NEXFLOW - Next generation oil-water flow models in production technologies

In the oil and gas production process so called multiphase flow pipes are used to transport produced oil, gas and water simultaneously in a single pipeline to the processing facility. Future multiphase production systems will face longer transport distances, high water production rates, complex flow effects caused by fluid mixing and introduction of production chemicals, and need to find energy efficient solutions. Within this context, handling and prediction of oil-water dispersions in the production lines is a main challenge that must be overcome in order to design and operate future production systems in an energy- and cost-efficient way. The primary objective of NEXFLOW was to improve the understanding of oil-water dispersions in the oil transport process. The effect of petrophysical fluid properties was studied systematically in pipe flow experiments. Advanced instrumentation was used to characterize dispersion flow on both the macro and micro scale. New knowledge was generated, and novel correlations developed for predicting dispersion behaviour. In parallel, simplified dispersion characterization methods were developed delivering fluid specific input parameters for the new correlations. New correlations are crucial for future improvements of multiphase pipe flow simulators. These simulators are powerful tools used during planning, design and operation. Improvements will help to realize longer multiphase transport lines and transport strategies for a wider range of production conditions as result of extended field lifetime and production from so called small satellite fields. The project was a collaboration between the partners SINTEF, NTNU and TotalEnergies EP Norge, focusing on the flow problematic, and was collaborating with a Brazilian partner project involving ISDB Flowtech, PUC Rio and ESSS, focusing on the generation and stability of dispersions in the production system. Regular status meetings were held throughout the project period with eager involvement of all partners and fruitful discussions and exchange of knowledge and results. Annual visits to our Brazilian partners were made and presentations at the annual November Conference given, as organized by The Research Council of Norway. During the project two experimental flow loop campaigns were conducted in a unique multiphase flow loop with extensive instrumentation and a test section length of more than 200m. Measurements comprised amongst others pressure drop, cross-sectional phase distributions, flow visualization, droplet sizes and wall wetting properties. The flow loop was equipped with an inlet choke valve which allowed creation and control of dispersions. Due to the length of the test section transient flow effects could be monitored which provided unique data for development and verification of new models. The first campaign focused on viscosity effects by testing oils with three different viscosities. The second campaign focused on the introduction of surfactant and by this stabilization of dispersion droplets. Spiking with a heavy crude oil was used to introduce natural surfactants and to mimic the complex chemistry of real production systems but without changing other physical properties. Higher crude concentration resulted in a clearly higher pressure drop (reduced drag reduction effect). Further, the in-flow separation was delayed for experiments with inlet choke mixing. A clear difference was found when compared with comparable experiments with the synthetic surfactant Span 83. While the crude oil surfactants provided a gradual in-flow separation, experiments with Span 83 resulted in a more sudden separation of the fluids. In parallel, different dispersion characterization methods were developed and tested. A custom-made bench scale rheology rig (pipe viscometer) was used for investigating the effect of different parameters on the effective viscosity and drag reduction phenomenon in dispersions, two mechanisms that are not well understood and typically missing in dispersion models. These experiments could confirm the behaviour observed in the flow loop. In addition, a stir tank was used to prepare dispersions and monitor dynamic droplet growth and separation when the stirring intensity was changed. Measured characteristics were comparable with the behaviour in the flow loop experiments. A steady-state gravity/turbulent diffusion model was developed in the project. The model considers in-situ droplet sizes as input and predicts concentration profiles within the pipe. Predictions were in very good agreement with the experiments. Through this work we found suitable closure laws for two of the main constituents for modelling entrainment of droplets/bubbles/particles, namely turbulent diffusion and gravitational drift. The results of this project were published in open access journal publications and presented at several international conferences.

The project has generated new knowledge about, new experimental data, new experimental test methods and new models for oil-water dispersion flow. The project results have clearly improved our knowledge about flow behaviour in the presence of oil-water dispersions. In particular understanding of the stability of dispersions in production pipelines and the influence of surface stabilizing chemicals (surfactants) on the dispersion stability and resulting flow behaviour has been extensively improved.This is important for evaluating the impact when introducing new production chemicals or producing fluids from new discoveries. The experimental test methods and setups applied in this project may be further developed and used in other projects or applications. In particular the developed bench scale characterisation methods for oil-water dispersions may be offered as a service for testing of production fluids and for deriving fluid specific input parameters for multiphase flow simulators. This will help to better tune models to field specific conditions and finally reduce uncertainty in predictions. The experimental data created in the project is unique with respect to the degree of detail and reliability. This is a result of the advanced instrumentation and size of the flow loop used. This data will serve as background for further research and development and finally improvement and verification of new multiphase flow models. New model correlations created and published during the project period are available for implementation in commercial and open source flow simulators. In fact, data and model approaches created in NEXFLOW were already successfully used in other projects aiming on improving and implementing new models in commercial Multiphase Flow simulators, such as LedaFlow. Further, the developed model concepts and frameworks have successfully been modified towards prediction of gas bubble flow and may be used for other particle flow. Main impact of the project will be a contribution to better planning and development of future oil fields, and in particular developments requiring long transport distances. It is expected that over 80% of future field developments on the NCS will be tie-back solutions with long multiphase transport distances (e.g. satellite fields and marginal discoveries far off existing infrastructure) (OG21 strategy document). Reduced uncertainty when predicting the expected pressure drop and flowability in such developments will improve the design and operational strategy considerably. This will result in improved energy-efficiency, reduced chemical consumption and more accurate cost estimates.

This proposal is submitted for the joint call between FINEP-RCN together with ISdB Flowtech as the Brazilian partner. Oil-water dispersions play an important role in the oil and gas production system as they have a direct effect on the pressure drop in transport lines. Reliable pressure drop predictions and multiphase flow model capabilities in general will lead to higher energy efficiency and cost reductions, and facilitate the development of longer transport lines and tiebacks. This will enable extended lifetime of fields, which during later years of production are exposed to low recovery rates and high water production. Risks of flexible field development concepts involving tie-in production from satellite fields will be reduced by reducing uncertainty of transport models and understanding of well-stream compatibility effects on flow characteristics. State-of-the-art models and commercial simulators are not fully predictive as effects of complex fluid chemistry and its interaction with the hydrodynamics of the flow are not represented. Currently, no physical mechanisms or model input parameters exist to account for the effects of surface chemistry, stability of interfaces and evolution of droplet sizes. Based on novel experimental methods closures for dispersion behavior will be developed. The main project result will be improved understanding of oil-water flow in transport pipelines. Unique pipe flow experiments with advanced measurement techniques will be conducted, to close the gaps in the current understanding. These data, together with the derived dispersion characteristics, can be used for dispersion flow model development and verification. The research topic is part of SINTEF's strategic institute commitment and builds on work from a strategic institute project at SINTEF. One PhD will be educated in the project. The PhD student will be encouraged to co-supervise at least 1 master student per year (both specialization project and master thesis).