| dc.description.abstract | The multiphase flow simulators OLGA 5 and LEDA version 1.05 are tested against experimental data and field data. Experimental data on gas-condensate flow in a high pressure flow loop (inclined pipe project), and on air-water flow in an S-shaped riser are used. The field data are from the Pecorade onshore field in the south west of France. The object of the study is to simulate the cases with realistic geometries and input parameters, to evaluate the computed results and compare them with existing data. The LEDA code is currently under development, and comparisons with existing flow simulators, like OLGA, is hence of great importance. The report focuses on the variables flow regime, pressure and liquid holdup, because these are important for achieving a realistic prediction of the flow. Both simulators achieve pressure gradient -and liquid holdup trends matching the inclined pipe experiments, and the major part of the values obtained are fairly accurate. Based on these cases, there is no obvious conclusion on when one simulator is better than the other. For some conditions the two codes predict different flow regimes; LEDA gives the correct one in most of these cases. However, it is found that a correct flow regime not automatically corresponds to a better prediction of pressure gradient and liquid holdup. With small pressure fluctuations and short slug periods, LEDA does not reproduce the terrain slugging observed in Pecorade field tests. A better rendition to the field data is achieved with OLGA by using the slugtracking module. With the S-riser configuration, LEDA s ability to simulate terrain induced slugging is tested in a simpler geometry. The resulting pressure- and holdup graphs imply annular flow instead of terrain slugging, which is observed experimentally. As an overall conclusion, LEDA appears as a promising development of a new multiphase flow simulator, but improvements are needed in simulation accuracy and speed. The simulation time in LEDA is considerably longer than in OLGA, and the results provided are for the present not good enough compared to real data. Both multiphase flow simulators need to improve the treatment of complex geometries, and the ability to predict terrain slugging. | nb_NO |
