The ONE GUYANA Floating Production Storage and Offloading (FPSO) unit has been designed to operate in deepwater conditions with a production capacity of 250,000 barrels of oil per day. The FPSO is designed for a 20-year operational life, handling complex processes including water and gas injection. Previous projects highlighted the operational and safety challenges posed by flow-induced vibrations in seawater overboard lines, prompting a comprehensive vibration review in the design phase to ensure system reliability and integrity.
This case study details the technical approach, findings, and targeted mitigations from a vibration analysis focusing on selected seawater overboard lines within the FPSO topsides modules. The emphasis was on lines downstream of major control valves, where vibration risks due to flow-induced turbulence (FIT) were suspected to be most severe.
Methodology
Modal Analysis
The initial phase involved a modal analysis of nine piping work packages, each associated with distinct overboard flow paths or critical control valves. The primary objective of this step was to identify natural frequencies and mode shapes, with particular attention to low-frequency modes (<7 Hz) that are susceptible to excitation by flow turbulence. The BOSfluids finite element modal solver, coupled with Ansys, was used for the analysis. Limitations inherent in the analysis included uncertainties in support structure stiffness, anisotropy in fiberglass sections, and the inability to model support gaps.
Each system’s susceptibility to vibration-induced fatigue was evaluated using the Energy Institute’s Guidelines for the Avoidance of Vibration-Induced Fatigue (2nd Edition, 2008), resulting in a Likelihood of Failure (LOF) score. The LOF is a relative indicator of risk; values above 0.3 prompt further investigation, and values above 0.5 require corrective action.
Computational Fluid Dynamics (CFD)
For piping sections exhibiting high LOF scores or geometrically driven concerns, CFD simulations provided further insight into flow-induced turbulence and the associated unsteady forces on pipe walls. Two turbulence modeling strategies were employed:
- Unsteady Reynolds-Averaged Navier-Stokes (URANS): Used for broader risk screening and identification of high-turbulence zones by resolving mean flow and turbulent kinetic energy.
- Large Eddy Simulation (LES): Applied selectively to high-risk regions, this method provided detailed, time-resolved force data for dynamic stress analysis.
CFD model domains were constructed to capture the most critical geometric features, with mesh refinement dictated by turbulent energy gradients. LES results were used to generate force spectra, which were subsequently applied in structural dynamic stress models to assess fatigue risk directly.
Key Findings
Modal Analysis Results
Among the nine work packages, three (Work Packages 1, 2/3, and 8) emerged as critical due to their combination of low natural frequencies and elevated LOF scores.
- Work Package 1 (VE200-WS001): High LOF (0.818) and natural frequencies near 7.32 Hz were found near the main sea water lift pump overboard valves (122-FCV-1113 A–D). Downstream bends had natural frequencies as low as 5.66 Hz, with long spans between supports exacerbating risk.
- Work Package 2/3 (TS091-WI004/TS091-WI006): The lowest frequency mode was 6.12 Hz, but the corresponding LOF (0.447) was below critical, indicating manageable risk provided small-bore connections are inspected.
- Work Package 8 (TS001-WS002): Downstream of the produced water cooler overboard valve (124-TCV-1502), a bend exhibited a natural frequency of 4.60 Hz and an LOF of 0.501, both indicative of a moderate risk zone.
Other work packages exhibited either higher natural frequencies or lower LOF scores and were not flagged for immediate mitigation, although a screening for small-bore connections is recommended where LOF > 0.3.
CFD and Dynamic Stress Analysis
- Work Package 1, Location 1 (Downstream of Control Valve): URANS simulations revealed strong shear layers and high turbulence arising from the modeled orifice (standing in for the globe valve). Turbulent kinetic energy was highest immediately downstream of the valve and at the subsequent T-section, confirming the high vibration risk observed in the modal analysis.
- Work Package 1, Location 2 (Downstream Bends): URANS indicated moderate turbulence at critical bends. LES provided time-resolved force data, which, when applied in a dynamic stress analysis, showed that maximum cyclic stresses at resonance were only 26% of the fatigue endurance limit—demonstrating an acceptable risk level even without additional supports.
- Work Package 8, Location 2 (Downstream Bend): CFD revealed moderate turbulence, with energy dissipating significantly by the second bend, suggesting a lower but non-negligible risk. The primary concern here was the low natural frequency rather than turbulence severity.
Mitigation Strategies
Support Arrangement Modifications
The central mitigation philosophy is to increase the natural frequencies of at-risk piping sections by locally stiffening support structures, thus reducing susceptibility to excitation by turbulence. All proposed changes were checked to ensure compliance with static stress limits.
- Work Package 1: Additional guide supports are to be installed just upstream of the critical control valves (nodes 1480, 2480, 3480, 4480). This modification increases the first natural frequency from approximately 7.3 Hz to 14 Hz, well above the threshold for FIT excitation.
- Work Package 8: Because direct stiffening at the critical bend (node 820) was impractical, horizontal movement of the trunnion at node 2090 will be restrained, raising the first natural frequency from 4.6 Hz to 6.7 Hz. Additionally, a guide support immediately downstream of the control valve (node 720) will eliminate problematic modes near 8.8 Hz.
Targeted Inspections
For systems where LOF exceeds 0.3 but no immediate design changes are planned, a detailed inspection of small-bore connections is mandated, as these are especially vulnerable to vibration-induced fatigue.
Conclusions and Recommendations
This study confirms that flow-induced turbulence, especially downstream of control valves, is the principal driver of vibration risk in the Yellowtail FPSO seawater overboard systems. Low-frequency modes at critical geometric locations—particularly at bends and near valves—are susceptible to excitation, but risk can be effectively mitigated through targeted support modifications.
Recommendations for future projects and design reviews:
- Incorporate guide supports near major control valves to preemptively address high turbulence zones.
- Implement the specific support modifications detailed herein for the current Yellowtail FPSO design (Sections 5.1 and 5.2 of the full report).
- Screen all piping systems with LOF > 0.3 for small-bore connection vulnerabilities and reinforce as necessary.
- Continue to combine modal and CFD analysis in the design phase for high-risk lines to enable evidence-based, cost-effective mitigation strategies.
The combined use of modal analysis and CFD, culminating in dynamic stress assessment, provides a robust methodology for identifying and mitigating vibration risks in critical FPSO piping systems. This approach ensures both operational reliability and long-term structural integrity in demanding offshore environments.