Faster simulation time continues to be a major industry priority for real-time application of integrated optimization asset modeling. Simultaneously, model sizes are rapidly becoming computationally more expensive. A new parallel simulation solution has been developed to quickly solve the largest and most complex modeling studies. This paper shows the benefits of the new parallel technique applied to a real case optimization. The described production optimization tool has the ability to integrate in a unique environment a whole production system, from wellbore to export and storage. A powerful evolutionary algorithm searches for the optimum field configuration that maximizes production. The tool computes the "fitness" of each solution, combines their properties in order to obtain new candidate solutions, and then selects the best individuals to allow the evolution of the population and detect the optimum. A master machine creates the individuals to be tested and multi-thread computation on a large number of simulation nodes allows to drastically reduce run times. The parallelism has been developed and applied to a complex integrated production optimization case study in order to prove the benefits of the novel architecture in terms of computational efforts. The tool was applied to the largest onshore oil field in Europe, with a very complex gathering network, and a large process plant composed of five treatment trains comprising acid gas removal and sulphur recovery unit. Comparison between the traditional and the parallel simulation architecture applied to the integrated production model of this asset shows a significant reduction in run times, decreasing from days to hours. These drastically reduced computational efforts have allowed the new simulator to include heavy computational methods to build more accurate models. Furthermore, the integrated production optimization tool has been developed to become an enterprise solution with an end-user friendly graphic interface linked to a common database for data handling and storage. This paper describes an innovative parallel simulation approach for an integrated production optimization tool. Reduced computational efforts and the possibility to build more accurate models are the main advantages of this tool, together with a newly developed friendly user interface. The benefits introduced by the novel parallel computational method developed for the tool guarantee real-time support to field production optimization even for the largest and most complex models.

This content is only available via PDF.
You can access this article if you purchase or spend a download.