(Editor’s note: Part 1 of this article was published in the June issue of P&GJ)
As cracking
threats gained prominence in integrity planning, ILI and assessment methods
were initiated and evolved. Attempts to apply non-destructive techniques
started in the 1970s, with commercial systems pioneered and introduced by Baker
Hughes for USCD (ultrasonic crack detection) for liquid pipelines in the 1990s
and EMAT (Electromagnetic Acoustic Transduction) for gas pipelines in the 2000s.
Integrity programs since then have evolved to reliably use high-performance
crack ILI systems [Ref. 14,18,19,20].
Reported
features from ILI related to cracking include “crack-like,” “crack field” and
“weld anomalies” as the primary feature types of interest as axially oriented
features (where a rupture threat from hoop stress is the expected failure
mode). EMAT may also include indicators of localized coating damage.
EMAT ILI systems for crack detection and characterization are available today across most diameters of long-distance pipelines (8-42 inches) (Figure 5).
An ironic factor
of early EMAT systems was the inability to discriminate material flaws and
discontinuities from cracking [Ref 20,21]. EMAT systems today address, and
report discriminated features distinctly. And for purposes of early detection
of stress concentrators as preferred in the blended hydrogen scenario, the
identification of inclusions, laminations and manufacturing flaws becomes
preferential in collected data. This ability is an example of potential
evolution of capabilities.
It is envisaged
that crack ILI will continue to evolve with Hydrogen pipelines as well as form
part of a robust integrity management program.
A core principle
of Engineering Critical Assessment (ECA) of cracks (including SCC, seam weld
cracks, girth weld cracking, laminations, and weld anomalies) is based in
fracture mechanics and fatigue crack growth methodologies [Ref 10] as well as an
understanding of reported ILI features and interpretations such as crack
profiling and crack field statistics. [Ref 14].
Updates to ILI
crack profiling was recently evaluated and validated [Ref 18] for achieving
reduced conservatism in fitness-for-service assessments (Figure 6). Such
approaches would apply to validation of cracking assessment in hydrogen
pipelines.
Hard Spots
Hard spots are
distinguished as localized areas of material in pipe that is typically of
higher “hardness” than the line pipe specification. As material hardness is
identified as a potential threat susceptibility condition for blended H2
operations, the identification of hard spot areas becomes a factor to address
higher risks of failure from defects due to a localized reduction in material
fracture toughness.
ILI technologies
specifically for localized material difference detection have existed since
the1990s, with even early ILI technologies being sensitive to localized
material differences since the 1970s. The most common measurement approach is
from remnant magnetic field detection and interpretation.
An example of an
ILI signal response for a confirmed hard spot area (Figure 7). Note that
signals are distinctive and repeatable. In some cases, other material
impurities may also be detectable.
Detected and reported areas of interest may also then be aligned with ILI crack, corrosion, deformation and strain data for a more comprehensive evaluation and assessment.
Corrosion, Gouging
Monitoring for
corrosion and external damage (mechanical damage) remains a prominent activity
for pipeline integrity as it predicates early detection of conditions,
primarily as corrosion and volumetric wall anomaly characterization but also
for external cracking (SCC) susceptibility.
As with cracking
threats, the threat of corrosion or gouging may be presumed to be an indicator
to cracking initiation and accelerated embrittlement crack failure, even from
internally permeated hydrogen. For gas pipelines, high-performance MFL (magnetic
flux leakage) (Figure 8), is recognized as the most reliable means of
ILI inspection for corrosion and gouging/damage threats.
With presumed
more conservative acceptability criteria for the case of flaws in hydrogen
pipelines, the presence of any such flaws through detection by ILI, is presumed
to trigger response and remediation activities. Defect specific sizing
tolerances per reported defect has been recently introduced.
As a per-defect
tolerance, it provides less conservatism and enables a better understanding of
risk and prioritization of severity. Such methods will also certainly likely be
advantageous to the critical assessment of metal loss features in the presence
of hydrogen.
Advanced MFL systems have the
capability for girth weld anomaly assessment, where circumferential crack-like
features may be detected and reported.
Pipeline Strain
There is a
premise and assumption that pipeline force cycling may not solely be from
pressure but potentially in conjunction with external forces/geotechnical
forces that may also cause pipeline damage, even with low cycling frequencies.
In today’s MFL
ILI platforms, they are typically run with associated high-performance “IMUs”
(Inertial Measurement Units) to collect motion measurements of the tool in the
pipeline which was pioneered by Baker Hughes in the late 1980s [Ref 22]. This
inertial measurement data can be processed to provide continuous GPS location
data of the center line of the pipeline, provide independent insight for
characterization of dents and other deformation features that may be present,
and be interpreted for representative bending strain (curvature) of the
pipeline at all points.
Abnormal or
unexpected levels of bending strain are indicators of external
force/geotechnical forces on the pipeline which may pose an imminent threat.
A primary use
for this method includes geotechnically active areas (prone to landslides),
areas near active geological faults or areas prone to large pipeline movements
due to seasonal changes such as muskeg/swamp conditions, offshore oceanic
forces, or ground frost-heave (such as in northern Canada and Alaska) [Ref 24].
And when such strain events are
coincident with time dependent flaws (like cracking and corrosion) such as
reported by MFL or EMAT systems, a more detailed engineering assessment is to
be considered, as assessments of time-dependent flaws in isolation are not
applicable [Ref 25].
Longitudinal Strain
New ILI
technology has been introduced that provides independent reporting of
longitudinal strain distinctly from conventional ILI IMU methods [Ref 26]. It
was initially motivated from a need to monitor for geohazard conditions beyond
bending strain itself, such as for potential initiation of buckling and
wrinkling due to compressive forces.
After H2 introduction, the strain
capacity of both base material and girth welds may be affected by H2
embrittlement, further increasing the risk of failure.
Girth Weld Assessment
Girth weld
anomaly analysis involves MFL, IMU and strain measurement data to provide
prioritized assessment for potential girth weld cracking. Abilities for the
identification of girth weld and spiral weld cracking based on advanced MFL ILI
technologies was previously presented [Ref 27].
The adoption of these methods has grown from recent history in North
America due to concerns of weld undermatching where a high rate of girth weld
failures since 2019 has occurred in USA both in active pipelines and in newly
constructed pipelines.
Causes are due to use of higher strength pipe than stated SMYS, and
inconsistent infield welding procedures leading to HIC and HAZ softening, which
significantly lowered the strain capacity for external forces and pipeline
movement [Ref 28]. As a hydrogen enabled issue, this should be presumed to be
more prominent and distinctive for hydrogen blended pipelines’ integrity.
Predictive Methods
All ILI
data-based integrity assessment methods will have practices for monitoring,
growth prediction and remaining life estimation. Criteria for feature response
and action to identified change (growth) are then according to operator and
industry practices.
Such methods are
envisaged to be adoptable, and necessary, for blended hydrogen pipeline
scenarios. Fundamentally for ILI based integrity programs, change detection is
the basis for growth rate estimation.
Within the evolution of pipeline
integrity practices, flaw growth methodologies started with corrosion. Baker
Hughes were principal authors in the generation of the primary industry
guidelines for corrosion growth including for deterministic and probabilistic
treatments [Ref 29].
The key aspects
of this method were to provide practical, consistent, and systematic means to
establish rates of growth. It also identified the opportunities for managing
localized and varied growth rates within a defect population, the direct use of
ILI-based signal change detection for more accurate growth rate estimation over
ILI reported box methods, and guidance to address and validate “very high” or
“improbable” rates of predicted growth in practical terms [Ref 30].
The concepts and
methods from corrosion (Figure 9) and have since been adapted and
applied for other threat types.
An increasing
direction of interest takes the form of more holistic assessment, where
coincident features and conditions to overall threats take form in integrity
programs, which address a high volume and variety of scenarios for potential
loss of containment through any means (Figure 10) [Ref 25, 30].
For hydrogen
pipeline integrity, such an approach may be necessary over conventional natural
gas methods, as the sensitivities of threats and their combinations have a
different interaction and set of outcomes over hydrocarbon scenarios.
Most significantly is the use and application of Machine Learning both to ILI signal interpretation but also to predictive analytics for significance of reported features and remaining defect populations. The availability of high count and variety of field validated samples lead to the opportunity for “Big Data” machine learning approaches [Ref 32,33].
Such techniques
and methods are naturally structured to manage large volumes of disparate data.
They may also involve engineering calculations, practices, and techniques as
guiding principles as well as independent explicit checks as boundary
conditions of predicted results and historic results.
Tool Readiness
In the context
of ILI systems history, the readiness of ILI for blended or pure Hydrogen
pipelines comes with the experience and perspective of readiness for other
pipeline products. Over time this has come to include methane gases, including
H2S, refined liquid products, CO2 & nonhydrocarbon gases, aromatics
(ethylenes), ammonia, etc. [Ref 34].
For the blended
hydrogen case, the first consideration is safety. As with hydrocarbon
pipelines, areas of highest risk are typically the ILI/pigging launch and
receive systems of pipelines, where conditions of explosive gas, air and fuel
are present.
The application
of ATEX practices for pipeline ILI/pigging activities has been used for several
decades, where typically risk of a potential explosive environment is mitigated
(removed) by purging of the facilities and access points, of the explosive gas
by a non-flammable gas. Of note is decompression behaviors of hydrogen
depending on pressure and state as would be found in ILI/pigging receiving
traps.
In more extreme
cases by analogy, there are similar and parallel procedures for ILI/pigging in
sour-service (H2S) and hazardous product lines that have also been used for
several decades.
For ILI tool
system compatibility and run endurance within a blended Hydrogen environment,
the assessment is like that for elements of the pipeline itself with initial
assessments dating back to the EU Natural project of 2006-2009 [Ref 3,34]. A
compatibility assessment includes the expected forms of Hydrogen as gaseous,
aqueous/ionic, and scenarios of the presence of water, etc. It reviews effects
on materials as the metallics, elastomers (seals), as notably mimics topics
noted in the conversion of service of a pipeline.
Monitoring of
ILI vehicles and components for damage is already built into active maintenance
practices. Materials and technologies for sealing such as in valves and
compression equipment are a source that ILI can also draw from, as such sealing
materials become qualified for hydrogen service. There is strong confidence
that compatibility can be addressed in ILI operations in hydrogen lines scaled
to match hydrocarbon and/or other hazardous product pipelines.
Operationally,
there is a significant change in ILI tool flow dynamics expected depending on
the percent of hydrogen in the product mix, particularly as it will affect gas
compressibility, drag, and bypass. For those ILI systems equipped with active
variable bypass “speed control” systems, operations at high flow rates will
need to be investigated and validated.
However, at one
time, that was also an unknown and concern of active ILI speed control systems
for hydrocarbon gas pipelines. Today, such systems are used several times every
day in pipelines around the world.
Summary
The progression
of current hydrocarbon integrity and inspection practices has occurred over the
last 50-plus years. From that experienced basis, their applicability to blended
hydrogen pipelines is expected to evolve to address the need of future energy
transition pipelines.
Further
stringent criteria and conservatism are envisaged for hydrogen pipeline
integrity at least until experiences with hydrogen pipeline operations as core
energy infrastructures become more prevalent and common.
Technologies
that have been developed and advanced for monitoring of conventional pipelines
are the basis for pipelines of the future energy transition. The application of
current crack ILI inspection technologies is mature and will have application
for hydrogen pipelines.
It is
anticipated that the multitude of technologies needed to manage the threats are
available and as a better understanding of critical flaw sizes and specific
threats are better established by the pipeline industry, technologies will
evolve to meet these potentially higher expectations. P&GJ
REFERENCES