Heuristic method of automated and learning control, and building automation systems thereof
Apparatuses, systems, and methods of physical-model based building automation using in-situ regression to optimize control systems are presented. A simulation engine is configured to simulate a behavior or a controlled system using a physical model for the controlled system. A data stream comprises data from a controlled system. A training loop is configured to compare an output of a simulation engine to a data stream using a heuristic so that a physical model is regressed in a manner that the output of the simulation engine approaches the data stream.
1. A controller for controlling a controlled system, the controller comprising:
a memory; and
a processor in communication with the memory and configured to:
identify a plurality of candidate control schemes for controlling the controlled system, wherein each of the plurality of candidate control schemes comprises at least one control action;
simulate a behavior of the controlled system using a physical model for the controlled system to predict an outcome of a control action of at least one of the plurality of candidate control schemes;
evaluate the plurality of candidate control schemes based on the simulation of the behavior of the control system;
select a selected control scheme of the plurality of the candidate control schemes based on the outcome of evaluating the plurality of candidate control schemes; and
apply the selected control regime to the controlled system.
2. The controller of claim 1 , wherein, in evaluating the plurality of candidate control schemes based on the simulation, the processor is configured to assess performance of the control action performance according to a cost function.
3. The controller of claim 2 , wherein the cost function is time variant.
4. The controller of claim 1 , wherein at least one of the plurality of candidate control schemes comprises a valuation of a respective control action.
5. The controller of claim 1 wherein simulating the behavior using the physical model comprises simulating the effects of at least one external factor on the controlled system.
6. The controller of claim 5 , wherein the at least one external factor comprises current weather conditions.
7. The controller of claim 5 , wherein the at least one external factor comprises a current occupancy.
8. The controller of claim 5 , wherein the at least one external factor comprises input from a user.
9. The controller of claim 5 , wherein the at least one external factor comprises a predicted external factor.
10. The controller of claim 9 , wherein the predicted external factor includes a future weather prediction.
11. The controller of claim 9 , wherein the predicted external factor includes an occupancy schedule.
12. A method executed by at least one processor for controlling a controlled system, the method comprising:
identifying, by the at least one processor, a plurality of candidate control schemes for controlling the controlled system, wherein each of the plurality of candidate control schemes comprises at least one control action;
simulating a behavior of the controlled system using a physical model for the controlled system to predict an outcome of a control action of at least one of the plurality of candidate control schemes;
evaluating the plurality of candidate control schemes based on the simulation of the behavior of the control system; and
selecting a selected control scheme of the plurality of the candidate control schemes based on the outcome of evaluating the plurality of candidate control schemes.
13. The method of claim 12 , wherein the step of evaluating the plurality of candidate control schemes based on the simulation comprises assessing performance of the control action performance according to a cost function.
14. The method of claim 13 , wherein the cost function is time variant.
15. The method of claim 12 , wherein the step of simulating the behavior using the physical model comprises simulating the effects of at least one external factor on the controlled system.
16. The method of claim 15 , wherein the at least one external factor comprises current weather conditions.
17. The method of claim 15 , wherein the at least one external factor comprises a current occupancy.
18. The method of claim 15 , wherein the at least one external factor comprises input from a user.
19. The method of claim 15 , wherein the at least one external factor comprises a predicted external factor.
20. The method of claim 19 , wherein the predicted external factor includes a future weather prediction.
21. The method of claim 19 , wherein the predicted external factor includes an occupancy schedule.
22. The method of claim 12 , further comprising applying the selected control regime to the controlled system.
23. A non-transitory machine-readable storage medium encoded with instructions for execution by at least one processor for controlling a controlled system, the non-transitory machine-readable storage medium comprising:
instructions for identifying, by the at least one processor, a plurality of candidate control schemes for controlling the controlled system, wherein each of the plurality of candidate control schemes comprises at least one control action;
instructions for simulating a behavior of the controlled system using a physical model for the controlled system to predict an outcome of a control action of at least one of the plurality of candidate control schemes;
instructions for evaluating the plurality of candidate control schemes based on the simulation of the behavior of the control system;
instructions for selecting a selected control scheme of the plurality of the candidate control schemes based on the outcome of evaluating the plurality of candidate control schemes; and
instructions for applying the selected control regime to the controlled system.
24. The non-transitory machine-readable storage medium of claim 23 , wherein the instructions for evaluating the plurality of candidate control schemes based on the simulation comprise instructions for assessing performance of the control action performance according to a time variant cost function.
25. The non-transitory machine-readable storage medium of claim 23 , wherein the instructions for simulating the behavior using the physical model comprise instructions for simulating the effects of at least one external factor on the controlled system.