Secure models for model-based control and optimization
In certain embodiments, a control/optimization system includes an instantiated model object stored in memory on a model server. The model object includes a model of a plant or process being controlled. The model object comprises an interface that precludes the transmission of proprietary information via the interface. The control/optimization system also includes a decision engine software module stored in memory on a decision support server. The decision engine software module is configured to request information from the model object through a communication network via a communication protocol that precludes the transmission of proprietary information, and to receive the requested information from the model object through the communication network via the communication protocol.
1. A computer-implemented method, comprising:
instantiating a plurality of model objects on a plurality of model servers, wherein each of the plurality of model objects comprises a model configured to model operation of a plant or process being controlled, and wherein at least one of the models contains protected information;
requesting information from the plurality of model objects via a communication network;
receiving the information from the plurality of model objects via the communication network;
generating an approximation of at least one of the plurality of the models of the plant or process being controlled, wherein the approximation is relevant to generating control commands;
generating control commands based at least in part on the approximation and the information received from the plurality of model objects without accessing the protected information;
transmitting the control commands to at least one of the plurality of model objects via the communication network; and
controlling operation of an industrial automation component in the plant or process based on the control commands.
2. The computer-implemented method of claim 1 , wherein each of the plurality of model objects comprises a parametric hybrid model.
3. The computer-implemented method of claim 2 , wherein each parametric hybrid model comprises an empirical model, a parameter model, and a parametric first-principles model.
4. The computer-implemented method of claim 2 , wherein the approximation comprises an approximation of the parametric hybrid model over a future horizon of interest.
5. The computer-implemented method of claim 1 , wherein the information received from the plurality of model objects does not include information tagged as protected information.
6. The computer-implemented method of claim 5 , wherein at least one of the model objects comprises a parametric hybrid model that utilizes the protected information to determine the approximation.
7. The computer-implemented method of claim 1 , wherein each of the model objects comprises an interface that precludes the transmission of protected information via the interface.
8. The computer-implemented method of claim 1 , wherein the information and the control commands are transmitted through the communication network via a communication protocol that precludes the transmission of protected information.
9. The computer-implemented method of claim 1 , wherein the information and the control commands are transmitted via a cloud.
10. A control/optimization system, comprising:
a plurality of model servers configured to store a plurality of instantiated model objects, wherein each of a plurality of model objects comprises a model configured to model operation of a plant or process being controlled;
a plurality of decision support servers configured to store a plurality of decision engine modules, wherein each of the decision engine modules comprises instructions and each of the plurality of decision support servers is configured to execute the instructions to:
request information from the plurality of model objects via a communication network;
receive the information from the plurality of model objects via the communication network, wherein the information comprises an approximation of each of the models;
generate control commands based at least in part on the information received from the plurality of model objects without accessing proprietary information utilized in the plurality of models objects, wherein the control commands are generated collectively by the plurality of decision engine modules; and
transmit the control commands to at least one of the plurality of model objects via the communication network to control operation of an industrial automation component in the plant or process.
11. The control/optimization system of claim 10 , wherein each of the plurality of model objects comprises a parametric hybrid model.
12. The control/optimization system of claim 11 , wherein the approximation comprises an approximation of each parametric hybrid model over a future horizon of interest.
13. The control/optimization system of claim 11 , wherein each parametric hybrid model comprises an empirical model, a parameter model, and a parametric first-principles model.
14. The control/optimization system of claim 10 , wherein the information received from the plurality of model objects does not include information tagged as proprietary information.
15. The control/optimization system of claim 10 , wherein each of the model objects comprises an interface configured to preclude the transmission of proprietary information via the interface.
16. The control/optimization system of claim 10 , wherein:
the plurality of model servers is configured to use the plurality of model objects to transmit the information through the communication network via a communication protocol that precludes the transmission of proprietary information: and
the plurality of decision support servers is configured to use the plurality of decision engine modules to transmit the control commands through the communication network via the communication protocol.
17. The control/optimization system of claim 10 , wherein:
the plurality of model objects each comprises a parametric hybrid model; and
the plurality of model servers is configured to use the parametric hybrid model of each of the plurality of model objects to determine the approximation based on the proprietary information.
18. A control/optimization system, comprising:
a model server comprising an instantiated model object, wherein the model object comprises a model configured to model operation of a plant or process being controlled, and wherein the model object utilizes proprietary information and comprises an interface configured to preclude transmission of the proprietary information via the interface; and
a decision support server comprising a decision engine module, wherein the decision support server is configured to use the decision engine module to:
request information from the model object through a communication network via a communication protocol that precludes the transmission of proprietary information;
receive the requested information from the model object through the communication network via the communication protocol, wherein the requested information comprises an approximation of the model;
generate control commands to control operation of an industrial automation component in the plant or process without accessing the proprietary information; and
transmit the control commands to the model object via the communication network to control operation of the industrial automation component in the plant or process.
19. The control/optimization system of claim 18 , wherein the model object comprises a parametric hybrid model.
20. The control/optimization system of claim 19 , wherein the approximation comprises an approximation of the parametric hybrid model over a future horizon of interest.
21. The control/optimization system of claim 19 , wherein the parametric hybrid model comprises an empirical model, a parameter model, and a parametric first-principles model.
22. The control/optimization system of claim 19 , wherein model server is configured to use the parametric hybrid model to determine the approximation based on the proprietary information.