The thermal management of a subsea hydrocarbon production system is a major task over the entire life cycle of the subsea production system (SPS) development. The thermal studies include heat transfer modeling and temperature-distribution calculations of the production fluid and hardware of the entire SPS stretched between the reservoir and the process facility. The thermal model involves the subsea environment as well as the SPS operating conditions (e.g. reservoirs, seawater, ambient and arrival temperatures and pressures). The major objective of the thermal management process is to aid the design of the SPS as well as to verify the production philosophy. The design includes component temperature qualifications and thermal insulation design, and the production philosophy includes maximum and minimum temperature during steady production and transient flow scenarios.
The thermal study applies analysis and simulation tools that are based on fundamental heat transfer theory. However, limitations of computation resources or project time lines lead to simplifications and assumptions in the thermal modeling approaches. One major assumption comprises the modification of the thermal conductivity of the modeled fluids to account for the transient free convection process in the subsea components (for example for the hydrocarbon fluid flow in the piping system or for the trapped fluids). This assumption reduces the calculation effort associated with the explicit modeling of the transient buoyancy and natural convection flow and heat transfer.
This paper focuses on the validation of this assumption. A combination of experimental and numerical results are presented and are utilized to recommend a robust thermal modelling procedure for subsea components that allows a favorable balance between conservatism and accuracy. The experimental data consists of both steady state and transient results from models of subsea components. Three-dimensional numerical conjugate Computational Fluid Dynamics/Heat Transfer (CFD/CHT) simulations were performed with assumed effective conductivity and with explicitly resolved free convection. Results are compared to the experimental data. The comparison between the experimental and model exit temperatures during steady state flow showed a good match with a 7% offset as a maximum value for the deviation between measured and calculated temperatures. In the transient results, the numerical model using an effective conductivity consistently gave a closer match than did the model explicitly modelling buoyancy. Given the closer match and the advantage of numerical efficiency, it is concluded that it is best to treat a production fluid as a solid with an effective conductivity when modelling transient events. Further test data are utilized to validate thermal modelling using the effective conductivity shows acceptable results accuracy of the cooldown behavior of subsea components. With increased confidence in the numerical model results, there is less of a need to apply over conservative boundary conditions. This in turn leads to a more robust product and decreased cost due to avoidance of over engineered designs.