Apply PC-SAFT Model in Industrial Scale Dynamic Simulation

Industrial scale applications usually involve much more components compare to systems used in academic research. A typical phase equilibrium calculation within industrial application can easily contain more than twenty components. When thermodynamic models such as Perturbed Chain Statistical Associate Fluid Theory (PC-SAFT) is applied in such simulation, computational cost becomes the major concern. Such concern can determine successful or failure of dynamic simulation when thousands of such phase equilibrium calculations must be done within transient time interval. In many cases, such transient time interval is a quarter second.

In order to apply advanced thermodynamic models such as PC-SAFT for industrial scale dynamic simulation, we initiated a bounding algorithm based on a pair of linear forms, which we named as “super-linear approach”. Such linear forms were created based on inverse function theorem. The pair of linear forms were then applied to bound the value of thermodynamic properties in extended operational range. With such linear forms, the property calculation becomes much more efficient compare to the effort of rigorously calculating it in the full operation range. The simulation accuracy was still guaranteed by an automatic rigorous update scheme, where the results of the pair of linear forms started to deviate.

The super-linear approach has been successfully applied to PC-SAFT in AVEVA dynamic simulation suite in commercial applications of huge-scale dynamic simulations of real processes including polymer production, which runs with a practical speed keeping the benefits of PC-SAFT. In general, it can be applied to any thermodynamic model and any type of phase equilibrium calculation.