1Above and Below: ILI based Pipe

Strain Assessment

Dr. Sebastian Hühn1, Daniel Bahrenburg2, Dr. Dietbert Wortelen1, Jörg

Grillenberger3, Thomas Beuker3
1
ROSEN Technology & Research Center, 2 ROSEN USA, 3 ROSEN Germany

Organized by

Proceedings of the 2023 Pipeline Technology Conference (ISSN 2510-6716).

www.pipeline-conference.com/conferences

Copyright ©2023 by EITEP Institute.

 Pipeline Technology Conference 2023, Berlin

1 ABSTRACT
Geological processes and weather events continuously reshape the surface of our planet. The
occurrence and extent of such events is not always predictable and easy to determine. Over the last
two decades the intensity of cataclysmic and routine weather events are increasing. As most pipelines
are buried, geological and sedimentary changes can impose additional strain to pipeline materials
ultimately resulting in pipeline geohazards. To guarantee safe pipeline operations these hazards need
to be considered during construction and the ongoing integrity management of existing assets.

Accounting for the influence of onshore or subsea land movement on structures is the primary goal of
any effective geohazard management program. Traditionally, geotechnical monitoring programs are
based on distributed point measurements in a defined area of ground movement. Active monitoring
of ground movement and pipeline stress states is typically performed with discrete monitoring
equipment, such as slope inclinometers or strain gauges. This can provide sufficient information for
management of simple pipeline geometries located within basic geological environments and, most
importantly, where hazards have already been identified. Frequently, hazards that result in pipeline
failures are unmonitored or in areas that do not appear to be problematic, particularly where the
surface expression is subtle and not identified by traditional surveillance methods.

To meet the demands placed on operators by advancing regulation and severe weather conditions,
more frequent and comprehensive appraisals of pipeline right-of-way corridors will be required.

This paper will demonstrate an ILI based pipe strain assessment method, which utilizes in-line inertial
mapping and a new developed electromagnetic axial stress measurement sensor to assess the
condition of the pipeline. Inertial mapping reveals the occurrence of discrete flexural loading and its
proximity to performance limits. Additionally, this study will show the ability of the new
electromagnetic based ILI technology to accurately measure stress resulting from axial loading on
pipelines. The integration of these two data collection techniques ultimately results in a high level of
confidence when diagnosing and characterizing active geohazard threats, even before they are
detected above ground.

This approach not only provides the ability to identify pipe strain, but also assists with setting
appropriate inspection intervals to track changes to pipeline integrity and hazard development. An
additional advantage of these complementary technologies is that they provide a robust baseline for
any geotechnical program and increased efficacy in the design of site surveys, the selection of
monitoring points and guidance toward the most appropriate remediation solution. By conducting
repetitive inspections the described ILI technique alone could also replace costly installations of
monitoring devices.

2 INTRODUCTION
Current efforts to monitor the presence and effects of geohazards on pipelines include a variety of
techniques. They can take the form of fixed equipment to monitor ground movement or strain over a
continuous time scale, ground survey or in-line inspection to acquire an instantaneous snapshot in
time, or a combination of many such methods. All are useful and provide valuable insight into the
integrity and stability of a pipeline system. However, it can be challenging for operators to maximize
the effectiveness of the individual platforms and techniques available to them, as well as to justify
utilization of each. This paper proposes a solution that allows for accurate and routine monitoring of
2
 Pipeline Technology Conference 2023, Berlin

geohazards using a combination of inertial measurement and micro-magnetic sensor response. The
benefit of employing this particular combination is a mixture of simplicity and repeatability.

3 GEOHAZARDS
It is standard practice to consider all environmental threats that may affect the pipeline during the
design stage. These include landslides (see Figure 1), geological faults, liquefaction, river crossings,
and both natural and man-made ground subsidence issues. Geological, geomorphological,
geotechnical, river and seismic specialists are normally involved in the identification and
characterization of these threats and the route is adjusted to avoid significant hazards as required.
Where it is not possible to avoid specific threats such as the crossing of active geological faults or
some types of landslide due to other constraints, mitigation is applied to reduce the loading on the
pipeline or to stabilize the hazard.

Figure 1. Pipeline Crossing Head Area of an Active Landslide

Geohazard threats also arise during the service life of the pipeline. Despite detailed studies of the
right-of-way at design, extreme or unexpected events leading to ground movement loads may occur.
In some cases, the hazard is subtle and may have existed prior to construction but was not identified.
In other cases extreme weather events or increasing soil temperature can activate geohazards.

Pipeline displacement due to ground loading will typically develop a combination of bending and
axial forces that can result in high strains and ultimately lead to failure through tensile fracture or
local buckling (Nyman, 2008). Pipelines that cross areas of high geohazard susceptibility can be
designed to withstand the higher load demands that could arise and this normally involves strain-
based design. However, the vast majority of onshore pipelines have not been designed for strain
limits, and this means the strain capacity may be impaired particularly due to uncertainties over the
properties of the girth welds (Andrews, 2018).

3
 Pipeline Technology Conference 2023, Berlin

A process is required to manage geohazards during the service life of the pipeline and this should
involve the main elements of identification, evaluation and mitigation (ISO, 2019). The use of
technologies that identify the presence and nature of threats at the earliest possible stage is clearly
of great interest to operators with assets crossing terrain that is susceptible to geohazards. The
integration of in-line inspection technologies as described in this paper increases the confidence in
early identification, interpretation and management of such threats.

3.1 INERTIAL MEASUREMENT UNITS

An inertial measurement unit (IMU) is an electronic device that uses accelerometers and gyroscopes
to measure linear and rotational accelerations. IMU’s are widely used in the automotive, aerospace
and oil and gas industries, enabling a precise determination of instantaneous position and
orientation in space. A typical IMU is composed of three accelerometers and three gyroscopes, which
are oriented as illustrated in

Figure 2. Each measurement taken by the orthogonally-mounted IMU sensors can be used to
describe the position of a body within a three-axis coordinate system across six degrees of freedom.

Inertial measurement units are now widely integrated into in-line inspection platforms to provide
pipeline operators with line position, anomaly location, and bending strain detection services
(Dotson, 2021). Above ground markers (AGM) are used during inertial data post-processing to aid in
calculation of pipeline displacement and to compensate for noise and drift in the dataset. Once an
IMU dataset is fully processed, measurements are used to calculate the instantaneous Easting,
Northing, elevation, pitch, azimuth, and roll at each measurement increment (CRES, 2017).

Figure 2. Illustration Defining IMU Degrees of Freedom

This is accomplished by making use of the definition of rotational acceleration (𝛼), which is a vector
quantity that represents the rate of change of angular velocity (𝜔) of a body in space. Using this
principle, angular displacement (𝜃) can be derived through integration using the following
relationship:

∬ 𝛼(𝑡)𝑑𝑡 = ∫ 𝜔(𝑡)𝑑𝑡 = 𝜃 (𝑡)

4
 Pipeline Technology Conference 2023, Berlin

An IMU is typically mounted on a smart pig platform and integrated with other technologies such as
circumferential/axial magnetic flux leakage, calliper, or others. The inertial system is located at the
center of the canister such that the coarse center line of the IMU device corresponds with the true
center line of the pipe body in a straight section. As tool position and orientation changes, the IMU
sensors deflect from the straight line and systemic deviations are recorded. The raw data is
processed with odometer data to remove false accelerations, reduce drift, and correlated to a linear
distance system for alignment with pipeline features and anomalies. Once complete, pipeline
geospatial information, along with the heading data and velocity, are derived to aid in additional
assessments (Czyz, 1994).

IMU’s can also be integrated into dedicated platforms, such as the ROSEN RoGeo PD tool
(Bahrenburg D. e., 2022). These standalone in-line inspection solutions allow operators and integrity
vendors to obtain inertial data at a substantially higher frequency using a simple and robust design,
which can be deployed in piggable segments with reduced impact to operations. For purely
dedicated configurations, an odometer wheel is not utilized. Data collection consists only of a time
record and inertial data. The data is linearly aligned and geospatially correlated to a fully navigated
baseline inspection prior to interrogation.

By collecting inertial data using one of the configurations outlined above, a bending strain
assessment can be conducted. A single IMU dataset can be used to conduct a bending strain analysis
and pipeline movement can be detected by performing run-to-run analyses of multiple datasets.
Elevated levels of bending strain are a common indication of ground movement, construction-related
activity or other external forcing acting on the pipeline. Pipeline movement assessments offer the
unique advantage of revealing bending strain changes between inspections and offer the ability to
demonstrate stability in areas where high bending strains have been recorded during an initial
inspection.

By solely reviewing and processing IMU data, it is not possible to accurately quantify the uniform
axial strain present in a segment of pipe due to either axial loading or elongation of the pipeline
caused by lateral displacement and partial or full fixity at the analysis boundaries. In order to
determine the total longitudinal strain due to displacement, both flexural and uniform axial
components must be accounted for. Because of this, it is desirable to employ an additional in-line
inspection technology capable of measuring the change in magnetic susceptibility of the pipe caused
by an applied force, which can be linearly correlated to longitudinal stress using known material
properties throughout the pipeline.

3.2 AXIAL STRESS/STRAIN MEASUREMENT

For many pipeline operators the determination of local loading conditions is becoming more and
more important to guarantee safe operations. Recent case studies have shown the ability of a micro-
magnetic technology sensor-equipped in-line inspection (ILI) tool to accurately measure uniform
longitudinal stress resulting from external loading on pipelines. In addition to determining the
loading conditions, this technology is also able to assess material properties such as yield strength or
toughness of the pipeline material. The new sensor technology has been designated MTS (Micro-
Magnetic Technology Sensor).

Essentially, micro-magnetic sensors consist of a magnetic yoke, an excitation and a receiving coil (see
Figure 3 left). When testing a specimen, a magnetic hysteresis curve is generated (see Figure 3 right),
containing information about the mechanical properties and the loading conditions of this specimen.

5
 Pipeline Technology Conference 2023, Berlin

Conventional micro-magnetic sensor applications are mainly used as handheld devices for static local
measurements and play a reliable role e.g. in quality control of steel plates during manufacturing
(Heutling, et al., 2004).

Figure 3. Sketch of a micro-magnetic sensor (left). The excitation generates a hysteresis loop in the specimen which is
measured as sensor response.

To cope with the challenging conditions during in-line inspections i.e. high tool velocities, low power
consumption and potential sensor lift-off, this sensor technology has been upgraded and adapted in
an iterative development process.

Numerous fundamental investigations of the sensitivity of the MTS technology to the stress applied
to a test specimen have been carried out in recent years. These laboratory scaled investigations
proved its basic applicability to this requirement. Among others, tensile tests were performed on
steel samples from different pipeline materials, where the stress in the specimen was recorded in
parallel with the MTS response. An execution and the result of such an experiment is shown in Figure
4. It can be clearly seen that the MTS response is monotonically dependent on the tensile stress
applied to the sample. In further tests, in which the specimens were loaded beyond the yield
strength, it was demonstrated that the MTS response to these elastic stress components saturates.
However, the plastic deformation of the specimen can still be monitored with the MTS by using
additional measurement components. These first experiments verify that the MTS technology allows
quantification of elastic stresses and strains via the modulus of elasticity of steel and discrimination
of plastic strains in steel.

6
 Pipeline Technology Conference 2023, Berlin

Figure 4. A tensile test with parallel MTS sensor measurement (right picture) with the result of this measurement
(left diagram).

To test the laboratory validated measurement sensitivity of the MTS technology towards the stress
state of a line pipe under the boundary conditions of an inline inspection, test runs were carried out
in operating pipelines. For this purpose, an inline inspection tool was developed ( Figure 5). The inline
inspection tool consists of eight evenly spaced sensors installed around the circumference of the
chassis, which allows the splitting of the longitudinal stress into bending and axial components. It is
designed for standalone inspections, but it can be connected with MFL-A, IMU and/or geometry
inline inspection units. To test the functionality of the tool, pull through tests were carried out
through a pull through section, which was subjected to longitudinal stresses by central deflection.
Within these successful tests stresses below yield were investigated at tool velocities between 0.5
and 5 m/s.

Figure 5. Sketch of 24" ILI Tool equipped with MTS devices (in turquoise).

7
 Pipeline Technology Conference 2023, Berlin

3.3 UNIFORM AXIAL STRAIN

Using known material properties, stress can then be correlated to strain. But this approach has some
limits. Only the elastic portion of stress/strain can be determined because, once the applied loading
has induced stress that approaches the material’s yield strength (YS), the magnetic response
recorded by the tool also approaches saturation. This is the point at which permanent, plastic
deformation of the pipe body and welds begin. As the material continues to deform, the ratio of
stress to strain decreases until the ultimate tensile strength (UTS) is reached. This is why plastic
deformation is considered “stress relieving” to some extent for displacement controlled loads. As the
applied load continues to deform the material beyond the UTS, stress decreases as strain increases
until rupture occurs. An example plot of this process is presented as the stress-strain curve in Figure
6.

Despite this, it is still possible to calculate the total strain within a plastically deformed segment using
a combination of MTS and IMU. Because it is assumed that the uniform axial component of
longitudinal strain will be elastic and smaller than that of the bending component, limiting
quantifiable stress to yield (𝜎𝑙𝑖𝑚 = 𝜎𝑌) is acceptable for most cases. The axial component can be
extracted from the MTS data, the flexural component can be extracted from the IMU data, and both
can be combined to determine total strain. Additional details are provided in the following section.

Figure 6. Example stress-strain plot

3.4 COMBINED LOADING

Superposition of the flexural and axial components of strain allows for a complete definition of the
strain state of a segment of pipe. Total strain can be calculated as

𝜖𝑡𝑜𝑡𝑎𝑙 = 𝜖𝑎 + 𝜖𝑏

8
 Pipeline Technology Conference 2023, Berlin

where 𝜖𝑎 and 𝜖𝑏 are the axial and bending components, respectively. Standard convention states
that tensile strains will be positive and compressive strains will be negative. In general terms, if strain
falls within the linear elastic region, 𝜖𝑡𝑜𝑡𝑎𝑙 can be determined using only the magnetic responses
recorded by the MTS. If plastic deformation has occurred, 𝜖𝑎 can be extracted from the MTS data and
𝜖𝑏 can be extracted from the IMU data. In the elastic region, comparison of IMU and MTS datasets
allows for crosscheck and validation of the results from both technologies.

When superimposing uniform axial and bending components, the total tensile/compressive stresses
will increase/decrease, depending on the direction of the uniform axial strain.

4 CASE STUDIES

4.1 CASE STUDY 1– AXIAL STRESS/STRAIN AT ACTIVE SLOPE

In this paper we want to present two case studies from prototype inspection surveys. The first was
conducted in a 24'' crude oil pipeline. This specific pipeline is particularly suitable for such a test run,
as it has already been monitored for axial and bending stresses since its construction using stationary
equipment. The more than 100 km long inline inspection survey was performed in combination with
an inertial measurement unit (IMU).

The relative stress profile from the pipeline derived from the MTS data showed one section within
the pipeline where significant axial stress was encountered. This section of the pipeline was
identified as a slope known to be subject to ground movement. Strain data from an operating dense
strain gage monitoring system was used to calibrate and validate the MTS stress profile. The MTS
relative axial stress showed a linear dependence on the axial strain values of the strain gage system.
This dependence was to be expected, since a linear relationship is present in the elastic region.
Furthermore, this relationship can be used to calibrate the MTS relative stress profile to the absolute
strain in the elastic range with only two parameters. In addition, the robustness of the calibration can
be checked with a leave-one-out-cross-validation: for this, the linear fit is performed on all but one
measured value, which is used to predict the axial strain value with the help of the calibration
parameters of the linear fit. This is performed iteratively for all measurements. After this procedure
one gets, as shown in Figure 7 on the left, a unity plot where each value was predicted by the
calibration parameters of the remaining values. From this plot it is clear that the deviation of the
prediction of the absolute strain done by the MTS stress measurement is less than 0.01% strain
(marked by dotted lines in Figure 7 left). On the right side of Figure 7, the calibrated raw data of the
MTS measurement is plotted against the log distance for the pipeline section addressed. The increase
and decrease in axial tensile strain along the 800 m slope is clearly seen in the MTS data and
confirmed by the strain gage monitoring data. The smaller spikes in the data can be explained by the
presence of field bends and the impact of local cold work hardening to the MTS measurement.

9
 Pipeline Technology Conference 2023, Berlin

Figure 7. Left diagram: Unity plot of MTS axial strain predictions produced by leave-one-out-cross-validation. The
achieved accuracy is better than 0.01% strain. Right diagram: MTS axial strain data adjusted by calibration (black
line) with reference measurements of the strain gage monitoring system (orange crosses).

In addition to the axial strain state of the pipeline, the MTS technology can also be used to gain
insight into the bending strain state and thus the total longitudinal strain state. Figure 8 shows the
MTS strain data of two locations within the investigated pipeline section, each of which is subject to
vertical bending. While at the first location no significant axial strain occurs (left diagram in Figure 8),
an axial strain acts on the pipeline at the second location (right diagram in Figure 8).

Figure 8. MTS absolute strain data from two locations where vertical bending acts on the pipeline. On the left
diagram without additional loading from axial strain, on the right diagram with axial strain marked by the blue
arrow is shown.

Figure 9 illustrates the observed consistency between IMU and axial stress measurement
technologies. Four plots are included, which show the bending and axial strains, quantified after
post-processing of both datasets. The first and third panel show the vertical and horizontal bending
strains based upon the collected inertial data. The second and fourth plots show the strains
calculated from the collected magnetic data oriented at 0°, 90°, 180°, and 270. Note the reduction in
magnitude of the strain reported by IMU when compared to MTS due to the absence of the axial
component of strain.

10
 Pipeline Technology Conference 2023, Berlin

Horizontal Bending Strain, IMU [%]

0.2

0.1

-0.1

Total Strains at 90° and 270°, MTS [%]

0.2

0.1

Sensor Orientation 90 Degrees Sensor Orientation 270 Degrees

-0.1

Vertical Bending Strain, IMU [%]

0.2

0.1

-0.1

Total Strains at 0° and 180°, MTS [%]

0.2

0.1

0
Sensor Orientation 0 Degrees Sensor Orientation 180 Degrees
-0.1

Figure 9: Strain Area Reflecting Agreement between IMU and MTS Technologies

4.2 CASE STUDY 2 – AXIAL STRESS/STRAIN IN AREA OF MINING SUBSIDENCE

In a further technology validation survey, which was carried out with the MTS ILI tool, the experience
towards the performance of the measurement system from the first presented case study was
confirmed. This survey was performed in Europe in a 28'' pipeline, which runs through an abandoned
mining district with ongoing and monitored mining subsidence. In this case, the sensor calibration for
the absolute strain was obtained by a tensile test made of a pipe material sample, which was
provided by the pipeline operator. The axial strain profile of the pipeline section exposed to mining
subsidence is shown in Figure 10. Over a length of several kilometers within the mining subsidence
area, the strain state of the pipeline is alternating between axial tension and compression. Due to the
active ground subsidence, the pipeline has been monitored by the operator for decades:

11
 Pipeline Technology Conference 2023, Berlin

measurements are carried out every year at different points on the pipeline using a BAM (German
Federal Institute for Material Testing) Stress-Measuring Extensometer. The in situ strain
measurements obtained within periodical pipeline inspection of the last decade are also presented in
Figure 10. There is a very good qualitative and quantitative consistency between the MTS data and
the extensometer data. The deviation of some points are a result of the time difference from the
different data collections: ILI tool run (2022) and extensometer measurements (2012-2021).

Figure 10. An extract of the axial strain profile of the second case study. The MTS data (black) are consistent with
the strain data measured with an extensometer within the last decade (orange crosses).

5 SUMMARY
This paper has presented a combined approach for routine monitoring and rapid detection of
geohazards and pipeline movement using a multifaceted strategy that involves collection of inertial
and magnetic response data. Through a relatively simple process, it is possible to calculate bending
strains using differential attitude information collected by an in-line inspection platform that includes
IMU. By integrating high-resolution non-harmonic micro-magnetic sensors, axial loading can be
detected and uniform axial strains can be accurately quantified. These two technologies provide an
accurate and direct measure of the full strain state of the pipeline, whether it is due to pipeline
construction procedures, anthropogenic influence, or geohazard activity.

Two case studies (in pipelines affected by an active slope and mine subsidence) have shown that the
MTS technology can determine the loading conditions of a pipeline within the linear elastic range
with an accuracy of 0.01% strain and a lateral resolution of 10 mm. Areas on the pipeline with plastic
strain can be identified and characterized as well by adding IMU results. Thus, the MTS technology
will be able to make an important contribution to geohazard management in order to meet the
demands placed on operators by advancing regulation and severe weather conditions (Bahrenburg,
2023).

12
 Pipeline Technology Conference 2023, Berlin

6 REFERENCES

Andrews, R. S. (2018). Assessment of stress based design pipelines experiencing high axial strains. . IPC2018-
78111. Calgary: 12th International Pipeline Conference.
Bahrenburg, D. e. (2022). Catch My Drift? The Big Implication of a Smaller IMU Platform. Pipeline Pigging and
Integrity Management Conference. Houston, Texas.
Bahrenburg, D., Young, A., Edwards, J., Singh, A., & Norman, J. (2023). Above & Below: A Holistic Geohazard
Monitoring Solution. Pipeline Pigging and Integrity Management Conference. Houston, Texas.
Beller, M., Eiken, T., & Thale, W. (2016). Integrity Assessment: Addressing the Issue of Available and Missing
Data from an In-Line Inspection Perspective. 3R International.
CRES. (2017). Management of Ground Movement Hazards for Pipelines. Retrieved from CRES-2012-M03-01:
http://www.cres-americas.com
Czyz, J. A. (1994). Computation of Pipeline-Bending Strains Based on Geopig Measurements. Pigging and
Integrity Monitoring Conference. Houston, Texas.
Dotson, R. e. (2021). Beware the Shapeshifter: A Repeatability Study on Pipeline Movement and Bending Strain
Assessments. Pipeline Pigging and Integrity Management Conference. Houston, Texas.
Heutling, B., Reimche, W., Bach, F., Stock, M., Kroos, J., Stolzenberg, M., & Schulz, S. (2004). Online material
characterization of hot-dipped steel sheet with electromagnetic harmonic analysis. Montreal: WCNDT
2004.
ISO, 2. (2019). Petroleum and natural gas industry - Pipeline transportation systems – Geological hazard risk
management for onshore pipelines. Geneva: International Organisation for Standardisation.
McInnes, L., Healy, J., & Melville, J. (2020). UMAP: Uniform Manifold Approximation and Projection for Dimension
Reduction. https://doi.org/10.48550/arXiv.1802.03426.
Molenda, D., & Thale, W. (2014). A novel approach for the pipe grade determination through in-line inspection
(ILI). IPC2014-33441.
Nyman, D. L. (2008). Mitigating geohazards for international pipeline projects: challenges and lessons learned.
IPC2008-64405. Proceedings of 7th International Pipeline Conference, American Society of Mechanical
Engineers.
Slater, S., Dwyer, K., Ginten, M., & Eiken, T. (2018). An integrated approach for the verification of pipeline
material using the latest state-of-the-art technologies and engineering processes. IPC2018-78776.

13