Axial walking is the gradual movement of the flowline over time, driven by thermal cycles. Factors such as seabed bathymetry, flowline length, and presence of lateral buckles influence the magnitude and direction of the axial walking. If not addressed, excessive flowline walking can displace end structures and overstress connected components, potentially causing failure.

Digital twins, when applied strategically alongside risk-based inspection, are a powerful tool for managing these challenges. By offering continuous monitoring and data-driven insights, digital twins enable proactive risk mitigation, early detection of anomalies, and improved readiness for life extension. This approach helps prevent unplanned failures, while enhancing overall operational reliability.

Traditional Flowline Integrity Management

A traditional robust flowline integrity management programme to address the risks of flowline buckling and flowline walking typically includes an inspection and monitoring campaign.

Inspection can include the following:

• Remotely-operated vehicle (ROV) general visual inspection (GVI) – ROV GVI can detect flowline buckling and berms, and large movements of end structures based on soil markings.

• Positional surveys – Positional surveys can be conducted using light detection and ranging (LiDAR). The results are used to quantify buckle shapes and sizes and end structure movement. These surveys are more expensive than GVI and flowline movement is unknown in the timeframes between inspections.

Monitoring approaches generally fall into two categories:

• In-situ instrumentation (i.e. strain and motion loggers) - In-situ instrumentation provides direct measurements and requires a thorough understanding of flowline behaviour to determine sensor placement. For HPHT flowlines, fatigue hotspots can differ from design predictions, complicating sensor placement. Additionally, the high capital and operational costs of installing and maintaining subsea sensors for up to 20 years make this approach less practical.

• Process-based monitoring (e.g., using pressure, temperature, and flow rate data) - Process data offers a more practical alternative but can lead to overly conservative fatigue estimates if used alone.

A Smarter Approach: Machine Learning and Digital Twins

A machine learning digital twin model, capable of near real-time monitoring of flowline structural response, can also optimise flowline integrity management. When integrated with a risk-based inspection (RBI) strategy, the digital twin provides operators with enhanced visibility into HPHT flowline behaviour, enabling them to identify and respond to anomalous trends quickly, optimise inspection intervals, and confidently improve long-term asset reliability.

The digital twin model is created using the following key data from design documentation:

• Flowline physical properties including insulation and route details

• Flowline stress and displacement response from FEA analysis

• Site specific ambient temperature along the flowline route

• Flow assurance profiles (steady state and transient) 

Raw measured data from subsea and topsides sensors is pre-processed and filtered to scrub spurious data.

Digital twin input data includes:

• Flow rates

• Pressure

• Temperature

• Manifold valve position in multi-well tiebacks

The digital twin estimates temperature distribution and structural response along the flowline using a segmented, Eulerian-based approach. Calibrated initially with downstream sensor data or flow assurance models, it is further refined over time using linear regression, improving accuracy as more operational data is processed. Initial model calibration uses downstream sensor data or flow assurance modeling where available. A linear regression algorithm continuously refines model accuracy as new data becomes available, improving prediction fidelity over time. It is important to re-calibrate the digital twin against measured LiDAR data over the course of the service life.

Digital twin outputs include:

• Temperature timeseries at locations of interest (virtual sensors)

• Thermal cycles histograms

• Stresses

• Accumulated fatigue damage

• Walking distance over time

• Buckle displacement over time

Case Study – Flowline Walking

2H has developed and implemented flowline digital twin models for a key client optimising their flowline integrity management program across multiple assets. These digital twins enabled them to plan inspections more efficiently as well as plan mitigation measures helping to prevent a jumper overstress scenario.

The case study focuses on a subsea field located in approximately 2,600ft water depth, which began production in 2015. The field consists of three wells, a central manifold, two PLETs, two rigid PLET jumpers connecting the manifold to the PLET, and two rigid flowlines.

During the design phase, thermo-mechanical finite element (FE) analysis identified axial walking as a potential concern for both flowlines. Although mitigation measures such as PLET anchoring were not considered necessary at the time, continuous monitoring was recommended for the life of the asset. To manage this risk, allowable walking distances were defined based on jumper stress limits, using both ASME B31.4 code requirements and the jumper’s material yield strength. These thresholds act as key performance indicators (KPIs) for jumper structural integrity, allowing the operator to compare actual walking, measured via surveys or predicted by the digital twin, against design limits.

Before the implementation of the digital twin, flowline walking was managed using a risk-based inspection (RBI) strategy and by monitoring thermal cycle trends. Inspections consisted of ROV GVI and positional surveys to measure jumper hub-to-hub movements and global axial movement. ROV GVI focused on detecting seabed changes around the PLET foundations, while positional surveys combined as-built acoustic data and subsequent LiDAR surveys to generate highly accurate 3D models.

In January 2022, a digital twin was introduced to enable predictive monitoring based on operational data. By the end of 2023, the model predicted 19.5 inches of cumulative walking for the South PLET, slightly exceeding the amber KPI limit of 19 inches for the rigid PLET jumper. The digital twin indicated an average walking rate higher than the original design estimate, driven primarily by an increased number of shutdown cycles in 2023. These findings prompted an anomaly-based LiDAR inspection in early 2024.

The 2024 LiDAR survey confirmed that measured movement closely matched with digital twin predictions. A fitness-for-service (FFS) assessment was then performed to evaluate jumper structural integrity and guide the decision on whether to mitigate the flowline walking or replace the jumper. The FFS concluded that the South PLET jumper can tolerate up to 42 inches of total displacement before reaching the 0.5% strain limit a threshold predicted to be reached in mid-2025 based on historical walking rates.

To prevent further walking, a mitigation plan was developed involveing the installation of double-stack concrete mattresses on the flowline at the PLET end. The combination of acceptable FFS results and continuous digital twin monitoring allowed the operator to make a fully data-driven decision to continue operating the field while executing the remediation plan.

Conclusions

Using a flowline digital twin to replicate the physical and thermal behaviour of subsea infrastructure allows operators to monitor critical assets in near-real-time. While it does not fully replace physical measurements from field positional surveys, a digital twin provides a robust framework for predicting in-service performance, enabling more targeted inspection, smarter remediation planning, and reduced unplanned downtime, all while ensuring structural integrity. This approach gives operators greater confidence in managing HPHT flowlines and offers a path to reduce engineering conservatism in future field developments.

For a deeper dive into the flowline digital twin and the case study discussed here, read OTC paper OTC-35694-MS “Using Flowline Digital Twin to Plan for Inspections and Remediation”, which was co-authored by the integrity management division of 2H (formerly Clarus Subsea Integrity).