Skouras E.,Statoil |
Solbraa E.,Statoil |
Lokken T.V.,Statoil |
Aaserud C.,Gassco AS
International Gas Research Conference Proceedings | Year: 2014
The PhaseOpt technology is a tool for online monitoring of hydrocarbon dew point in rich gas pipelines. Accurate prediction of hydrocarbon dew point temperature and pressure is significantly important to obtain effective utilization of the natural gas pipelines and process plants without hydrocarbon liquid drop-out. A discussion covers a description of PhaseOpt technology; main technology elements in PhaseOpt; Cricondenbar specification in rich gas pipelines and process plants; business drivers for PhaseOpt technology; and results of the field tests carried out to evaluate and qualify the PhaseOpt technology. This is an abstract of a paper presented at the International Gas Union Research Conference (Copenhagen, Denmark 9/17-19/2014).
Munkejord S.T.,Sintef |
Bernstone C.,Vattenfall |
Clausen S.,Gassco AS |
De Koeijer G.,Statoil |
Energy Procedia | Year: 2013
The present paper concerns the importance of a combined modelling and experimental effort to develop physicsbased combined thermodynamic and transient flow models for CO2-transport pipelines. Such models need to handle both multiple components and two-phase flow, which can occur both during normal operation and transient situations, such as first fill and depressurization. Moreover, these models can provide useful input to risk analyses and design of mitigation actions of undesirable incidents like pipeline rupture and well blow-outs. This paper discusses various physical phenomena, design issues and solutions by using the experience from actual cases encountered by Vattenfall, Gassco and Statoil. Among the key issues are the minimum temperature in the pipe wall during depressurization and the magnitude of pressure oscillations during transient operation.
Dugstad A.,Institute for Energy Technology of Norway |
Halseid M.,Institute for Energy Technology of Norway |
Morland B.,Institute for Energy Technology of Norway |
Sivertsen A.O.,Gassco AS
Energy Procedia | Year: 2013
When dense phase CO2 is depressurised and forms a two phase gas/liquid system, impurities will partition between the two phases and go preferentially to the phase where their solubility is highest. Partitioning and depressurisation experiments run at 4 and 25 °C showed that water, H2S and SO2 accumulated while O2 was depleted in the remaining liquid CO2 phase when the system was depressurised via the gas phase. When the water solubility is exceeded, a third aqueous phase can form. The accumulation of impurities increased the corrosivity of the remaining liquid phase and carbon steel specimens exposed in the autoclaves were corroded after 3 days exposure. Lowest corrosion rates (< 0.1 mm/year) were measured in a system with CO 2 and water (489 and 1222 ppmv) only. The corrosion rate is reasonably low because the water phase quickly becomes saturated with corrosion products that reduce the corrosivity. The situation was different when the CO2 contained SO2 (138 ppmv) and NO2 (191ppmv) as these gases accumulate in the water phase, reduce the pH and increase the reactivity of the aqueous phase. The specimens in the SO2 experiments got covered with a black film and the corrosion rates were about 0.1 mm/year while the specimens in the NO2 experiment developed a brown rusty layer of corrosion products and the corrosion rate was about 1 mm/year. These high corrosion rates only last until the impurities are consumed, the replenishment and availability of impurities becomes therefore an important issue. The accumulation of impurities (including water) was maximum 5- 10 times the original concentration in the experiments where the liquid phase was reduced to 10-20 % of the original volume. The ratio of the remaining to the original liquid phase volume can be much lower in a long pipeline and a much larger accumulation can be foreseen. If the corrosion rate consumes most of the available impurities in the pipeline, the corrosion rate can be much higher than the corrosion rate measured in the present experiments.
Abrahamsen E.B.,University of Stavanger |
Aven T.,University of Stavanger |
Iversen R.S.,Gassco AS
Proceedings of the Institution of Mechanical Engineers, Part O: Journal of Risk and Reliability | Year: 2010
Safety management and uncertainty management have traditionally been completely separate functions. The two disciplines are, to a large extent, based on different scientific pillars. The safety discipline typically produces frequency estimates of specific hazardous events, whereas the uncertainty discipline produces prediction intervals based on probability distribution quantiles, in addition to mean values. Furthermore, the safety discipline has a focus on risk acceptance criteria, whereas the uncertainty discipline makes top-ten and similar lists to rank the most critical uncertainty aspects. These differences make it difficult to obtain an integrated framework for safety management and uncertainty management. However, the recent introduction of risk perspectives highlighting the uncertainty component of risk has provided an improved basis for development of such an approach. By seeing risk as a two-dimensional concept covering events and consequences on the one side and uncertainties on the other, the content and boundaries of risk assessments are changed. The gap between the two disciplines can, to a large extent, be bridged. The purpose of the current paper is to present and discuss an integrated framework for these disciplines and traditions, based on this risk perspective. The starting point is petroleum operations, but the discussion is, to a large extent, general.
Clausen S.,Gassco AS |
Energy Procedia | Year: 2012
A numerical benchmark study has been performed, comparing results from the OLGA pipeline simulator to data obtained using the multi-stage (MUSTA) centred scheme. The case considered is a depressurization of a pipeline. Care was taken to compare the same flow model and thermodynamic equation of state. The two methods appear to converge on fine grids, but on coarse grids, the MUSTA method is more accurate. © 2012 The Authors. Published by Elsevier Ltd.