ABSTRACT This study evaluates the performance of three industry‐standard simulation tools—Aspen HYSYS (V14), Multiflash (V7.5), and OLGA (V2024.1)—by comparing predicted hydrate dissociation temperatures with experimental data obtained from a PVT sapphire cell. The analysis focuses on systems representative of typical pipeline operating conditions (80–200 bar) containing water, monoethylene glycol (MEG), and lower concentrations of sodium chloride, which are commonly encountered inhibitors in natural gas transmission systems. Differences in thermodynamic models and equations of state were examined to assess their ability to capture the combined inhibitory effects of MEG and dissolved salts. The results demonstrate that all investigated software packages overpredict hydrate dissociation temperatures in the presence of salts, indicating that the full inhibitory effect of salts is not adequately represented. For salt‐free systems, Aspen HYSYS with the glycol fluid package provided the most reliable predictions, whereas Multiflash using the Peng–Robinson equation of state yielded the closest agreement with experimental data for salt‐containing mixtures. However, even this model underestimates salt inhibition. The low salt concentration range (1–3 wt%) was found to exhibit the largest model inaccuracies. An adjustment‐factor analysis suggested that the salt inhibition is underestimated by a factor of 3–4 in commonly used software models for very low (~1 wt%) concentrations. Overall, this study provides a systematic validation and quantification of model bias under representative pipeline conditions, highlighting limitations in current thermodynamic formulations and identifying software configurations that offer improved hydrate prediction accuracy for inhibited systems.
Bloomfield et al. (Thu,) studied this question.