24. Convergence measurement equipment with tilt meter.

Figure 24 illustrate the procedure to measure the convergence of the tunnel. The distance between the reference points is computed with the help of a virtual triangle. The vertices of this virtual triangles are: the two reference points and the joint between the long and short arm. There is one tilt meter to measure the tilt of the long arm and another one to measure the tilt of the short arm. If the two reference points are displaced, the shape of the triangle changes. Taking into account the measures of the two tilt meters, by simple geometric computation, the convergence of the tunnel is derive


[2011/05/31] Page 22 of 35

2.2. Measurements and evaluation of defects in track geometry Track geometry defects are the source for high dynamic forces exchanged between the train and the rails. Measurement systems for track geometry were described, with detail in deliverable D1.1. Only a short review will be included in this section.

2.3. Non-destructive evaluation methods for railway components Evaluation methods for NDT were described, with detail in deliverable D1.1. Only a short review will include in this sections.

2.3.1. Ultrasonic inspection

Ultrasonic techniques belong to the most commonly used NDE methods with a wide variety of application fields. They are mostly used in pulse-echo mode so that only one-sided access to the structure under investigation is necessary Non-contact ultrasonics One of the main problems in ultrasonic testing in general is the need for coupling agents in order to ensure a sufficiently high energy transfer and signal-to-noise ratio. Especially for high-speed application, for instance in rail track inspection, these coupling agents lead to several problems like supply problems and speed reduction. For this reason several non-contact techniques are also available in principle.

2.3.2. Acoustic Inspection Techniques

While ultrasonic techniques typically use the system response to active excitations, acoustic methods only “listen” to natural sound sources like the rolling noise. If hollow shafts are available (e.g. in some high-speed trains) hollow shaft integrated acoustic sensor systems can be used to detect defects in wheel sets of the rolling stock

2.3.3. Electromagnetic Inspection

Beside ultrasonic inspection techniques a wide variety of electromagnetic techniques is available. Below some of them with a high relevance for railway inspection are shortly described.

2.3.3.1. Inspection using pulsed eddy currents Eddy current measurements have been a standard technique for a long time for finding cracks in metals either on the surface or within the material

2.3.3.2. Alternating Current Field Measurements (ACFM) The Alternating Current Field Measurement (ACFM) is now widely accepted as an alternative to magnetic particle inspection [e.g. Papaelias et al, 2010]. The ACFM technique is capable of both detecting and sizing surface breaking cracks in metals based on the skin effect. It can be applied for the detection of near-surface defects.

2.3.3.3. Ground Penetrating Radar (GPR) Ground Penetrating Radar is a geophysical method that uses radar pulses in exactly the same way than ultrasonic pulses are used to image subsurface.

2.3.4. Thermographic inspection

Infrared thermography also belongs to the well-known non-contact NDT techniques.


[2011/05/31] Page 23 of 35

2.3.5. Radiographic inspection

X-ray applications in the railway sector are very rare which is mainly caused by the large dimensions of the mostly metallic samples (low penetration depth) and the necessary safety regulations for ionising radiation.

2.3.6. Inspection using visual cameras

The main goal for using images of the track is to eliminate, or reduce as much as possible, the visual inspection done by workers walking along the track to detect any fault, missing components, etc.

2.3.7. Distributed Optical Fibres

Distributed optical fibre sensors, capable of temperature and strain measurements, are based on stimulated Brillouin scattering and rely on the interaction between two lightwaves and an acoustic wave in the optical fibre As the acoustic velocity depends on temperature and strain, it is possible to map out the distribution of temperature or strain by monitoring changes in the Brillouin frequency shift along the fiber length.

2.4. Application matrix The following table summarizes the applicability of the discussed inspection and monitoring techniques to find a specific kind of defect. The green colour indicates that the defect can be identified by the corresponding technique. The red colour shows that an identification of the defect is not possible. If it is still unclear if a specific technique is able to find the defect or not, the yellow colour is used instead. In some case additional remarks are also given. It is planned to update this matrix regularly during execution of the project in order to take new developments and experience from the field measurements into account.


[2011/05/31] Page 25 of 35

Inspection technique Type of defect

Visual inspection

Laser

Optical cameras

Ultrasonics (active)

Acoustic inspection

(passive, e.g. hollow-shaft)

Thermography Eddy current

Alternating Current

field (ACFM)

Radar (GPR)

Acceleration sensors

Distributed optical fibres

Damage in rails: Longitudinal vertical crack (UIC 113)

only surface-

near

Tache ovale (UIC 211) expected Horizontal crack (UIC 212) Running surface disintegration (UIC 221)

Short-pitch corrugation (UIC 2201)

Long-pitch corrugation (UIC 2202)

Lateral wear (UIC 2203) Running surface shelling (UIC 2221)

Gauge-corner shelling (UIC 2222)

Head checks Bruising (UIC 301) Faulty machining (UIC 302)

Transverse and horizontal cracks in weldings (UIC 411, 412, 421, 422, 431, 432)

only surface-near

only surface-

near

only surface-

near

Rail wear (Track profile deformation)

Damage of sleepers (timber, steel, concrete)


[2011/05/31] Page 26 of 35

Inspection technique Type of defect

Visual inspection

Laser

Optical cameras

Ultrasonics (active)

Acoustic inspection

(passive, e.g. hollow-shaft)

Thermography Eddy current

Alternating Current

field (ACFM)

Radar (GPR)

Acceleration sensors

Distributed optical fibres

Defects in fastenings (broken bolts, gaps, breakage)

Longitudinal geometry track defect (LD)

Transverse geometry defect (TD)

Horizontal geometry defect (HD)

Gauge deviations Track twist Deterioration of ballast Deterioration of subballast Change of subgrade (vegetation, weed)

Damage of structures: bridge degradation (foundations, piers, decks)

Several NDT

methods

Not from train

Not from train

Defects in retaining walls Damage of tunnels (geotechnical balance and humidity)

Damage of cuttings and embankment

Damage of drainage ditches

Table 1. Defects and instrumentation


[2011/05/31] Page 27 of 35

ANNEX A −−−− Glossary of non-destructive testing techniques

Nondestructive testing (NDT), also known as Nondestructive inspection (NDI) and Nondestructive evaluation (NDE) are techniques that can be classified into various methods of nondestructive testing, each based on a specific scientific principle. These methods are deployed into various techniques. Not all techniques have proved to be valid for all industrial applications. Along this document some terminology related to non-destructive techniques (NDT) for testing items and systems of the railway infrastructure appear. The purpose of this appendix is to provide a brief description of the NDT, which can also be easily found in many texts and database (Moore and Udpa, 2008).

• Ultrasonic techniques: This group of techniques is usually based on elastic waves in the ultrasonic frequency range.

� Acoustic emission testing (AE or AT). Acoustical analysis can be done on a sonic or ultrasonic level. Ultrasonic techniques make possible to predict deterioration earlier than conventional techniques. AEs are commonly defined as transient elastic waves within a material caused by the release of localized stress energy. Hence, an event source is the phenomenon which releases elastic energy into the material, which then propagates as an elastic wave. Acoustic emissions can be detected in frequency ranges under 1 kHz, and have been reported at frequencies up to 100 MHz. Rapid stress-releasing events generate a spectrum of stress waves starting at 0 Hz and typically falling off at several MHz. Sonic monitoring equipment is less expensive, but it also has fewer uses than ultrasonic technologies. Sonic technology is useful only on mechanical equipment, while ultrasonic equipment can detect electrical problems and is more flexible and reliable in detecting mechanical problems.

o Ultrasonic testing (UT), very short ultrasonic pulse-waves with center frequencies ranging from 0.1-15 MHz and occasionally up to 50 MHz are launched into materials to detect internal flaws or to characterize materials. UT is typically used in pulse-echo or pulse transmission mode. Conventional UT usually needs a fluid coupling agent to launch enough energy into and detect strong echoes out of the specimen. Non-contact versions of UT are based on air-coupled ultrasound and Electro Magnetic Acoustic Transducers (EMATs).

o Laser ultrasonics (LUT) is another non-contact technique where both excitation and detection is based on laser systems. The generation lasers are short pulse (from tens of nanoseconds to femtoseconds) and high peak power lasers. The physical principle is thermal expansion or ablation. In the thermoelastic regime the ultrasound is generated by the sudden thermal expansion due to the heating of a tiny surface (in case of metals) or in a vloume of the material (e.g. fluid media) by the laser pulse. If the laser power is sufficient to heat the surface above the material boiling point, some material is evaporated (typically some nanometres) and ultrasound is


[2011/05/31] Page 28 of 35

generated by the recoil effect of the expanding material evaporated. In the ablation regime, a plasma is often formed above the material surface and its expansion can make a substantial contribution to the ultrasonic generation. Consequently the emissivity patterns and modal content are different for the two different mechanisms. The ultrasonic echoes from inside the specimen are usually detected by laser Doppler vibrometers or other interferometers.

o Time of flight diffraction ultrasonics (TOFD) is a very sensitive and accurate method for nondestructive testing of welds for crack-like defects. TOFD is a computerised system that was invented in the UK in the 1970s for the nuclear industry by Dr. Maurice Silk. It usually uses the crack-tip diffraction and mode converted echoes to determine location, depth and size of a crack. The use of TOFD enables crack sizes to be measured more accurately, so that expensive components could be kept in operation as long as possible with minimal risk of failure.

o Guided Wave testing (GWT) employs mechanical stress waves – usually in the ultrasonic range - that propagate along an elongated structure while guided by its boundaries. This allows the waves to travel a long distance with little loss in energy. This technique is also known as Guided Wave Ultrasonic Testing (GWUT) or Long Range Ultrasonic Testing (LRUT), though it is very different to conventional ultrasonic techniques due to the dispersive nature of the guided modes for instance..

• Electromagnetic testing (ET). In this group of techniques electric currents or electromagnetic fields or both are induced inside a test object and the electromagnetic response is observed. If the test is set up properly, a defect inside the test object creates a measurable response.

o Alternating current field measurement (ACFM) is an electromagnetic technique for detection and sizing of surface breaking cracks. It works on all metals and, not requiring direct, electrical contact, works through coatings.

o Eddy-current testing (ECT). It uses electromagnetic induction to detect flaws in conductive materials. However, the surface of the material must be accessible. It has some disadvantages as the finish of the material affects the accuracy, the depth of penetration into the material is limited, and flaws that lie parallel to the probe may be undetectable. The principle is a circular coil carrying currents placed in proximity to the test specimen (electrically conductive). The alternating current in the coil generates a changing magnetic field which interacts with the test specimen and generates eddy currents. Variations in the phase and magnitude of these eddy currents can be monitored using a second 'search' coil, or by measuring changes to the current flowing in the primary 'excitation' coil. Variations in the electrical conductivity or magnetic permeability of the test object, or the presence of any flaw, will cause a change in the eddy currents and a corresponding change in the phase and amplitude of the measured current.

o Pulsed eddy-current. In this technique the classical sinusoidal current mode is substituted by a transit signal such as a pulsed current. In fact, the pulsed character of the electric excitation allows a high peak value of the density of


[2011/05/31] Page 29 of 35

eddy currents in the material; also the broadband signal contains optimized frequencies to probe the sample over a more extended depth.

• Infrared and thermal testing (IR). Infrared monitoring and analysis has the widest range of application (from high- to low-speed equipment), and it can be effective for spotting both mechanical and electrical failures; some consider it to currently be the most cost-effective technology. IR is a technique that can detect internal voids, delaminations, and cracks in concrete structures such as bridge, decks, and other civil constructions. The infrared thermographic scanning system can measure and view temperature patterns based upon temperature differences as small as a few hundreds of a degree Celsius. It may be performed during day or night, depending on environmental conditions and the desired results. All objects emit electromagnetic radiation of a wavelength dependent on the object’s temperature. The frequency of the radiation is inversely proportional to the temperature. In infrared thermography, the radiation is detected and measured with infrared imagers (radiometers). The imagers contain an infrared detector that converts the emitting radiation into electrical signals that are displayed on a color or black & white computer display monitor. A typical application for regularly available IR Thermographic equipment is looking for “hot spots”. There are many other terms widely used all referring to infrared thermography: a) Video Thermography and Thermal Imaging draw attention to the fact that a sequence of images is acquired and is possible to see it as a movie; b) Pulse-Echo Thermography and Thermal Wave Imaging are adopted to emphasize the wave nature of heat; c) Pulsed Video Thermography; d) Transient Thermography; e) Flash Thermography. For the excitation of the thermal field different techniques are available e.g. flash lights, infrared lamps, inductive excitation or ultrasound sonotrodes.

• Laser testing

o A Profilometer is a measuring instrument used to measure a surface's profile, in order to quantify its roughness. Vertical resolution is usually in the nanometre range, though lateral resolution is usually poorer. An optical profilometer is a non-contact method for providing much of the same information as a stylus based profilometer. There are many different techniques which are currently being employed, such as laser triangulation, and confocal microscopy. The Fiber-based optical profilometer version scans surfaces with optical probes which send light interference signals back to the profilometer detector via an optical fiber. Fiber-based probes can be physically located hundreds of meters away from the detector enclosure, without signal degradation. Profilometry is a technique used in road pavement monitoring (AASHO Road Test) for a long time. It uses a distance measuring laser (suspended approximately 30 cm from the pavement) in combination with an odometer and an inertial unit (normally an accelerometer to detect vehicle movement in the vertical plane) that establishes a moving reference plane to which the laser distances are integrated. The inertial compensation makes the profile data more or less independent of the speed the profilometer vehicle during the measurements, with the assumption that the vehicle does not make large speed variations and the speed is kept above 25 km/h or 15 mph. The profilometer system collects data at normal highway speeds, sampling the surface elevations at intervals of


[2011/05/31] Page 30 of 35

2–15 cm (1–6 in), and requires a high speed data acquisition system capable of obtaining measurements in the kilohertz range. Many road profilers are also measuring the pavements cross slope, curvature, longitudinal gradient and rutting. Some profilers take digital photos or videos while profiling the road. Most profilers also record the position, using GPS technology.

• Vibration analysis (VA) is one of the most productive dynamic methods but can also be the most expensive component of a PdM program. Vibration analysis, when properly done, allows the user to evaluate the condition of equipment and avoid failures. The latest generation of vibration analyzers comprises more capabilities and automated functions than its predecessors. Many units display the full vibration spectrum of three axes simultaneously, providing a snapshot of what is going on with a particular machine or structure. But despite such capabilities, not even the most sophisticated equipment successfully predicts developing problems unless the operator understands and applies the basics of vibration analysis.

• In-situ Visual inspection (IVI) and Remote Visual inspection (RVI). The IVI is a common method of quality control, data acquisition, and data analysis. Visual inspection, used in maintenance of facilities, means inspection of equipment and structures using either or all of human senses such as vision, hearing, touch and smell. Visual inspection typically means inspection using raw human senses and/or any non-specialized inspection equipment. Inspections requiring ultrasound, X-ray equipment, infra-red, etc are not typically considered as visual inspection as these inspection methodologies require specialized equipment and training. The RVI refers to a special branch of visual inspection where the operator does not enter the inspection area and takes help from visual remote cameras, such as video fiberscopes, Pan-Tilt-Zoom cameras and others.


[2011/05/31] Page 31 of 35

ANNEX B −−−− Acronyms

AC Alternating Current ACFM Alternating Current Field Measurement

AE or AT Acoustic Emission Testing AI Artificial Intelligence

ANN Artificial Neural Network BOTDA Brillouin Optical Time Domain Analysis

CBR Case Based Reasoning CPU Central Processing Unit

ECT Eddy-Current Testing EMATs Electromagnetic Acoustic Transducers

ERA European Rail Agency ES Expert Systems

ET Electromagnetic Testing FPGA Field Programmable Gate Array

GPR Ground Penetrating Radar GWT Guided Wave Testing

GWUT Guided Wave Ultrasonic Testing HD Horizontal Defect

ICE Intercityexpress IR Infrared And Thermal Testing

IVT In-Situ Visual Inspection LD Longitudinal Defect

LRUT Long Range Ultrasonic Testing LUT Laser Ultrasonics

MFL Magnetic Flux Leakage ML Machine Learning

NDE Non-Destructive Evaluation NDT Non-Destructive Technique

NMR Nuclear Magnetic Resonance RCF Rolling Contact Fatigue/Rail Contact Fatigue

RVI Remote Visual Inspection SBS Stimulated Brillouin Scattering

SME Small To Medium-Sized Enterprises TD Transverse Defect

TOFD Time Of Flight Diffraction Ultrasonics UT Ultrasonic Testing

VA Vibration Analysis


[2011/05/31] Page 32 of 35

References and Bibliography

Aamodt, A., Plaza, E. (1994). Case-Based Reasoning: Foundational Issues, Methodological Variations, and System Approaches. Artificial Intelligence Communications 7, no. 1, 1994, pp. 39-59. Aharoni, R., and E. Glikman, (2002) “A novel high-speed rail inspection system”, ECNDT 2002. AII (2011). Acquired Intelligence Inc. http://www.aiinc.ca/ (last accessed on February 2011). Al-Qadi, I., J. Rudy, J. Boyle, E. Tutumluer, (2006) “Railroad Ballast Fouling Detection Using Ground Penetrating Radar - A New Approach Based on Scattering from Voids”, ECNDT 2006. Bernini, R., M. Fraldi, A. Minardo, V. Minutolo, F. Carannante, L. Nunziante and L. Zeni, (2006) “Identification of defects and strain error estimation for bending steel beams using time domain Brillouin distributed optical fiber sensors”, Smart Materials and Structures, 15 612–622 (2006) Bernini, R. A. Minardo, L. Zeni, (2007) “Accurate high-resolution fiber-optic distributed strain measurements for structural health monitoring”, Sensors & Actuators: A. Physical, 134, 389-395 (2007) Bernini, R., A. Minardo, L. Zeni, (2008) “Vectorial dislocation monitoring of pipelines by use of Brillouin-based fiber-optics sensors”, Smart Materials and Structures, 17 15006-15014 (2008) Blouin, A., D. Lévesque, S. Kruger, J. Monchalin, (2009) “Laser ultrasonics for NDE in the automotive industry”, NDT in Canada 2009 National Conference, Aug 25-27, 2009 London, Ontario, Canada. Bouye, J., (2010) “The Innovation and Modernization of Radiography in the Nuclear Industry”, NDT.net 2010. CORDIS (2011). Community Research and Development Information Service. Information on all EU-supported R&D activities. http://cordis.europa.eu/ (last accessed on January 2011). DataEngine (2011). Software Tools for Intelligent Data Analysis. http://www.dataengine.de/english/sp/index.htm (last accessed on February 2011). Dey, A., Thomas, H.M., and Pohl, R. (2008) The important role of eddy current testing in railway track maintenance, 17th WCNDT, Shanghai, China Elballouti, A., S. Belattar, N. Laaidi, (2010) “Thermal Non-Destructive Characterization of the Spherical Defects in the Roadways”, ECNDT 2010.


[2011/05/31] Page 33 of 35

Esveld, C. (2001). Modern Railway Track, 2nd edition, Delft University of Technology, MRT Productions. Ferreira, L. and M. Murray (1997), Modelling Rail Track Deterioration and Maintenance Current Practices and Future Needs, Transport Reviews, 17, 207-221. Frankenstein, B.; Hentschel, D.; Schubert, F.; Pridöhl, E , (2005) “Hollow Shaft Integrated Monitoring System for Railroad Wheels”, In: Advanced sensor technologies for nondestructive evaluation and structural health monitoring: 8 - 10 March 2005, San Diego, California, USA, Bellingham/Wash.: SPIE, 2005 (SPIE Proceedings Series 5770), pp.46-55, 2005. Grimberg, R., A. Savin, R. Steigmann, C. Comisu, (2010) “GPR – Electromagnetic Sensor Array Data Fusion”, ECNDT 2010. Heckel, T., H. Thomas, M. Kreutzbruck, S. Rühe, (2009) “High Speed Non-Destructive Rail Testing with Advanced Ultrasound And Eddy-Current Testing Techniques”, Indian National Seminar & Exhibition on Non-Destructive Evaluation NDE 2009, December 10-12, 2009. Hokstad, P., H. Langseth, B. Lindqvist, and J. Vatn (2005), Failure Modeling and Maintenance Optimisation for a Railway Line, International Journal of Performability Engineering., 1, 51-64. INNOTRACK (2008). Rail inspection technologies. Report prepared for EU funded project INNOTRACK, Project no. TIP5-CT-2006-031415. INNOTRACK (2008b) The state of the art of the simulation of vehicle track interaction as a method for determining track degradation rates – Part One – Strategic Models. INNOTRACK deliverable D1.3.2. Iqbal, Z. (2010) A Study of AI techniques for Railheads, Vegetation, Switches & Crossings. Master Thesis. Reg.No. E38440. Högskolan Dalarna. Jarmulak, J., Kerckhoffs, Eugene J.H., vant’t Veen, P., (2001). Case-based Reasoning for Interpretation of Data from Nondestructive Testing. Engineering Applications of Artificial Intelligence 14, 2001, pp. 401 -417. Jemec, V. and J. Grum, (2010) “Automated Non-Destructive Testing and Measurement Systems for Rails”, ECNDT 2010. Jovanovic, S., Esveld, C., (2001): ECOTRACK: An objective condition-based Decision Support System for long-term track M&R Planning directed towards reduction of Life Cycle Costs. 7th International Heavy Haul Conference, Brisbane, Australia. Larsson, D. (2004). A study of the track degradation process related to changes in railway traffic. PhD Thesis. Lulea University of Technology. Sweden.


[2011/05/31] Page 34 of 35

Martın, O., Lopez, M., Martın, F. (2007). Artificial neural networks for quality control by ultrasonic testing in resistance spot welding. J. Mater. Process. Technol. 183, 226–233. Moore, P.O.(ed); Udpa S. S. (technical ed) (2008). Nondestructive Testing Handbook. 3rd edn. American Society of Nondestructive Testing. Negnevitsky, M. (2005). Artificial intelligence: A guide to intelligent systems, Second Edition, London: Addison-Wesley. Nunziante, L., M. Fraldi, M.C. Pernice, A. Gesualdo, L. Zeni and K.M. Mahmoud, (2010) “Distributed optical fibre sensor measurements on rods and bridge cable wires – Part II: Experimental”, Bridge Structures 6, 49–63, DOI:10.3233/BRS-2010-002 (2010) Papaelias, M. Ph.; Roberts, C.; Davis, C. (2008). A review on nondestructive evaluation of rails: state-of the art and future development. Proc. IMechE. Vol. 222. Part F J. Rail and Rapid Transit 2008. P.367-384 Papaelias, M. Ph.; Davis, C.; Roberts, C.; Blakeley, B.; Lugg, M. (2010). INTERAIL: Development of a Novel Integrated Inspection System for the Accurate Evaluation of the Structural Integrity of Rail Tracks – Implementation of the ACFM Rail Inspection Module. 10th European conference and exhibition on non-destructive testing (10thECNDT). Moscow 2010. Peng, C., L. Wang, X. Gao, Z. Wang, Q. Zhao, Y. Zhang, J. Peng, K. Yang, (2010) Dynamic Wheel Set Defect Inspection by Applying Ultrasonic Technology”, ECNDT 2010. Peter, J. (1999) Introduction to expert systems, Third edition, Harlow, England: Addison Wesley Longman. Picarelli, L., L. Zeni, (2009) Discussion on “Test on application of distributed fibre optic sensing technique into soil slope monitoring” by B. J. Wang, K. Li, B. Shi and G. Q. Wei”, Landslides, 6:361–363 DOI 10.1007/s10346-009-0169-0 (2009) Podder, T. (2010) Analysis & study of AI techniques for automatic condition monitoring of railway track infrastructure. Master Thesis. Reg. No: E3845D. Högskolan Dalarna. Podofillini, L., E. Zio, and J. Vatn (2006), Risk-informed Optimisation of Railway Inspection and Maintenance Procedures, Realibility Engineering and System Safety, 91, 20-35. Rockstroh, B., W. Kappes, F. Walte, (2008) Ultrasonic and eddy current inspection of rail wheels and wheel set axles”, WCNDT 2008. Salzburger, H., L. Wang, X. Gao, (2008) In-motion ultrasonic testing of the tread of high-speed railway wheels using the inspection system AUROPA III”, Proceedings of WCNDT 2008.


[2011/05/31] Page 35 of 35

Sambath, S., Nagaraj, P., Selvakumar, N. (2011). Automatic Defect Classification in Ultrasonic NDT Using Artificial Intelligence. Journal of Nondestructive Evaluation, Volume 30, Number 1, pp. 20-28. Sawadisavi, S., J. Edwards, E. Resendiz, J.M. Hart, C.P.L Barkan, Ahuja, N. (2009) Machine-Vision Inspection of Railroad Track. Proceedings of the TRB 88th Annual Meeting, Washington, DC, January 2009. Shapiro, S. C. (ed.) (1991). Encyclopedia of Artificial Intelligence, 2nd Ed. New York: John Wiley. Snyder, W. E., Qi, H. (2010). Machine Vision. Cambridge University Press. Štarman, S., and V. Matz, (2008) “Automatic System for Crack Detection Using Infrared Thermographic Testing”, WCNDT 2008. StatSoft (2011). StatSoft - Software for Statistics, Data Mining, Predictive Analytics, Credit Scoring. http://www.statsoft.com/ (last accessed on February 2011). Willems, H., B. Jaskolla, T. Sickinger, O. Barbian, F. Niese, (2010) “Advanced Possibilities for Corrosion Inspection of Gas Pipelines Using EMAT-Technology”, ECNDT 2010. XpertRule (2011). XpertRule Software Ltd. http://www.xpertrule.com/ (last accessed on February 2011). Yella, S., Dougherty, M. S., Gupta, N. K. (2006). Artificial intelligence techniques for the automatic interpretation of data from non-destructive testing, Insight Journal of the British Institute of Non-Destructive Testing, vol 48, no 1, 2006. Zhou , Z., H. Zhang, D. Wie, (2010) “Application of Pulse Compression Technique in Air-Coupled Ultrasonic Non-Destructive Testing on Composite Material”, ECNDT 2010.

Documents

SEVENTH FRAMEWORK PROGRAMME THEME SST.2010.5.2.1 ... · Automated and cost effective railway infrastructure maintenance Automated and Cost Effective Maintenance for Railway Contract: