Process optimization by grouping mixed integer nonlinear programming constraints

1. A system, comprising:

one or more computers comprising one or more processors and one or more non-transitory computer readable media, the one or more non-transitory computer readable media comprising instructions stored thereon that when executed cause the one or more computers to:

generate, by the one or more processors, a plurality of model components implementing at least one first-principle equation, wherein the at least one first-principle equation represents at least one process unit, and wherein two or more of the plurality of model components comprise a group;

generate, by the one or more processors, a plurality of switch components each comprising a corresponding one of the plurality of model components and each implementing a mixed integer nonlinear programming (MINLP) behavior of the corresponding one of the plurality of model components; wherein each switch component includes a group identifier parameter of the group and a group complement parameter encoding an operational constraint; and

generate, by the one or more processors, an MINLP solver component configured to switch each of the plurality of model components of the group between an active state and an inactive state based at least in part on the group identifier parameter and the group complement parameter during a simulation.

2. The system of claim 1 ,

wherein the group identifier parameter comprises a common value among each of the two or more of the plurality of model components comprising the group.

3. The system of claim 1 ,

wherein the plurality of model components each exhibit a complementary switching behavior during the simulation when the group complement parameter is in a marked state.

4. The system of claim 1 ,

wherein each of the plurality of switch components are made aware of a group leader unit model component of the plurality of model components.

5. The system of claim 1 ,

wherein the group comprises a first set of model components each having an unmarked group complement parameter and a second set of model components each having a marked group complement parameter.

6. The system of claim 1 ,

wherein the group comprises at least one model component not comprising one of the plurality of switch components.

7. The system of claim 1 ,

wherein each of the plurality of switch components further includes one or more logical constraints.

8. A method for generating an optimal operating state of a process comprising:

providing one or more computers comprising one or more processors and one or more non-transitory computer readable media, the one or more non-transitory computer readable media including instructions stored thereon that when executed implement steps of:

simulating, by the one or more processors, an operation of a plurality of unit models in conjunction with a mixed integer nonlinear programming (MINLP) solver, the plurality of unit models each representing one of a plurality of process units via at least one first-principle equation;

grouping, by the one or more processors, two or more of the plurality of unit models into a group based on at least one operating constraint, each of the plurality of unit models of the group having an MINLP switch implementing MINLP behavior, and each MINLP switch encoding the at least one operating constraint;

switching, by the one or more processors, the MINLP switch of each of the plurality of unit models of the group between an active state and an inactive state during said simulating based at least in part on the at least one operating constraint; and

generating, by the one or more processors, an operating state of a process based on said simulating, said grouping, and said switching, wherein the operating state satisfies the at least one operating constraint.

9. The method of claim 8 ,

wherein each MINLP switch encodes the at least one operating constraint as a group identifier parameter; and

wherein the instructions further comprise a step of assigning a common value to the group identifier parameter of each unit model of the group.

10. The method of claim 9 ,

wherein each MINLP switch further encodes the at least one operating constraint as a group complement parameter, and wherein the instructions further comprise a step of assigning at least one of a marked state and an unmarked state to each group complement parameter, wherein a unit model exhibits complementary switching behavior with respect to other unit models of the group when the group complement parameter of the unit model is in a marked state, and wherein a unit model exhibits similar switching behavior with respect to other unit models of the group when the group complement parameter of the unit model is in an unmarked state.

11. The method of claim 8 ,

wherein said grouping comprises making each of the plurality of unit models of the group implicitly aware of a group leader unit model.

12. The method of claim 8 , wherein said grouping comprises partitioning each of the plurality of unit models of the group into a first set of non-complementary unit models and a second set of complementary unit models.

13. The method of claim 8 ,

wherein the instructions further comprise a step of associating the plurality of unit models of the group with at least one unit model not having the MINLP switch.

14. The method of claim 8 ,

wherein each MINLP switch encodes the at least one operating constraint as a logical constraint.

15. A system comprising:

one or more non-transitory computer readable media, the one or more non-transitory computer readable media including instructions stored thereon that when executed are configured to cause one or more computers to:

generate, by one or more processors, a plurality of model modules each defining a plurality of unit models, each unit model representing a process unit within a process via at least one first-principle equation during an execution of the plurality of model modules by the one or more processors, wherein two or more of the plurality of model modules comprise a group;

generate, by one or more processors, a plurality of mixed integer nonlinear programming (MINLP) switch modules each transitioning a corresponding model module of the group between an active state and an inactive state based at least in part on a parameter encoding at least one operating constraint, the plurality of MINLP switch modules each implementing an MINLP behavior of the corresponding model module; and

generate, by one or more processors, a solver module simulating an operation of each of the plurality of model modules and each of the plurality of MINLP switch modules during an execution of the solver module to generate an operating state of a process, wherein the operating state satisfies the parameter encoding the at least one operating constraint.

16. The system of claim 15 ,

wherein the parameter comprises a group identifier parameter; and

wherein a common value is assigned to the group identifier parameter of the plurality of model modules of the group.

17. The system of claim 16 ,

wherein the parameter further comprises a group complement parameter; and

wherein a model module exhibits complementary transitioning behavior with respect to other model modules of the group during the execution when the group complement parameter of the model module is in a marked state.

18. The system of claim 15 ,

wherein each of the plurality of model modules of the group are made aware of a group leader model module.

19. The system of claim 15 ,

wherein the group comprises a first set of non-complementary model modules and a second set of complementary model modules.

20. The system of claim 15 ,

wherein the parameter comprises one or more logical constraints encoding the at least one operating constraint.